Front cover

Introducing IBM FileNet

Business Process Manager

Business process design

Integration with external

systems and services

Step-by-step
instructions

Wei-Dong Zhu
Eric Adel
William Benjamin
Imtiaz A Khan
Mike Marin
Mark Yingling

ibm.com/redbooks
International Technical Support Organization

Introducing IBM FileNet Business Process Manager

August 2008

SG24-7509-00
 Note: Before using this information and the product it supports, read the information in
“Notices” on page ix.

First Edition (August 2008)

This edition applies to Version 4, Release 0 of IBM FileNet Business Process Manager (product
number 5724-R76).

© Copyright International Business Machines Corporation 2008. All rights reserved.

Note to U.S. Government Users Restricted Rights -- Use, duplication or disclosure restricted by GSA ADP
Schedule Contract with IBM Corp.
Contents

Notices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Trademarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
The team that wrote this book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Become a published author . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii
Comments welcome. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiv

Chapter 1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Business process management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.1 Transactional processes and content centric processes . . . . . . . . . . 3
1.2 IBM FileNet Business Process Manager overview . . . . . . . . . . . . . . . . . . . 4
1.3 Process design with BPM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.3.1 Process definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.3.2 Steps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.3.3 Routes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.3.4 Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.3.5 Deadlines and timers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.3.6 Content events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.4 BPM application development options . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.4.1 IBM FileNet Workplace (or Workplace XT) . . . . . . . . . . . . . . . . . . . . 10
1.4.2 Business Process Framework (BPF) . . . . . . . . . . . . . . . . . . . . . . . . 11
1.4.3 Electronic forms (eForms or Lotus Forms) . . . . . . . . . . . . . . . . . . . . 11
1.4.4 Portal user interface through IBM FileNet P8 portlet. . . . . . . . . . . . . 12
1.4.5 Web Application Toolkit (WAT) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.4.6 Process Engine API . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.5 BPM integration to external systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.5.1 Component Integrator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.5.2 Web services. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.5.3 Business rules engines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.6 BPM process monitoring and analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.6.1 Process Simulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.6.2 Process Analyzer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.6.3 IBM FileNet Business Activity Monitor . . . . . . . . . . . . . . . . . . . . . . . 16
1.7 BPM applications and tools summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
1.8 The IBM FileNet P8 family of products . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
1.8.1 IBM FileNet content products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
1.8.2 IBM FileNet process products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

© Copyright IBM Corp. 2008. All rights reserved. iii

 1.8.3 IBM FileNet compliance products . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
1.8.4 IBM FileNet system management tools . . . . . . . . . . . . . . . . . . . . . . 21

Chapter 2. Solution example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

2.1 Scenario background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.2 Claim approval process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.2.1 Initiating a claim (steps 1 and 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.2.2 Back office mailroom clerk task (step 3) . . . . . . . . . . . . . . . . . . . . . . 28
2.2.3 Claim processing (steps 4 through 6) . . . . . . . . . . . . . . . . . . . . . . . . 29
2.2.4 Financial controls (part of steps 5 through 6 and 7) . . . . . . . . . . . . . 30
2.2.5 Workflow process maps for the claim approval process . . . . . . . . . . 32
2.3 Summary of benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

Chapter 3. System architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

3.1 IBM FileNet P8 core engines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.2 Layered architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.2.1 Presentation tier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.2.2 Middleware tier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.2.3 Data tier. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.3 Process Engine (PE) applications and other supporting applications . . . . 43
3.3.1 Process Simulator system architecture . . . . . . . . . . . . . . . . . . . . . . . 44
3.3.2 Process Analyzer system architecture . . . . . . . . . . . . . . . . . . . . . . . 44
3.3.3 Business Activity Monitor system architecture . . . . . . . . . . . . . . . . . 45
3.3.4 Business Process Framework system architecture . . . . . . . . . . . . . 46
3.4 Service Oriented Architecture (SOA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.5 Process Engine logical view . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

Chapter 4. Process design with BPM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

4.1 Basic concepts in process design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4.1.1 Tasks, participants, and roles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4.1.2 Workflow properties (workflow groups and attachments) . . . . . . . . . 57
4.1.3 Steps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
4.1.4 Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
4.1.5 Milestones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
4.1.6 Deadlines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
4.1.7 Task reassignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
4.1.8 Process voting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
4.2 BPM supported data types and system functions . . . . . . . . . . . . . . . . . . . 62
4.2.1 Supported data types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4.2.2 System functions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

iv Introducing IBM FileNet Business Process Manager

4.3 Inter-related processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
4.4 Interaction with external systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

Chapter 5. Using Process Designer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

5.1 Process Designer user interface overview . . . . . . . . . . . . . . . . . . . . . . . . 68
5.1.1 Map pane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
5.1.2 Properties pane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
5.1.3 Step Palettes pane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
5.1.4 Toolbar icons. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
5.2 Creating workflow definition overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
5.3 Setting up workflow properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.4 Creating steps in a workflow map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
5.4.1 General step . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
5.4.2 Submap step . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
5.4.3 System step . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
5.4.4 Component step . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
5.4.5 Launch step. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
5.4.6 Start step . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
5.5 Creating routes with Process Designer . . . . . . . . . . . . . . . . . . . . . . . . . . 111
5.5.1 Setting conditions for routes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
5.6 Advanced topics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
5.6.1 Workflow inheritance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
5.6.2 Default system submaps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
5.6.3 Expression Builder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122

Chapter 6. Implementing business processes: Case study . . . . . . . . . . 127

6.1 Building business processes overview . . . . . . . . . . . . . . . . . . . . . . . . . . 128
6.2 Defining the basic process object model. . . . . . . . . . . . . . . . . . . . . . . . . 129
6.3 Starting a new workflow process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
6.3.1 Defining workflow properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
6.3.2 Saving the workflow process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
6.4 Building the process workflow map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
6.4.1 Adding steps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
6.4.2 Adding routes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
6.4.3 Setting up conditional routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
6.4.4 Validating the workflow process map . . . . . . . . . . . . . . . . . . . . . . . 157
6.5 Finishing the step properties definition . . . . . . . . . . . . . . . . . . . . . . . . . . 159
6.5.1 Setting up a component step . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
6.6 Building a submap. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
6.6.1 Creating the Claim Setup submap . . . . . . . . . . . . . . . . . . . . . . . . . 172
6.6.2 Setting up system steps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
6.6.3 Linking the submap to the main map . . . . . . . . . . . . . . . . . . . . . . . 184
6.6.4 Checking in and transferring your workflow process map. . . . . . . . 186

Contents v
 6.7 Creating a workflow subscription. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
6.7.1 Exposing a system property and mapping it in a workflow . . . . . . . 199
6.8 Testing the basic process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
6.8.1 Monitoring workflow processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
6.8.2 Testing the workflow process from beginning to end . . . . . . . . . . . 212
6.9 Creating and calling a Web service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
6.9.1 Creating a workflow process as a Web service . . . . . . . . . . . . . . . 231
6.9.2 Calling the Web service in a workflow process . . . . . . . . . . . . . . . . 244
6.9.3 Create a subscription to launch your process . . . . . . . . . . . . . . . . . 256
6.9.4 Testing the process with Web services . . . . . . . . . . . . . . . . . . . . . . 256

Chapter 7. Integration with external systems and services with BPM . . 257
7.1 Component Integrator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258
7.1.1 Content Engine operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
7.2 Web services. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
7.2.1 Process orchestration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
7.3 Rules connectivity framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267

Chapter 8. Implementing Component Integrator and Web services . . . 271

8.1 Understanding Component Integrator . . . . . . . . . . . . . . . . . . . . . . . . . . . 272
8.1.1 Adapter and security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
8.2 Implementing Component Integrator. . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
8.2.1 Creating and configuring a component queue . . . . . . . . . . . . . . . . 274
8.2.2 Importing component queue operations . . . . . . . . . . . . . . . . . . . . . 279
8.2.3 Configuring Component Manager . . . . . . . . . . . . . . . . . . . . . . . . . . 283
8.2.4 Using components in a workflow. . . . . . . . . . . . . . . . . . . . . . . . . . . 285
8.3 Understanding Web services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
8.3.1 Web services adaptor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
8.4 Implementing a process to invoke a Web service . . . . . . . . . . . . . . . . . . 288
8.4.1 Configuring the UDDI registry list . . . . . . . . . . . . . . . . . . . . . . . . . . 289
8.4.2 Defining Web services information in workflow properties . . . . . . . 291
8.4.3 Defining an Invoke step in a workflow . . . . . . . . . . . . . . . . . . . . . . . 293
8.5 Implementing a process as a Web service . . . . . . . . . . . . . . . . . . . . . . . 296
8.5.1 Defining Web services information in workflow properties . . . . . . . 296
8.5.2 Using the Receive/Reply steps in a workflow . . . . . . . . . . . . . . . . . 297
8.5.3 Finalizing Web services operations and transferring the workflow
definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300
8.5.4 Publishing the workflow process as a Web service . . . . . . . . . . . . 301

Chapter 9. Planning and designing a BPM solution . . . . . . . . . . . . . . . . 303

9.1 Planning a BPM solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
9.1.1 Discovering and understanding the process . . . . . . . . . . . . . . . . . . 306
9.1.2 Preparing for a walkthrough . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
9.1.3 Requirements gathering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

vi Introducing IBM FileNet Business Process Manager

 9.1.4 Interviews and questionnaires. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309
9.1.5 XYZ Corporation as-is auto claim approval process . . . . . . . . . . . . 310
9.2 Functional design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312
9.2.1 Transforming requirements into a solution . . . . . . . . . . . . . . . . . . . 312
9.2.2 XYZ Corporation future (to-be) auto claim approval process . . . . . 313
9.2.3 Locking down the requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . 317
9.3 Detailed design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317
9.3.1 Solution components. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318
9.3.2 Modeling the process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320
9.3.3 Technical architecture overview . . . . . . . . . . . . . . . . . . . . . . . . . . . 321
9.3.4 Cost analysis metrics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323
9.3.5 Final walkthrough . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324

Appendix A. Additional material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325

Locating the Web material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
Using the Web material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326
System requirements for downloading the Web material . . . . . . . . . . . . . 326
How to use the Web material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326

Related publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327

IBM Redbooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
Online resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
How to get Redbooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
Help from IBM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 328

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329

Contents vii
viii Introducing IBM FileNet Business Process Manager
Notices

This information was developed for products and services offered in the U.S.A.

IBM may not offer the products, services, or features discussed in this document in other countries. Consult
your local IBM representative for information on the products and services currently available in your area.
Any reference to an IBM product, program, or service is not intended to state or imply that only that IBM
product, program, or service may be used. Any functionally equivalent product, program, or service that
does not infringe any IBM intellectual property right may be used instead. However, it is the user's
responsibility to evaluate and verify the operation of any non-IBM product, program, or service.

IBM may have patents or pending patent applications covering subject matter described in this document.
The furnishing of this document does not give you any license to these patents. You can send license
inquiries, in writing, to:
IBM Director of Licensing, IBM Corporation, North Castle Drive, Armonk, NY 10504-1785 U.S.A.

The following paragraph does not apply to the United Kingdom or any other country where such
provisions are inconsistent with local law: INTERNATIONAL BUSINESS MACHINES CORPORATION
PROVIDES THIS PUBLICATION "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESS OR
IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF NON-INFRINGEMENT,
MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Some states do not allow disclaimer
of express or implied warranties in certain transactions, therefore, this statement may not apply to you.

This information could include technical inaccuracies or typographical errors. Changes are periodically made
to the information herein; these changes will be incorporated in new editions of the publication. IBM may
make improvements and/or changes in the product(s) and/or the program(s) described in this publication at
any time without notice.

Any references in this information to non-IBM Web sites are provided for convenience only and do not in any
manner serve as an endorsement of those Web sites. The materials at those Web sites are not part of the
materials for this IBM product and use of those Web sites is at your own risk.

IBM may use or distribute any of the information you supply in any way it believes appropriate without
incurring any obligation to you.

Information concerning non-IBM products was obtained from the suppliers of those products, their published
announcements or other publicly available sources. IBM has not tested those products and cannot confirm
the accuracy of performance, compatibility or any other claims related to non-IBM products. Questions on
the capabilities of non-IBM products should be addressed to the suppliers of those products.

This information contains examples of data and reports used in daily business operations. To illustrate them
as completely as possible, the examples include the names of individuals, companies, brands, and products.
All of these names are fictitious and any similarity to the names and addresses used by an actual business
enterprise is entirely coincidental.

COPYRIGHT LICENSE:

This information contains sample application programs in source language, which illustrate programming
techniques on various operating platforms. You may copy, modify, and distribute these sample programs in
any form without payment to IBM, for the purposes of developing, using, marketing or distributing application
programs conforming to the application programming interface for the operating platform for which the
sample programs are written. These examples have not been thoroughly tested under all conditions. IBM,
therefore, cannot guarantee or imply reliability, serviceability, or function of these programs.

© Copyright IBM Corp. 2008. All rights reserved. ix

Trademarks
IBM, the IBM logo, and ibm.com are trademarks or registered trademarks of International Business
Machines Corporation in the United States, other countries, or both. These and other IBM trademarked
terms are marked on their first occurrence in this information with the appropriate symbol (® or ™),
indicating US registered or common law trademarks owned by IBM at the time this information was
published. Such trademarks may also be registered or common law trademarks in other countries. A current
list of IBM trademarks is available on the Web at http://www.ibm.com/legal/copytrade.shtml

The following terms are trademarks of the International Business Machines Corporation in the United States,
other countries, or both:

DB2® Lotus® WebSphere®

Domino® OmniFind™ Workplace™
FileNet® Redbooks®
IBM® Redbooks (logo) ®

The following terms are trademarks of other companies:

Cognos, and the Cognos logo are trademarks or registered trademarks of Cognos Incorporated, an IBM
Company, in the United States and/or other countries.

FileNet, and the FileNet logo are registered trademarks of FileNet Corporation in the United States, other
countries or both.

Network Appliance, SnapLock, and the NetApp logo are trademarks or registered trademarks of NetApp,
Inc. in the U.S. and other countries.

IBM FileNet, and the IBM FileNet logo are registered trademarks of IBM FileNet Corporation in the United
States, other countries or both.
J2EE, Java, JavaScript, JDBC, JSP, Streamline, and all Java-based trademarks are trademarks of Sun
Microsystems, Inc. in the United States, other countries, or both.

Excel, Expression, Internet Explorer, Microsoft, Visio, Windows, and the Windows logo are trademarks of
Microsoft Corporation in the United States, other countries, or both.

Other company, product, or service names may be trademarks or service marks of others.

x Introducing IBM FileNet Business Process Manager

Preface
This IBM® Redbooks® publication provides a basic introduction to IBM FileNet®
Business Process Manager (BPM) V4.0. BPM enables organizations to create,
modify, manage, simulate, and analyze content centric business processes. One
key advantage of BPM is its ability to work with active content, which refers to the
ability of content to trigger or affect business processes.

In this book, we cover the key elements that make up a business process,
including tasks, participants, roles, steps, routing, and deadlines. We describe
how to use Process Designer (a BPM application) to design your business
processes. In addition, we provide step-by-step instructions on how to implement
a use case business process scenario (an auto claim approval process).

BPM supports integration with external systems and services through

Component Integrator, Web services, and Rules Connectivity Framework. In this
book, we discuss these integration options and provide instructions on how to
implement the use case by demonstrating some of the options.

At the end of the book, we discuss, from a high level, the planning and designing
of content centric BPM applications.

This book is useful for system architects, process analysts, and designers who
require an understanding of IBM FileNet Business Process Manager. It also
serves as a practical guide for those who want detailed instructions in order to
implement a BPM system.

The team that wrote this book

This book was produced by a team of specialists from around the world working
at the International Technical Support Organization, San Jose Center.

Wei-Dong Zhu (Jackie) is an Enterprise Content Management (ECM) Project

Leader with the ITSO in San Jose, California. She has more than 10 years of
software development and management experience in accounting, image
workflow processing, and digital media distribution. Jackie holds a Master of
Science degree in Computer Science from the University of Southern California.
She joined IBM in 1996 and has managed and led the production of numerous
Enterprise Content Management Redbooks in the area of IBM Content Manager,
OnDemand, CommonStore, Document Manager, Records Manager,
WebSphere® Commerce Module, IBM Content Analyzer (previously known as
OmniFind™ Analytics Edition), IBM FileNet Content Manager, and more.

© Copyright IBM Corp. 2008. All rights reserved. xi

 Eric Adel is a Senior Technical Instructor with the IBM Enterprise Content
Management (ECM) Education group. He has 15 years of software development
experience involving artificial intelligence, data mining, content management,
and business process management. In the ECM Education group at IBM, he is
the current subject matter expert for IBM FileNet P8 Content Manager API
Programming and IBM FileNet Business Process Manager API Programming.
Eric holds a Master of Science degree in Electrical Engineering and has finished
all course work for his Ph.D. in Object Recognition and Artificial Intelligence.
His areas of expertise include IBM FileNet P8 content management, business
process management, artificial intelligence, and data mining, as well as EAI,
Web services, SOA, and XML. He has designed projects for both the J2EE™
and .NET platforms.

William Benjamin is a Solutions Architect and Information Technology

Specialist for IBM's Enterprise Content Management software segment. He has
over 10 years of experience working in content management ranging from
application development to systems architecture, software engineering, quality
assurance, customer education, and technical consulting. William is a subject
matter expert in the area of Content Integration and has participated or led in the
full spectrum of the software development life cycle from idea inception to post
implementation support.

Imtiaz A Khan is a Solution Architect and a Certified Consulting Information

Technology Specialist for the IBM Enterprise Content Management software
segment. He has 26 years of working experience in the imaging and information
management industry. He holds a Bachelor of Engineering and Associate degree
in Information Technology. Imtiaz joined IBM in 2000. He holds various industry
certifications such as Records Management Practitioner from the Association of
Information Technology AIIM, as well as MIT accreditation from AIIM and CDIA+
certification from CompTIA.

Mike Marin is a Distinguished Engineer at IBM Software Group and the Product
Architect for IBM FileNet Business Process Manager. He holds a Master of
Science degree in Computer Science, specializing in Artificial Intelligence and
has more than twenty years of experience designing and developing system
software. The last eleven years, he has been developing business process
management and workflow products, and participating in standard organizations
including OMG, OASIS, and WfMC, working on business process management
and workflow standards. He has edited and contributed to the definition of
several of these standards. Mike is a Fellow of the WfMC and has received the
WfMC Excellence Award for his technical contributions to the WfMC
standardization efforts.

xii Introducing IBM FileNet Business Process Manager

 Mark Yingling is a Certified Information Technology Specialist for IBM's
Enterprise Content Management software segment. He has 20 years of
information management and software development experience. He holds a
Bachelor of Science degree in Computer Science from Colorado State
University. Mark joined IBM in 2001. He is certified by the Association for
Information and Imaging Management as an Enterprise Content Management
Practitioner.

We would like to extend special thanks to the following people who have made
this project possible:

Richard Palfini
Steve Mason
Manoj Puthenveeti
Darik Siegfried
IBM Software Group, Costa Mesa, California

We also thank the following people for their contributions to this project:

Charles Burnett
Jingli Chen
Khoi Dang
Quynh Dang
Victor Falco
Kevin Fally
Evangeline Fink
Steven Kelly
Patrice Lewis
Jane Luo
Mark Muller
Ken Rong
Michael Samiley
Bernadine Sisneros
Chuck Snow
Christopher Timbreza
IBM Software Group, Costa Mesa and Sacramento, California

Become a published author

Join us for a two- to six-week residency program! Help write a book dealing with
specific products or solutions, while getting hands-on experience with
leading-edge technologies. You will have the opportunity to team with IBM
technical professionals, Business Partners, and Clients.

Preface xiii
 Your efforts will help increase product acceptance and customer satisfaction. As
a bonus, you will develop a network of contacts in IBM development labs, and
increase your productivity and marketability.

Find out more about the residency program, browse the residency index, and
apply online at:
ibm.com/redbooks/residencies.html

Comments welcome
Your comments are important to us!

We want our books to be as helpful as possible. Send us your comments about

this book or other IBM Redbooks® in one of the following ways:
򐂰 Use the online Contact us review Redbooks form found at:
ibm.com/redbooks
򐂰 Send your comments in an e-mail to:
[email protected]
򐂰 Mail your comments to:
IBM Corporation, International Technical Support Organization
Dept. HYTD Mail Station P099
2455 South Road
Poughkeepsie, NY 12601-5400

xiv Introducing IBM FileNet Business Process Manager

 1

Chapter 1. Overview
In this chapter we provide a high level overview of the IBM FileNet Business
Process Manager (BPM) and an introduction to the IBM FileNet P8 family of
products.

We cover the following topics:

򐂰 Business process management
򐂰 IBM FileNet Business Process Manager overview
򐂰 Process design with BPM
򐂰 BPM application development options
򐂰 BPM integration to external systems
򐂰 BPM process monitoring and analysis
򐂰 BPM applications and tools summary
򐂰 The IBM FileNet P8 family of products

© Copyright IBM Corp. 2008. All rights reserved. 1

1.1 Business process management
A business process is a set of activities performed by people and machines to
achieve a business goal. Business processes have been used as long as
businesses have been in existence because every aspect of a business involves
processes. Organizations achieve their objectives by implementing and
controlling a diverse set of business processes. Most of these processes are not
automated, or are only partially automated.

Business process management involves techniques and tools to manage and

automate business processes that are performed by people, applications, and
external sources. Automating and streamlining business processes helps
organizations to get the work done effectively and efficiently. Managing,
transforming, and optimizing business processes are the primary focus of
business process management.

An ideal business process management solution should provide the following

basic functions:
򐂰 Process automation
򐂰 Workflow
򐂰 Integration with external systems and applications
򐂰 Process optimization

Process automation
Processes are performed by people and applications. Automated processes
save time when compared with manual processes. The goal of process
automation is to provide technology to coordinate how people and applications
work together to perform processes. The technology should provide enough
functionality and be robust enough to be trusted with business critical processes.

Process automation can be achieved by automating a set of activities in a

process using custom applications or business process management products.

Workflow
Workflow is a structured way of performing tasks by an individual or a team.
Examples of workflow include auto insurance claims that require the efforts of
multiple people to process. The workflow process involves a policy holder filing a
claim, a field agent processing the claim, an adjustor adjusting the claim amount,
and finally the claimant either receiving the claim amount or the claim being
rejected.

2 Introducing IBM FileNet Business Process Manager

 A business process management solution should provide the workflow
capability. In addition, it must also provide the management and simulation of
processes, content integration, and the ability to interface with external systems.

Integration to external systems and applications

Integration to external systems and applications enables processes to interact
with automated applications and systems that are already in place. The
integration leverages the investment in existing applications and systems, and
enables quick and easy deployment of business process management
applications.

Process optimization
Process optimization refers to measuring, monitoring, simulating, and analyzing
business processes, and ultimately improving the processes. By providing the
continuous process optimization capability in a business process management
solution, organizations can increase their responsiveness, identify issues before
they become problems, and maximize the performance of their business
operations.

1.1.1 Transactional processes and content centric processes

Business process management applications generally fall under two categories:
򐂰 Transactional processes
򐂰 Content centric processes

Transactional processes heavily rely on application integration and

system-to-system processing. These are processes that interconnect with the
enterprise applications. The processes are short lived and might not require
human interaction.

Content centric processes are highly collaborative and require users or systems
to make decisions based on content during the processes.

Content is any unstructured information ranging from simple e-mails, pictures,

documents, and instant messages to complex forms, contracts, and reports.
Content can include any file format such as scanned images and electronic
documents, rich media, XML, and even business processes.

Content centric processes involve people and content and are usually longer
lived than transactional processes. Any addition or change to content, such as
new document creation, can trigger the associated process to act on the content,
thus the processes are considered content centric.

Chapter 1. Overview 3
 Most applications require both content and transactional processes to run
together to achieve a business goal. As organizations start automating
processes, they require both types of processes.

1.2 IBM FileNet Business Process Manager overview

IBM FileNet Business Process Manager (BPM) enables organizations to create,
modify, manage, simulate, and analyze business processes. One of the key
advantages of IBM FileNet Business Process Manager is its ability to work with
active content.

Active content refers to the ability of content to trigger or affect business

processes. For example, content such as a loan application or a claim
application, received via fax, e-mail, or Web, can trigger a process in IBM FileNet
Business Process Manager to process the loan or claim automatically. Active
content includes the delivery of information in a way that is more unified,
accurate, and contextual to the business processes. It enables a wide variety of
process management scenarios.

IBM FileNet Business Process Manager can be used in many areas, including
banking, insurance, manufacturing, health industry, and government. Examples
in the banking and insurance industries include applications for processing loans,
claims, credit card approvals and policy underwriting.

IBM FileNet Business Process Manager addresses the primary focus of business
process management (that is, managing, transforming, and optimizing business
processes). It provides the following functions:
򐂰 Process automation
򐂰 Process modeling and designing
򐂰 Process integration
򐂰 Process monitoring and optimization

Process automation
IBM FileNet Business Process Manager (BPM) provides a Process Engine and a
set of associated components and applications that enable easy implementation
of robust business process management solutions designed for use by business
users. BPM also provides a set of APIs to enable custom programming for
automating processes and a flexible environment for process improvements.

Process modeling and designing

IBM FileNet Business Process Manager makes a distinction between process
modeling and process designing. Process modeling is done for any business

4 Introducing IBM FileNet Business Process Manager

process in the organization, including manual processes that are not expected to
be automated. The goal of process modeling is to document and gain an
understanding of the process. Process designing, on the other hand, is used to
create executable processes.

Process models can be created by tools such as WebSphere Business Modeler,

or Microsoft® Visio®. WebSphere Business Modeler is a tool specifically
designed for process modeling and to simulate those processes. Microsoft Visio
is a drawing tool that can be used to document business processes. In either
case, the modeled processes can be used in IBM FileNet Business Process
Manager.

IBM FileNet Business Process Manager offers Process Designer, a tool that
creates executable processes. It can import Microsoft Visio diagrams and read
XML process definition language (XPDL) files that are generated by WebSphere
Business Modeler. In addition, Process Designer can read XPDL files generated
by other tools. Process Designer can use these models as a starting point to
create executable processes. IBM FileNet Business Process Manager also
offers Process Simulator, a tool that is used to simulate these processes to
identify bottlenecks and to play with “what if” scenarios.

We discuss process design in more detail later in this book.

Process integration
IBM FileNet Business Process Manager provides process integration capabilities
with external applications and systems through Component Integrator, rules
engine framework, and Web services. These integration options accelerate the
speed of business process management solution development and reduce the
overall costs associated with the development and maintenance of processes.

We discuss process integration in more detail later in this book.

Process monitoring and optimization

Business processes have to be managed and monitored so they deliver the
desired business goal. The monitoring of business processes can be done on
two levels; either at the individual process instance or across all processes. An
example of an individual process is the mortgage process for account 123.
Monitoring the instance for account 123 will show the current state of that
process instance.

Individual process instances can be monitored in several ways. IBM FileNet

Business Process Manager includes a tracker tool that displays the status of a
particular process in a graphical environment similar to the Process Designer.
There is also a milestone feature that can be used to define process milestones
during the process modeling. This can be used to monitor the status of the

Chapter 1. Overview 5
 process. In the mortgage case, some milestones might include application
received, in process, approved, and rejected. In this scenario, a business analyst
can be assigned as the tracker for a particular mortgage process and the
customer can be provided with access to the milestone view of the process.

For monitoring and analysis of all the processes, IBM FileNet Business Process
Manager offers Process Analyzer and Business Activity Monitor (BAM). Process
Analyzer enables users to evaluate process workload, discover processing
trends, and identify bottleneck areas. Business Activity Monitor lets managers
monitor various aspects of their business operations, set thresholds, and
automate actions to react to thresholds being exceeded.

Information collected by process monitoring can be used for optimization. For

newly designed processes, simulation is a technique that can be used to play
with “what if” scenarios and detect bottlenecks before placing the process into
production.

IBM FileNet Business Process Manager offers Process Simulator, which can be
used to analyze processes and evaluate how they behave in different scenarios.
Process Simulator can be populated with data from Process Analyzer to work
with processes that are already in production.

1.3 Process design with BPM

Process design starts by identifying the activities and the order in which those
activities must be performed to accomplish the business process. The outcome
of this is a graph that documents the business process and can be used to
automate it. IBM FileNet Business Process Manager offers the Process Designer
tool, which can be used to design processes and to import processes from
WebSphere Business Modeler and Microsoft Visio.

Process design and process improvement are iterative activities because

business processes must evolve and adapt to changing business requirements.
IBM FileNet Business Process Manager supports the iterative activities by
allowing for multiple versions of a process and providing easy management of
process evolution.

1.3.1 Process definition

For process models to become executable in the Process Engine, it is necessary
to add execution details. An executable process model is called a process
definition in BPM and it is stored in an XML Process Definition Language

6 Introducing IBM FileNet Business Process Manager

 (XPDL), or a PEP file. Process Designer can be used to generate fully
executable processes.

A process definition describes the activities and resources required to

accomplish a business process. It consists of a directed graph, with a series of
process activities or steps (nodes) connected together by a series of routes
(arcs), which define the sequence in which the steps are executed. Steps and
routes are organized into reusable maps. Maps, steps, and routes can be
annotated with comments. See Figure 1-1.

Map

Annotation
Step

IBM Software Group | Lotus softwar

Route

Figure 1-1 Process Designer showing a map with steps, routes and a step annotation

There can be multiple process definitions per business process to support

multiple versions. Process definitions can also inherit from other process
definitions. Inheritance allows organizations to specialize processes to create
consistency across a set of related processes. The process definitions are stored
in the IBM FileNet P8 Content Engine repository.

1.3.2 Steps
Steps represent specific business or system activities. Activities can be
performed by an individual user (the participant), by a group of users, or by an
automated application.

Chapter 1. Overview 7
 There are several types of steps:
򐂰 Launch step: The first step in a process. Every process has this step.
򐂰 General step: Represents a general activity to be processed by a participant
(or a group of users), or an automated process. It can be categorized as
follows:
– Participant step: A step that has an associated participant or a group of
users, all of whom must process the work item to complete the step. The
identity of these users can be defined at runtime through the use of
groups.
– Work queue step: A step that is assigned to a work queue (see definition
below) instead of a specific participant.
– Unassigned step: A step that has no effect in the process and can be used
for routing or documentation purposes.
򐂰 System step: Represents one or more functions to be performed by the
system. For example, a system step might include assigning data field
values, creating a new process instance, or suspending a process for a
specified period of time.
򐂰 Submap step: Calls another map in the current process definition.
򐂰 Component step: Performs operations in an external application or system.
It is accomplished by IBM FileNet Business Process Manager’s Component
Integrator.
򐂰 Web services step: Invokes or implements Web services. IBM FileNet
Business Process Manager uses Web services to integrate to external
applications and services.

Work queue
A work queue holds work items that can be completed by a user from some user
groups or by an automated process. Assigning work to a work queue rather than
to an individual participant provides flexibility as to who can perform the specific
activity at a step.

1.3.3 Routes
Routes define the order of execution for a series of steps based on specific rules
(using workflow data fields) and user responses (Approve and Deny as
examples). You can specify to always take a route or to only take the route when
a condition is met.

8 Introducing IBM FileNet Business Process Manager

 Routes can be used to define optional branches in a process map. For example,
in Figure 1-1 on page 7, the Evaluate step has two optional outgoing routes, one
labeled Approved and the other Reject. If a user approves the application, the
Approved route is taken. If the user rejects the application, then the Reject route
is taken.

Routes can also be used to define parallel branches where the steps are
executed in parallel.

1.3.4 Maps
A map represents the sequences of steps and routes required to complete a
process. You can define maps using Process Designer or import them from other
process design tools, such as Microsoft Visio.

A complex process can be broken down into simpler and reusable components
with each component being represented by a map. A process, therefore, can use
a collection of maps. The ability to break down processes into smaller
components makes complex processes easier to build and understand. It also
results in a significant reduction in the total cost of development. The reuse of the
maps ensures the consistency of processing and reduces time and costs
associated with the deployment of new processes.

A process definition that has been deployed can be executed by another

process. Processes that are used this way are called subprocesses.
Subprocesses are different than maps in that subprocesses are complete
processes, while maps are process segments.

1.3.5 Deadlines and timers

Users can create deadlines and timers to ensure that work is processed in a
timely manner. A deadline provides a time-based scheduling constraint, which
requires that a step or process be completed within a certain amount of time. The
deadline can be relative to the time that the step was routed to the participant, or
to the time that the process was launched.

A participant with a deadline can receive a reminder of the pending deadline

through e-mail. When the deadline is passed, a visual reminder is displayed in
the participant’s inbox, and an e-mail can be sent to a group of users.

A timer specifies a period of time before it is triggered. Multiple timers can be

created, terminated, and modified during a process. A timer specifies a sub-map
that is executed when the timer expires.

Chapter 1. Overview 9
1.3.6 Content events
Processes can be automatically launched when objects in a content repository
are created or changed. In general, any content event can be used to launch or
to trigger a process. For example, if a new customer (object) is created, a
process that manages new customers can be automatically launched.

1.4 BPM application development options

There are several ways in which P8 Web applications can be developed using
one or a combination of the following development options:
򐂰 IBM FileNet Workplace™ or Workplace XT1 are powerful J2EE applications
that can be customized to incorporate business logic.
򐂰 IBM FileNet Business Process Framework (BPF) allows customers to
develop and deploy quickly and efficiently a fully customized Web application
with interactive user interface and no additional coding.
򐂰 Electronic forms (eForms or Lotus® Forms) provide build-in business logic,
integration with back-end systems with data validation, and an easily
customizable user interface that enables quick development and deployment
of form-centric business processes.
򐂰 Portal applications can be provided through portlets.
򐂰 Web Application Integration Toolkit (WAT) provides an extensible framework
of reusable modules that facilitates the development of HTML based
applications.
򐂰 Process Engine API provides access to all BPM functionality.

1.4.1 IBM FileNet Workplace (or Workplace XT)

IBM FileNet Workplace (or Workplace XT) is a Web application for users,
process designers, and system administrators that provides enterprise content
management and business process management functions. Workplace comes
with a rich set of content centric functions including the ability to browse and
search for documents and process in-boxes, and provides access to IBM FileNet
Business Process Manager tools such as Process Designer and Tracker.

1
Workplace XT is the latest release of the Workplace application. Although this book deals
exclusively with Workplace, many of the features and functions that are discussed for Workplace
are also applicable for Workplace XT. For simplification, we mention just Workplace in many parts
of the book.

10 Introducing IBM FileNet Business Process Manager

 Workplace can be used as-is. You can also customize it for your business
requirements, or use it as a sample for custom application.

1.4.2 Business Process Framework (BPF)

Business Process Framework (BPF) provides a highly configurable framework
for rapidly developing IBM FileNet Business Process Manager applications.
It presents IBM FileNet Business Process Manager functions based on a
case management model. A case involves documents and processes that are
used in conjunction for a business application.

BPF supports a broad set of functions, including data entry, layout with built-in
validations, and data types. An IBM FileNet Business Process Manager Web
application can be created with simple configuration rather than coding. The
framework provides user interface components that can be configured and
customized easily, thus reducing development time and costs.

As a rapid application development tool, BPF is generally used in implementing

IBM FileNet Business Process Manager systems where time is a critical factor.
BPF is also a good fit in situations where business requirements are unclear and
are changing frequently, so an agile solution development is necessary. BPF is
very flexible in adapting to changes in the business requirements.

1.4.3 Electronic forms (eForms or Lotus Forms)

Forms increase productivity by minimizing the amount of time spent in creating
user interfaces and ensuring a rapid user acceptance through the visually rich
eForms or Lotus Forms user interface.

The user interface for a particular step in a process can be created using the
eForms or Lotus Form Designer. The resulting forms can be automated and
tightly integrated with the Process Engine without custom programming.

The form filled out by a user in a process step can be saved and stored in the
content repository as a document, which can then become a record of the user
action. In addition, data in the form can be used to populate the document
metadata or update an external database.

Forms provide features such as automatic calculations, data validations, and

error messages to assist users, and to help prevent data entry errors.

Chapter 1. Overview 11
1.4.4 Portal user interface through IBM FileNet P8 portlet
IBM FileNet P8 provides a set of predefined portlets that are compatible with
industry standard Web portals. A portlet is a user interface component that
encapsulates a small set of functions.

IBM distributes the source code for the predefined portlets, so their capabilities
can be modified and extended. In addition, users can create their own portlets
using the Content Engine and Process Engine Java™ APIs.

1.4.5 Web Application Toolkit (WAT)

Web Application Toolkit (WAT) provides an extensible framework and reusable
modules for building Web applications. The WAT reusable user interface
component facilitates the development of a robust Web-based application with a
user interface that requires little or no DHTML/JavaScript™.

WAT includes a rich set of APIs that can be used to create custom Web
applications or to integrate IBM FileNet P8 components with existing
applications. IBM FileNet Workplace is an example of a Web application that has
been developed using the toolkit.

1.4.6 Process Engine API

IBM FileNet Business Process Manager provides a full set of APIs, called
Process Engine APIs, which enable you to create applications that can take
advantage of all the IBM FileNet Business Process Manager functions. The APIs
provide configuration, management, and execution functionality. All the IBM
FileNet Business Process Manager tools and applications are written using the
same APIs exposed to you.

1.5 BPM integration to external systems

Access to external systems, applications, or databases can be established in
various ways, including but not limited to IBM FileNet P8 Component Integrator,
Web services, and business rules engines.

1.5.1 Component Integrator

Component Integrator makes it possible to interact with an external entity, a
component such as a Java object, or Java Messaging System (JMS) from a

12 Introducing IBM FileNet Business Process Manager

 process step. It includes adaptors to communicate events from the Process
Engine to external entities such as Java objects.

Note: IBM FileNet Business Process Manager consists of Process Engine,

Content Engine, and Application Engine. For information about these engines,
refer to Chapter 3, “System architecture” on page 35.

1.5.2 Web services

Web services are self-contained, self-described, modular services that can be
published, located, and invoked across the Web. In a typical Web services
scenario, a business application sends a request to a service at a given URL
over HTTP. The service receives the request, processes it, and returns a
response. With IBM FileNet Business Process Manager, a business process can
both send and receive Web services calls. Steps in a process can invoke Web
services.

Web services calls can also start a business process in IBM FileNet Business
Process Manager. A business process can be exposed as a Web service.

1.5.3 Business rules engines

Business rules engines allow customers to extract and manage business rules
outside the applications and processes. The rules framework allows business
rules engine vendors to integrate their engines into IBM FileNet Business
Process Manager.

In a process, business rules can be used in two distinct ways. The first is to
control the process itself. For example, steps can be skipped or re-executed. The
second is to implement application logic at a particular step. For example, a
business rule can be used to calculate the price of a discounted item.

The IBM FileNet P8 Platform can be integrated with industry standard rules
engines enabling process designers and business analysts to associate rules
with individual steps in a process.

1.6 BPM process monitoring and analysis

IBM FileNet Business Process Manager provides three ways to analyze
business processes: Process Simulator, Process Analyzer, and Business Activity
Monitor (BAM). The Process Simulator is used to analyze the process during the

Chapter 1. Overview 13
 development phase. The Process Analyzer is used for dynamic, chart-based
analysis of process activity at runtime. The IBM FileNet Business Activity Monitor
is used to monitor various aspects of in-progress business operations including
not only the executing processes but associated applications.

As shown in Figure 1-2, the life cycle of process improvement starts by analyzing
the current process that can be manual or automated, and identifying areas for
improvement. Simulation technology can be used to validate the process. Then
the improved process can be deployed into production, where it can be
monitored using Process Analyzer or BAM.

Deploy Analyze

IBM Software Group | Lotus software

Validate Identify

Simulate
3
Figure 1-2 Round trip process improvement

1.6.1 Process Simulator

Process Simulator is an application (tool), bundled with IBM FileNet Business
Process Manager, that is used to validate a process before placing it into
production. There are two scenarios where users would use simulation. The first
case consists of validating a brand new process, where the user provides the
information for the simulation based on experience and knowledge of the
process domain. The second case consists of validating enhancements to an
existing process, where historical data from the BPM repository is used to feed
the simulation. In either case, the user would simulate what-if scenarios and then
analyze the results of the simulation to validate the process. That way, an analyst
can test different scenarios to improve the business process before deploying it
in a production environment.

14 Introducing IBM FileNet Business Process Manager

 Figure 1-3 shows a process simulation user interface.

Figure 1-3 Process simulation

1.6.2 Process Analyzer

Process Analyzer is an application (tool), bundled with IBM FileNet Business
Process Manager, that is used to discover business processing trends, to
establish a baseline for productivity measurement, and to identify bottlenecks.
Process Analyzer monitors and analyzes events generated by the Process
Engine. The Microsoft Analysis Services, an On-Line Analytical Processing
(OLAP) solution, is used to store the aggregated results of the analysis. Process
Analyzer uses Microsoft Excel® as the default reporting tool for viewing and
analyzing the data stored inside the OLAP database. It uses Excel as a low-cost
common application that has a rich set of chart and reporting functions.

You can create your own custom reports, extending the reporting functions by
including business data that is tracked in the Process Engine event logs. In
addition to Microsoft Excel, if you require sophisticated reporting capabilities, you
can use IBM Cognos® Analysis Studio or IBM Cognos Reports.

Chapter 1. Overview 15
 Figure 1-4 shows the sample reports based on Process Analyzer output.

Figure 1-4 IBM FileNet P8 Process Analyzer output

1.6.3 IBM FileNet Business Activity Monitor

IBM FileNet Business Activity Monitor (BAM) is based on IBM Cognos
technology, and can be used to monitor various aspects of the in-progress
business operations in almost real-time. You can set thresholds and automate
actions to react to thresholds being exceeded. The business operations can be
steps that are managed by the Process Engine or activities that are managed by
other enterprise applications. You can configure BAM to collect and correlate
data from various data sources and generate a holistic analytical view of the
business operations.

Key Performance Indicators (KPIs) are financial or non-financial performance

parameters that organizations set based upon their corporate guidance. You can
set thresholds that reflect KPIs and configure BAM to send alerts and perform
actions if thresholds are violated.

Figure 1-5 shows the BAM Dashboard which illustrates various business
activities such as case arrivals, arrivals by hour, cases by product, along with
respective service level agreement (SLA) performance.

16 Introducing IBM FileNet Business Process Manager

 BPM Analytic Data
with thresholds

Real-time updates

Figure 1-5 Business Activity Monitor in IBM FileNet Business Process Manager

1.7 BPM applications and tools summary

Here is a summary of the IBM Business Process Manager applications and tools:
򐂰 Process Designer
򐂰 Process Simulator
򐂰 Process Analyzer
򐂰 Tracker (also known as Process Tracker)
򐂰 Configuration Console (also known as Process Configuration Console)
򐂰 Task Manager
򐂰 Workplace or Workplace XT
򐂰 Business Process Framework
򐂰 Electronic forms (eForms or Lotus Forms)
򐂰 Business Activity Monitor

The last three items on the list do not come with IBM FileNet Business Manager.
They are separate components that work with and support IBM FileNet Business
Manager solutions (as we have covered briefly in previous sections of this book).

Chapter 1. Overview 17
 The main tool to do process design in IBM FileNet Business Manager is Process
Designer. We discuss how to use this tool in more detail in Chapter 5, “Using
Process Designer” on page 67. Although other supporting applications and tools
are also important in improving, monitoring, and managing your business
process, it is beyond the scope of this book to cover them.

1.8 The IBM FileNet P8 family of products

In addition to IBM FileNet Business Process Manager, the IBM FileNet P8 family
of products includes back-end services, development tools, and applications that
address enterprise content and process management requirements. The IBM
FileNet P8 Platform enables users to automate business processes, access and
manage all forms of content, and automate records management to help meet
compliance requirements. It is a reliable, scalable, and highly available
enterprise platform that enables users to capture, store, manage, secure, and
process information to increase overall operational efficiency.

The IBM FileNet P8 Platform uses an open, standards-based J2EE architecture,

and is designed to fit in a variety of environments including Microsoft.NET. It
supports a large selection of databases, applications, operating systems, portals,
security frameworks, storage, systems management tools, and Web server
environments.

The IBM FileNet P8 family of products are divided into three major categories:
򐂰 Content
򐂰 Process
򐂰 Compliance

There is also a set of tools that the IBM FileNet P8 Platform provides for system
management.

1.8.1 IBM FileNet content products

The IBM FileNet P8 family of products provides comprehensive enterprise
content management functions and offers a number of products in the content
management category.

18 Introducing IBM FileNet Business Process Manager

 IBM FileNet Content Manager
IBM FileNet Content Manager manages highly customized and complex content
types while maintaining secure control over metadata and compliance activities.
At the core of IBM FileNet Content Manager are the repository services for
capturing, managing, and storing your business-related digital assets. Multiple
repositories, called object stores, can be created and managed within a single
system to serve the business requirements. Object stores can be configured to
store content in a database, a file system, a fixed content device (such as an IBM
FileNet Image Services repository, Network Appliance™ SnapLock®, or
Centera), or a combination of these options.

An object store is capable of storing a variety of business-related data — for

example, an auto insurance claim, a customer mortgage loan account, or
information about business partners. It can also store any type of structured or
unstructured content such as XML documents, Microsoft Office documents, Web
pages, photos, voice data, images, process definitions, templates, and more.

Content Federation Services (CFS)

CFS provides the capabilities to integrate with and manage content in both IBM
FileNet and non-IBM FileNet content repositories.

Content Federation Services for Image Services (CFS-IS)

Content Federation Services for Image Services (CFS-IS) natively integrates and
federates content from Content Engine and Image Services repositories. CFS-IS
enables Content Engine to use Image Services as another content storage
device. Users of IBM FileNet P8 applications have full access to content stored in
existing Image Services repositories.

Anything that is created in Workplace or created programmatically via the

Content Engine APIs can be stored in the Image Services permanent storage
infrastructure. Existing Image Services content is preserved and usable by
Image Services applications, and reusable by the IBM FileNet P8 family of
products, such as IBM FileNet Workplace and IBM FileNet Records Manager,
without duplication and without change to existing applications. The location of
document content is transparent to all applications.

1.8.2 IBM FileNet process products

The IBM FileNet P8 family of products provides enterprise process management.
The products offered in this category include IBM FileNet Business Process
Manager, IBM FileNet Business Process Framework, IBM FileNet eForms, and
IBM FileNet Business Activity Monitor. For a brief description of these products,
see 1.4, “BPM application development options” on page 10 and 1.6, “BPM
process monitoring and analysis” on page 13.

Chapter 1. Overview 19
1.8.3 IBM FileNet compliance products
IBM offers a cohesive compliance framework that combines e-mail and file
system capture with records policies and BPM processes. Using the IBM FileNet
P8 family of products, compliance can be enforced automatically without user
intervention.

There are three core products under the compliance framework: IBM FileNet
Records Manager, IBM FileNet Records Crawler, and IBM FileNet Email
Manager.

IBM FileNet Records Manager

The IBM FileNet P8 Platform and IBM FileNet Records Manager provide the
software components together in a single system architecture to assist
organizations in achieving records management compliance, with no user
involvement. The completely automated declaration and classification of records
is called ZeroClick in the IBM FileNet compliance framework. IBM FileNet
Records Manager helps solve regulatory compliance and records management
issues by helping organizations easily capture, organize, monitor, retrieve, and
retain records for improved decision making and adherence to regulatory
requirements.

Records can be automatically captured, declared, and classified, using any one
of the following methods:
򐂰 Document centric: The declaration and classification of records are based
on documents. When a file is saved to a particular folder in the IBM FileNet
P8 content repository, the system responds to certain criteria to declare and
classify the document as a record. The process can be set up to be
completely transparent to users, eliminating the necessity for additional steps
or manual decisions associated with records declaration and classification.
򐂰 Event based: The declaration and classification of records are based on
events. Following certain events or transactions associated with Web-based
or line-of-business applications, records can be automatically captured and
classified, based on pre-defined criteria or rules, such as the type of
transactions or metadata associated with the transactions.
򐂰 Process-centric: The declaration and classification of records are based on
business processes. At pre-determined points in a business process, records
can automatically be identified, declared, and precisely classified, based on
pre-defined criteria or data from the business process.

IBM FileNet Records Manager reduces the risk of litigation and provides
business continuity by enforcing corporate compliance procedures, storing only
records that are required and only for as long as they are required, and ensuring
that expired records are destroyed in a legally acceptable manner.

20 Introducing IBM FileNet Business Process Manager

 IBM FileNet Records Crawler
IBM FileNet Records Crawler is a capture tool (or application) that scans through
file systems and, based on specified rules, captures and stores electronic files
into Content Engine where they can be declared and classified as records.

IBM FileNet Email Manager

IBM FileNet Email Manager is similar to IBM FileNet Records Crawler but
processes e-mails on Exchange, Domino® and Groupwise e-mail servers. It
helps organizations easily capture, organize, monitor, retrieve, retain and share
e-mail content for improved decision making and adherence to regulatory
requirements.

1.8.4 IBM FileNet system management tools

The IBM FileNet P8 Platform offers tools that provide monitoring of system
components and applications such as the Process Engine, database, IBM
FileNet P8 Image Services. The system management tools include IBM FileNet
P8 System Monitor and IBM FileNet System Manager.

IBM FileNet System Monitor (FSM)

The IBM FileNet P8 Platform supports enterprise wide and mission critical
applications in which the availability of a system is vital. In order to achieve high
availability and maximum uptime, organizations must have hardware and
software that supports a high availability and disaster recovery infrastructure. In
addition, it is important to have a system to monitor all areas of software
operation and report any problems.

IBM FileNet System Monitor automates the monitoring of the entire IBM FileNet
P8 environment, including software components, application servers, databases,
log files, network communication and devices, and the full range of IBM FileNet
storage libraries.

IBM FileNet System Manager

IBM FileNet System Manager is a tool for collecting and distributing performance
data on the entire IBM FileNet P8 family of products, independent of the
operating system. The performance data gathered is available through a
dashboard.

Chapter 1. Overview 21
22 Introducing IBM FileNet Business Process Manager
 2

Chapter 2. Solution example

In this chapter we describe a scenario where IBM FileNet Business Process
Manager is used to optimize and streamline an existing business process that is
mostly manual, slow, and error-prone.

We cover the following topics:

򐂰 Scenario background
򐂰 Claim approval process
򐂰 Summary of benefits

© Copyright IBM Corp. 2008. All rights reserved. 23

2.1 Scenario background
Consider a fictitious company, XYZ Corporation. It is a large financial services
organization that offers a wide range of financial products and services through a
large network of field offices. They are headquartered in the United States with
satellite offices worldwide.

XYZ Corporation is looking to improve business processing and streamline

inefficiencies in their claim approval process. Their current process is partially
manual, time intensive, and prone to errors.

The current process is dependent on the documentation that comes from the
field offices. If there is a mistake, such as missing signatures, or if a document
takes too long to get into the office, it impacts the entire claim approval process.
Due to the mostly manual process, it takes a long time to get claim settlements
out to the policy holders. In addition, it is difficult to find and group all of the
documents associated with a policy holder and their open or closed claims, and
forms often get lost in the mail or in the mail room. With a long and error-prone
claim approval process, XYZ Corporation faces unsatisfied customers, and the
risk of losing customers and market share.

The process is content intensive and requires people to make decisions based
on that content; therefore, IBM FileNet Business Process Manager is the perfect
solution to streamline this business process.

2.2 Claim approval process

Let us examine how the claim approval process works in the XYZ Corporation
and see how IBM FileNet Business Process Manager can improve the overall
business process.

A policy holder has an auto insurance policy with XYZ Corporation. The policy
holder has been involved in an automobile accident and submits a claim with the
XYZ Corporation. Once the claim is submitted, it will either be approved or
denied. If the claim is approved, the policy holder will get a check to cover repairs
to their vehicle and can secure a rental car while their car is being repaired. If the
claim is denied, the policy holder will be notified and will not get paid.

24 Introducing IBM FileNet Business Process Manager

 Figure 2-1 shows the business context diagram for the claim approval process. It
illustrates the sequence of steps (numbered) throughout the entire process, the
people involved in the process, and how content (a claim package) flows through
the process to be used or processed.

3. Mailroom Clerk receives the

claim via fax, timestamps the claim
package, searches for policy
document, gets the name of Field
2. CSR takes Agent based on location of loss,
information, and and sends the claim package to
sends to head Field Agent via US mail
office via fax

1.Customer calls
and submits a
claim
Front Office
CSRs

7. Accounting
Clerk gets the
approved claim
8. Filing Clerk
package and writes
gets the closed
a check and sends
file and files it.
it to the claimant.
The closed file is
then sent for filing.

Back Office
Mailroom Clerks
Adjustors
6. Supervisor reviewes
Supervisors
Accounting Clerks 4. Field Agent gets the
the claim and either claim package, reviews
approves or rejects it. Filing Clerks it, associates police
Approved claims go to report and other related
Accounting Department
documents, then sends it
for payment to Adjustor

Field Agents
5. Adjustor receives the claim
package, reviews it, and either
approves or rejects the claim or
escalates it to Supervisor.
Approved claims go to
Accounting Department for
payment

Figure 2-1 Business context diagram for auto claim process

Chapter 2. Solution example 25

 When a customer submits a claim to the XYZ Corporation, several questions
arise within the business process and business practice, including these:
򐂰 Will the claim be approved or denied?
򐂰 If the claim is denied, what was the reason?
򐂰 How does supporting documentation get to the right person at the right time?
򐂰 What are the factors that help the XYZ Corporation make decisions more
efficiently and accurately on approving or denying claims?
򐂰 How can the turn-around time for the business process be more efficient?

Using effective business process management practices, with IBM FileNet

Business Process Manager, helps to automate and streamline the claims
approval process. Let us look into the individual stages of the process and see
how BPM improves this overall business process.

Based on Figure 2-1, the claim process can be broken down into the following
logical stages:
򐂰 Initiating a claim (steps 1 and 2)
򐂰 Back office mailroom clerk task (step 3)
򐂰 Claim processing (steps 4 through 6)
򐂰 Financial controls (part of steps 5 through 6 and 7)

2.2.1 Initiating a claim (steps 1 and 2)

An insurance claim is initiated at XYZ Corporation when a call is placed to a
Customer Sales Representative (CSR) from a customer. This is step 1 in
Figure 2-1. The customer (or the claimant) should be a policy holder whose
policy is in force. The CSR (also called agent for convenience) who takes the call
from the claimant has the responsibility to verify information about the claimant
as well as information about where and when the incident occurred. The agent
fills out all of the necessary forms to track this information and then faxes the
documents to the XYZ Corporation back office for processing. This is step 2 in
Figure 2-1 on page 25.

The current process to initiate a claim is manual and error-prone. It relies on

manual forms and paperwork to be filled out by an agent. The filed form and
paperwork for the claim are essential for starting the process, which is driven by
that content. In the current manual process, the content is passive: The agent
has to file the form and fax the information over to manually start the claim
process. In the automated solution, the content should take a more active role in
driving the business process.

26 Introducing IBM FileNet Business Process Manager

Improvements using BPM technology
Business processes in IBM FileNet Business Process Manager are capable of
interacting with content in various stages. IBM FileNet Business Process
Manager allows business processes to monitor and subscribe to content in a
repository. If a new piece of content is stored in the repository, or an old piece of
content is changed, a business process can be launched automatically. It also
allows content to be attached to the business processes. This makes the content
aware of the business processes. Using IBM FileNet Business Process
Manager, we no longer treat content and process as separate items; we
integrate them to create more effective, streamlined business process
management.

The agent takes the call from the customer and creates documentation about the
incident, such as where the auto loss occurred and when. The agent also faxes
the documentation about the loss to the back office. The faxed information
creates a document on the receiving end with its own life cycle and retention
rules. This is the content that is crucial to the approval process. The information
that is documented by the agent is what starts the claim approval process.
Because this filled-out document starts the claim approval process, it is clear that
when a new document of this type is created, a new claim process could be
started automatically. The combination of creating the initial claim document,
populating it with information, and faxing the document to the back office are the
key pieces that start the claim approval process.

We have identified the components that start the process. Now we can
streamline the process and optimize the workflow by unlocking the value of the
content. Leveraging components such as electronic forms (eForms), we can
automate the paperwork that the agent is required to fill out to initiate the claim.
Leveraging database lookups within the electronic forms, we can pull up the
customer’s information such as the address, make and model of the vehicle, and
the VIN number.

The electronic form can take items such as customer number, policy number, or
customer name, and retrieve the additional information required to start the claim
approval process. The initial claim document can be combined with additional
supporting documents about the claim and stored in the content repository
together. If there is ever a need to look up and review any of the customer’s
documents, we can find them all filed in the same system and using a single
search interface. IBM FileNet Business Process Manager leverages the stages
and life cycle of the content and makes it an active part (active content) of the
business process.

Chapter 2. Solution example 27

 Features and capabilities of BPM utilized
The features and capabilities of IBM FileNet Business Process Manager utilized
at this stage are:
򐂰 Electronic form filling (eForms)
򐂰 Database lookup
򐂰 Active content (electronically filled claims), which automatically triggers a new
claim approval process

2.2.2 Back office mailroom clerk task (step 3)

The XYZ Corporation back office works on the claim after it has been created by
an agent. After an agent fills out the necessary paperwork about the customer’s
loss, the information is sent to the back office. The back office’s mailroom clerk
receives the claim via fax, timestamps the claim package, searches for a policy
document, and gets the name of the field agent based on the location of loss.
Finally, the back office mailroom clerk sends the claim package to the field agent
via US mail. This is step 3 in Figure 2-1 on page 25.

Improvements using BPM technology

Content is an important driver in this step in the process. The information about
the customer and the loss must be documented and classified. The claim
package must be forwarded to a field agent for review and process.

IBM FileNet Business Process Manager has a routing capability that facilitates
the routing of content to other users, processes, or systems. In addition to
automatically routing the claim package to a field agent, we can also leverage
the conditional routing capability to assign the right field agent to handle the
claim, based on data such as the policy holder and the location of the loss. IBM
FileNet Business Process Manager can remove time consuming tasks such as
interoffice mail routing, manual lookup, and assignment of field agents. We can
leverage database lookups and Web services to find out how to get the right
information to the right person at the right time.

Features and capabilities of BPM utilized

The features and capabilities of IBM FileNet Business Process Manager utilized
at this stage are:
򐂰 Electronic routing of work packets (claim packages) to the right people
򐂰 Database lookup
򐂰 Automation of tasks (such as field agent assignment based on business
rules)
򐂰 Web services to get the right information to the right person at the right time

28 Introducing IBM FileNet Business Process Manager

2.2.3 Claim processing (steps 4 through 6)
A field agent receives the claim package. The field agent is responsible for
reviewing the claim package and checking for completeness and errors in the
claim. As part of the claim process, the field agent must gather supporting
documents about the customer’s loss. This supporting information can consist of
eye-witness statements, pictures from the loss location, and police statements.
The field agent is responsible for getting the supporting documents and content
to the adjustor. After the supporting documents are collected, they have to be
filed with the other documentation that has already been created for the claim.
This is step 4 in Figure 2-1 on page 25.

Delays in getting the supporting documents to the adjustor can slow or halt the
claim approval process. After receiving the claim package, the adjustor reviews
the entire claim package and decides to either approve or reject the claim.
Alternatively, the adjustor escalates the claim to a supervisor. The approved
claim goes to the accounting department for payment and record keeping. This is
step 5 in Figure 2-1 on page 25.
The adjustor might rely on several external systems to get all of the information
required to complete the customer’s claim. Working with the field agent to
process the claim, the adjustor uses the supporting documents to determine the
validity of the loss or the amount of monetary compensation that the policy holder
is entitled to receive.

If the claim escalates to the supervisor, the supervisor either approves it or

rejects it. The approved claim goes to the accounting department for payment
and record keeping. This is step 6 in Figure 2-1 on page 25.

The steps in this claim process can be delayed by dependencies on surface and
interoffice mail to get the claim packages and loss information to the field agent
and the adjustor. Other areas that can slow the process include the utilization of
external systems to locate other information required to process the claim. If the
field agent does not assign an adjustor when all the required documents are
available, the claim process can be further delayed. Automation in assigning an
adjustor and getting the right information to the field agent and the adjustor are
beneficial in removing latency from these steps.

Improvements using BPM technology

IBM FileNet Business Process Manager streamlines this part of the process by
providing automated mechanisms for the initial claim packages to get to the field
agent and later to the adjustor’s desktop. The field agent no longer has to work
with faxed documents that might be illegible and require several hours to check
for completeness and errors. Validation can be performed at the entry time and
at the application level. This way, field agents can be assured that the work is
assigned to them with the right data.

Chapter 2. Solution example 29

 IBM FileNet Business Process Manager optimizes the collection of supporting
documents by providing the field agent with a mechanism to submit and attach
content to the claim folder. When the content is added to the claim folder, IBM
FileNet Business Process Manager locates the related documents about the loss
and files them in a location that contains all of the content for the claim. This
eliminates the necessity for using costly methods such as surface or interoffice
mail to get the supporting documents to the field agent and the adjustor.

IBM FileNet Business Process Manager can also automate the classifying and
processing of this content, thus ensuring that it is filed in the correct customer
folders. This enables fast discovery of content for future claims and can help
satisfy record management and compliance requirements.

In addition to the above capabilities, IBM FileNet Business Process Manager can
interact with external systems, data sources, and applications to get the
information required to process the claim. External systems can provide the
customer’s information, such as additional policies, previously submitted claims,
and pending claims, to the field agent and the adjustor.

Features and capabilities of BPM utilized

Features and capabilities of IBM FileNet Business Process Manager utilized at
this stage are:
򐂰 Electronic routing of work packets to the right people as soon as possible
򐂰 Database lookup (for existing policy information and historical data)
򐂰 Active content with document attachment
򐂰 Foldering (adding documents to a folder or work package)
򐂰 Search the content in the system
򐂰 Interaction with the external systems and data sources to get the right
information for processing

2.2.4 Financial controls (part of steps 5 through 6 and 7)

The adjustor examines the claim package and its supporting documentation and
determines the monetary amount of the claim. The supporting documentation,
repair estimates, notes, and annotations that have been added throughout the
process are all relevant to making a determination of the amount of the loss. In
addition, the witness statements and police reports determine whether the
customer is liable for the loss.

The supervisor informs the adjustor if the claim is rejected. The adjustor then
informs the field agent of the decision. A letter outlining the reasons for the
rejection is generated and it is sent to the customer. The field agent notifies the
customer (either by phone or email) of the decision.

30 Introducing IBM FileNet Business Process Manager

If a claim is approved, the adjustor or the supervisor sends the claim to the
accounting department. The accounting department generates a payment and
sends it out to the customer. A letter is generated and sent, notifying the
customer about the approved status and stating when the payment will be sent.
The accounting department then archives the payment information along with the
claim for future audit or retrieval. This is step 7 in Figure 2-1 on page 25.

Improvements using BPM technology

IBM FileNet Business Process Manager can integrate electronic forms, content,
and business processes to create solutions that are logical and intuitive for users
to use. The adjustor uses the content in the claim package to determine whether
the claim amount is within the limit to be automatically approved. If the monetary
amount of the claim is within a pre-determined range, the adjustor has the ability
to automatically approve it. If the claim amount exceeds that limit, the adjustor
escalates it to a supervisor. The supervisor has the additional tools and authority
to make decisions on whether or not to approve a claim.

Using IBM FileNet Business Process Manager, we can use the routing capability
to automate the way the work goes from an adjustor to a supervisor. The routing
rule can be implemented with conditions that make it seamless for work to be
routed from the adjustor to the supervisor. After the work is completed (approved
by the supervisor if required and by the adjustor), it can resume its progress
along its defined path in the workflow, or it can be sent back to the point where it
started. IBM FileNet Business Process Manager also has the capability to
escalate the claim to another person such as a manager, if a certain amount of
time has expired (for example, three days).

IBM FileNet Business Process Manager has the ability to interact with external
systems and processes to automate the creation of letters or the generation of
payment for customers. After the letters, invoices, or purchase orders are
created, IBM FileNet Business Process Manager can automatically file the letter
along with the other documents from the claim in one place, making it simple to
find all related content items for the customer or the claim.

Features and capabilities of BPM utilized

The features and capabilities of IBM FileNet Business Process Manager utilized
at this stage are:
򐂰 Routing with routing rules to automatically route work packets to the
appropriate people
򐂰 Wait and resume capability in a business process
򐂰 Automation of tasks (such as printing of letters to customers)
򐂰 Interaction with external systems (such as interaction with an accounting
system for check printing)

Chapter 2. Solution example 31

 򐂰 Filing a collection of related documents into a folder (for example, all claim
related documents can be filed into one place for easy retrieval)

2.2.5 Workflow process maps for the claim approval process

With IBM FileNet Business Process Manager, we can create workflow process
maps that automate and facilitate the claim approval process.

Figure 2-2 shows the main workflow process map that drives the claim approval
process.

Figure 2-2 Main workflow process map for the claim approval process

The activities of setting up a claim case that involves creation of a case (folder)
for the claim, assigning of a field agent, gathering of supporting documents, and
assigning of an adjustor are encapsulated in a workflow subprocess, which is
implemented as a separate submap. Figure 2-3 illustrates this claim setup
submap.

Figure 2-3 Claim set up workflow process submap for the claim approval process

32 Introducing IBM FileNet Business Process Manager

 We cover process design, including the creation of a workflow map and
submaps, in more detail in later chapters of the book.

2.3 Summary of benefits

In this claim approval process example, we showed that by using IBM FileNet
Business Process Manager, we can create the content (claim document) and
automatically start a business process (claim approval process). IBM FileNet
Business Process Manager allows an organization to build processes that can
interact with content to provide higher-value (more accurate and efficient)
solutions. It provides the structure required to optimize the claim approval
process.

By automating the process and using routing of the work to different people,
work (the claim package) can move seamlessly between field agents, adjustors,
supervisors, and the accounting department. IBM FileNet Business Process
Manager also interacts with external systems, making the generation and
distribution of letters to policy holders a simple, automated process. The process
is made more efficient and less error prone. Reporting and audit and archival are
simplified.

The features and capabilities of IBM FileNet Business Process Manager

showcased in this solution example include:
򐂰 Electronic form filling (integration with eForms, another IBM FileNet P8
product)
򐂰 Active content: A new claim automatically triggering a new business process
򐂰 Active content: Document attachment (for the supporting documents) to the
existing work packet
򐂰 Electronic routing of work packet to the right people as soon as possible
Example:
– Routing of work packet (claim package) to an adjustor or agent depending
on where the step is in the process
– Routing of work packet based on the claim amount or other routing rules
– Escalation in routing, if a certain condition is true
򐂰 Wait and resume capability for a business process
򐂰 Database lookup (for existing policy information and historic data)
򐂰 Foldering, filing, and adding documents to a folder or work packet
򐂰 Efficient search of the content in the system

Chapter 2. Solution example 33

 򐂰 Automation of tasks
Example:
– Assigning a field agent to a claim based on the location of the accident
– Assigning an adjustor to a claim based on business rules
– Printing a letter of the claim status to the customer
򐂰 Web services and interaction with the external systems and data sources to
get the right information to the right person at the right time

34 Introducing IBM FileNet Business Process Manager

 3

Chapter 3. System architecture

In this chapter we describe the underlying architectures for the IBM FileNet P8
Platform and IBM FileNet Business Process Manager (BPM). We show how
BPM technology under the IBM FileNet P8 Platform is leveraged to deliver
automated content centric processing to businesses.

We cover the following topics:

򐂰 IBM FileNet P8 core engines
򐂰 Layered architecture
򐂰 Process Engine logical view

© Copyright IBM Corp. 2008. All rights reserved. 35

3.1 IBM FileNet P8 core engines
There are three primary engines that make up the IBM FileNet P8 Platform. For
our purposes, we define an engine as a collection of services and components,
which perform a set of related functions. Although an engine is composed of
many parts, we view it as a single functional unit. Understanding the complexity
of how each engine works is not necessary to gain an appreciation for how a
business process is defined, developed, and managed in the IBM FileNet P8
environment. However, it is important to know what each engine does.

The three core engines for the IBM FileNet P8 Platform are:
򐂰 Content Engine: The Content Engine provides software services for
managing different types of business-related unstructured content, which we
refer to as objects. It manages one or more object stores. An object store is a
repository for storing objects in an IBM FileNet P8 environment.
򐂰 Process Engine: The Process Engine components allow you to create,
modify, and manage automated business processes. These processes are
performed by applications, enterprise users, or external users, such as
partners and clients.
򐂰 Application Engine: The Application Engine hosts the Workplace Web
application, Workplace Java applets, and Application Programming Interfaces
(APIs). It is the presentation tier for both process and content.

Figure 3-1 illustrates the three IBM FileNet P8 engines and their relationships.

Application Engine

Process Engine Content Engine

Figure 3-1 IBM FileNet P8 engine relationships

36 Introducing IBM FileNet Business Process Manager

3.2 Layered architecture
IBM FileNet Business Process Manager is designed around content centric
process management and ease of use paradigms. The design goal is to provide
a BPM solution that can be easily used by business analysts to take control of
the business processes of their organizations. Business analysts are responsible
for authoring, managing, and monitoring the processes implemented in the IBM
FileNet Business Process Manager solution.

IBM FileNet Business Process Manager is designed as a four-tier architecture as

shown in Figure 3-2:
򐂰 Client tier
򐂰 Presentation tier
򐂰 Middleware tier
򐂰 Data tier

Web Client
Application
Client Tier

Application Engine Presentation Tier

Process
Engine Middleware Tier

Process Data Tier

Engine
Database

Figure 3-2 IBM FileNet Business Process Manager four-tier architecture

Chapter 3. System architecture 37

 Most organizations install IBM FileNet Business Process Manager solutions
using different servers for each tier, but tiers can be collocated into fewer
servers.

In the IBM FileNet P8 Platform, the functionality is exposed using the same tiered
architecture with other components in each tier. Those extra components provide
valuable features for the BPM functionality, including content management,
electronic forms, records management, collaboration functionality, and more.
Figure 3-3 shows a high level architecture of the IBM FileNet Business Process
Manager components.

Presentation Tier

Java Connector
...
PE Applications eForms

JMS Connector CE Operations BAM

Forms Dashboard
Workplace
Manager
Component Integrator

Java API

Middleware Tier
WS API

Content Process
Engine Engine Process Process Rules
BAM
Analyzer Simulator Framework

Data Tier

Content Process Process

Directory Engine Engine OLAP Analyzer
Service Database Database Database

Figure 3-3 High level architecture of IBM FileNet Business Process Manager
components

38 Introducing IBM FileNet Business Process Manager

3.2.1 Presentation tier
The presentation tier contains most of the IBM FileNet Business Process
Manager applications and tools that users need to interact with the system. The
APIs, external communication, and customer-written business logic are exposed
in this tier. In IBM FileNet P8, this tier is implemented by the Application Engine
and it is hosted in a Java Application Server.

Component Integrator
The Component Integrator is used to integrate Java or Java Message Service
(JMS) components for use in processes. Components are registered in the
Process Configuration Console to make them available in Process Designer.
Figure 3-4 shows a Java component that has been registered.

Application Engine

Java Component

Java adaptor

Component Integrator
Service Manager

Process Task Manager

Process Engine

Figure 3-4 Registered Java component with Component Integrator

Chapter 3. System architecture 39

 Process Orchestration
Process Orchestration provides the Web services functionality of IBM FileNet
Business Process Manager. It provides the ability for processes to invoke Web
services (WS), and to receive Web services calls. Most of the process
orchestration functionality is implemented in the middleware tier, but the
communication part is implemented at the presentation tier.

Figure 3-5 shows the Process Orchestration system architecture.

Application Engine

Incoming message
WS Listener

WS Adapter
Outgoing message
Component Integrator
Service Manager

Process Task Manager

Process Engine

Figure 3-5 Process Orchestration system architecture

The Process Orchestration component can be started by the Task Manager and
it is composed of a Web services Listener that receives all the incoming
messages and a Web services Adaptor that is used to send outgoing messages.

40 Introducing IBM FileNet Business Process Manager

3.2.2 Middleware tier
The middleware tier contains the core of the BPM system as shown in
Figure 3-6.

Middleware tier
WS API

Content Process Business

Engine Engine Rules
Activity
Framework
Monitoring
CE Kernel PE Kernel

Process Email Process Process

Auto launch Notification Analyzer Simulator

Figure 3-6 Middleware tier

The IBM FileNet BPM middleware tier includes the following components:
򐂰 Content Engine (CE) Kernel: The core Content Engine.
򐂰 Process auto-launch: The event-driven component that launches process
instances in the Process Engine system. This component can be configured
to launch a specific process version in response to specific events. For
example, filing a document in a folder can launch a particular process in the
Process Engine system.
򐂰 Process Engine (PE) Kernel: The core engine of the BPM system that
includes the execution sub-system.
򐂰 E-mail Notification: Enables automatic transmission of e-mail to users when
specified process-related events occur. E-mail notification can also be used to
track processes.
򐂰 WS API: Provides a Web services API to the Process Engine and to the
Content Engine.
򐂰 Rules Framework: Provides a framework for rules engines to integrate with
IBM FileNet Business Process Manager. The framework uses a well-defined
rules interface that must be implemented to integrate with rules engines.
򐂰 Process Analyzer: Provides analysis capabilities to determine cycle times,
find trends and bottlenecks, and generate reports and charts to analyze the
processes deployed in the Process Engine system.

Chapter 3. System architecture 41

 򐂰 Process Simulator: Allows customers to simulate processes and perform
what-if scenarios with hypothetical or historical data.
򐂰 Business Activity Monitor: Provides monitoring of various aspects of the
in-progress business processes in the context of other business applications
in almost real-time.

3.2.3 Data tier

The data tier contains the BPM databases and directory service as shown in
Figure 3-7.

Data Tier

Content Process Process Process

Directory Engine Engine Analyzer Analyzer
Service Database Database Database OLAP

Figure 3-7 Data tier

The data tier includes the following components:

򐂰 Directory Service: Provides authentication. The directory service
functionality is accessed through Content Engine. Content Engine supports
several LDAP providers and single-sign-on (SSO) functionality. (Note that
in versions prior to Version 4.0, IBM FileNet Business Process Manager had
direct access to the LDAP directory.)
򐂰 Content Engine database: Contains object stores. Each object store
contains content and metadata about the content.
򐂰 Process Engine database: Contains configuration and running instances of
processes.
򐂰 Process Analyzer database: Stores analytical data of executing processes.
򐂰 Process Analyzer OLAP: Contains aggregation of data stored in the
Process Analyzer database for the purpose of enabling quick analytical
charting and reporting.

42 Introducing IBM FileNet Business Process Manager

3.3 Process Engine (PE) applications and other
supporting applications
The Process Engine applications are implemented as applets and are hosted by
IBM FileNet P8 Workplace in the Application Engine. With the exception of the
Task Manager and Process Simulator applications, all the other applications use
the Process Engine APIs. You can write similar applications with these APIs.

Figure 3-8 shows some of the Process Engine applications.

Process
Process Process Task
Simulation
Designer Administrator Manager
Designer

Process Process
Step
Configuration Tracker Simulation
Processors
Console Console

Figure 3-8 Some Process Engine applications

The Process Engine applications include:

򐂰 Process Designer: Used to create, modify, and deploy processes.
򐂰 Process Configuration Console: Used to prepare and configure an IBM
FileNet Business Process Manager system.
򐂰 Process Administrator: Used to query and modify executing processes.
򐂰 Tracker (also known as Process Tracker): Used to look at the history of an
executing process.
򐂰 Process Simulation Designer: Used to create simulation scenarios of
business processes.
򐂰 Process Simulation Console: Used to manage and execute simulation
scenarios.
򐂰 Task Manager (also known as Process Task Manager): Used to manage
some aspects of the IBM FileNet Business Process Manager system
including the Component Integrator.
򐂰 Step Processors: Used to provide users with an interface to complete steps
in a process. There are generic step processors provided by the IBM FileNet
Business Process Manager environment. eForms and Lotus Forms can also
be used as step processors. Specialized step processors can also be written
using the Process Engine APIs.

Chapter 3. System architecture 43

3.3.1 Process Simulator system architecture
Process Simulator can be accessed from the Process Simulation Designer and
the Process Simulation Console. The Designer is used to create simulation
scenarios that can be saved and versioned in the Content Engine repository. The
Console is used to manage the scenarios and to execute simulations. The result
of a simulation execution is twofold: the simulation analytics are stored in a
Process Analyzer database and the simulation object, containing animation
information, is stored in the Content Engine for future animations.

Figure 3-9 shows the Process Simulator system architecture.

Application Engine
Simulation Simulation
Designer Console

Scheduler
Content Process
Engine Publisher Engine

Process Simulator

Content Process Process

Engine Analyzer Engine
Database Database Database

Figure 3-9 Process Simulation system architecture

3.3.2 Process Analyzer system architecture

Process Analyzer has an Event Dispatcher that retrieves the events from the
event logs in the Process Engine database, and a Publisher that processes the
events. In addition, the Publisher retrieves the configuration data required to
process the events. The configuration information includes the process
definitions, queues, and users who are present in the event. At the end of the
publishing interval, the Publisher updates the fact tables in the database.

44 Introducing IBM FileNet Business Process Manager

 The statistical data in the fact tables is used to build the OLAP cubes to provide
the information required by reporting tools such as Microsoft Excel, IBM Cognos
Analysis Studio, or IBM Cognos Reports.

Figure 3-10 shows the Process Analyzer system architecture.

Event Dispatcher

Publisher

Process Analyzer

Process Process
Analyzer OLAP Engine
Database Database

Figure 3-10 Process Analyzer system architecture

3.3.3 Business Activity Monitor system architecture

Business Activity Monitor (BAM) is based on IBM Cognos technology. It collects
data from Process Analyzer through its built-in connection. It can also collect
data from external applications through standard connections such as JMS,
JDBC™, Web services, and EAI. The collected data can be either raw, as in the
case of Process Analyzer, or context data. Context data is data used to correlate
with raw data. For example, a raw data record containing an account number can
be used to retrieve the account user name which corresponds to the context
data.

The BAM component has an in-memory streaming capability, that is configured

to generate in-memory views and OLAP cubes. Views of the data can be
created, and rules can be applied. The rules can be used to calculate Key
Performance Indicators (KPIs) and alerts.

Chapter 3. System architecture 45

 The BAM Dashboard can be configured to display information stored in the views
and cubes, and to create thresholds based on KPIs, configure alerts, and to
perform actions if a threshold is violated. Figure 3-11 shows the BAM Dashboard
system architecture.

BAM Dashboard

Data Application Workbench

Process Analyzer In memory streaming database

Raw
Data
JMS / JDBC Alerts
Views
& Rules
Web Services Cubes
Context KPIs
Data
EAI

Figure 3-11 Business Activity Monitor system architecture

3.3.4 Business Process Framework system architecture

Business Process Framework (BPF) is a configurable framework for developing
BPM Web client applications. The resulting Web client provides a feature-rich
interface using HTML, Java Server Pages (JSP™), JavaScript pages, and other
objects.

BPF applications are built based on a case paradigm. For example, an auto
insurance claim can be considered a case. The generated application uses both
Content and Process Engine APIs. The application user interface is
metadata-driven and is built dynamically upon user request. The metadata
describing the user interface layout is stored in the BPF configuration database
and is managed via the BPF Explorer configuration tool.

46 Introducing IBM FileNet Business Process Manager

 Figure 3-12 shows the BPF system architecture.

Application Engine
BPF Web
Workplace BPF Explorer
Application

Content Process
Engine Engine

Content Process BPF

Engine Engine Configuration
Database Database Database

Figure 3-12 Business Process Framework system architecture

3.4 Service Oriented Architecture (SOA)

IBM FileNet Business Process Manager provides a flexible architecture that
organizations can leverage to implement business processes as Web services
within enterprise architectures. IBM FileNet Business Process Manager solutions
can connect processes with applications and systems using Web services and
other standard integration methods. Organizations can increase the speed of
development and create new business process management applications to
support larger Service Orientated Architecture (SOA) initiatives.

IBM FileNet Business Process Manager processes can be published and

consumed as Web services. The Process Engine leverages Web services
technology to enable process designers to define message interactions internally
between several IBM FileNet Business Process Manager business processes,
as well as externally between IBM FileNet Business Process Manager and other
systems.

Chapter 3. System architecture 47

 Web services
Web services are the primary technology used in implementing Service Oriented
Architecture. Web services have characteristics that make them the ideal
building block for service oriented architecture. Web services are made up of a
common set of protocols and standards. These define how Web services are
published, discovered, and implemented in a standard format.

Figure 3-13 shows how a basic Web service is used. A client looks up a Web
service from a UDDI registry and obtains its WSDL. It then invokes the Web
service. In addition to UDDI, IBM FileNet Business Process Manager also
supports the IBM WebSphere Service Registry and Repository (WSRR).

Figure 3-13 Web services usage

Simply having a Web service or a Web services API does not imply Service
Oriented Architecture. It is the ability to orchestrate these Web services that is
fundamental in building or integrating with Service Oriented Architecture. The
ability to consume and publish Web services to create higher value services is an
essential aspect of a Service Oriented Architecture. IBM FileNet Business
Process Manager provides the facilities to participate in a Service Oriented
Architecture.

48 Introducing IBM FileNet Business Process Manager

 Consuming Web services
An external Web service can be consumed by a business process by the use of
an Invoke step. The Invoke step binds the process to the Web service. The ability
to consume Web services in business processes allows the reuse of
components from existing applications and this is key in building a Service
Oriented Architecture.

Publishing Web services

IBM FileNet Business Process Manager provides constructs for publishing
processes as Web services. A Web service can be exposed with the use of a
Receive step. There can also be one or more Reply steps used to handle the Web
services reply.

Publishing business processes as Web services allows business processes

defined in IBM FileNet Business Process Manager to become reusable in other
applications.

IBM FileNet Business Process Manager can inter-operate with enterprise

architectures such as Enterprise Services Buses (ESB) and Web services
management systems. IBM FileNet Business Process Manager can also help
organizations automate and manage business processes and can be leveraged
to build Service Oriented Architecture environments.

3.5 Process Engine logical view

At the core of IBM FileNet Business Process Manager is the Process Engine
(PE). From a logical view, the Process Engine can be seen as a collection of
isolated regions1, each one containing queues, rosters, and event logs.
Processes are compiled and stored in an isolated region. Each process is
defined as a collection of maps and the fields required to define the process
behavior. Maps are a collection of steps organized into a directed graph required
to complete a process. Figure 3-14 illustrates the Process Engine logical view.

1
For an explanation of isolated regions, refer to “Isolated regions” on page 50.

Chapter 3. System architecture 49

 Process Engine System Process Definition
Fields

Map
queue roster Event
log
Queue Roster Step
Log Step Step
Step

Isolated Region

Compile
queue roster Event
log
Queue Roster
Log
Work
Item

Isolated Region Process Instance

Figure 3-14 Process Engine system logical view

Process server
A process server is a physical machine designated as a member of a Process
Engine system. A Process Engine system is a farm of Process Engine servers,
with one or more servers. A process server can only be a member of one
Process Engine system. The process server runs the Process Engine software.

Isolated regions
An isolated region (as illustrated in Figure 3-14) is a logical subdivision of the
Process Engine database. Isolated regions are often used to provide separate
development, testing, and production environments which enable developers
and testers the ability to efficiently utilize shared computing resources and
provide the necessary isolation between these disparate activities. An isolated
region consists of queues, rosters, and event logs.

50 Introducing IBM FileNet Business Process Manager

Queues
A queue is a database structure that holds work items. There are three types of
queues: user queues, work queues, and system queues:
򐂰 A user queue is a queue that holds work items waiting to be processed by a
specific user.
򐂰 A work queue is a queue holding work items that can be completed by one of
a number of users, rather than by a specific participant, or work items that can
be completed by an automated process.
򐂰 System queues are queues holding work items that are undergoing or waiting
for processing by the Process Engine server.

A roster is a database structure that stores the current location and other
information about a work item. Process rosters provide the Process Engine
software with an efficient way to locate a specific work item. An event log is a
database structure that contains information about system-level events related to
work item processing.

Process definition
A process definition (as illustrated in Figure 3-14 on page 50) describes a
business process. It includes information such as fields and maps used in the
process.

A map is a representation of a process that defines the sequences of steps

needed to complete the business process. A step represents a stage in a
process where a distinct, well-defined action is performed. On a process map,
each step represents a specific activity or task in the business process described
by the map. Routes between the steps indicate one or more possible processing
paths. A field is a property that can be used in a process to hold data.

Process instance
A process instance corresponds to a single execution of a process definition.
For example, the auto insurance claim for Account 120-021 is an instance of the
claim process definition, where the Account field has the value 120-021. A
process instance consists of work items.

A work item is a single unit of work composed of a collection of fields. Work items
transverse the process map, moving the data required by the process from step
to step as indicted by the map.

Chapter 3. System architecture 51

52 Introducing IBM FileNet Business Process Manager
 4

Chapter 4. Process design with BPM

In this chapter we discuss the basic elements and concepts that are used in
designing business processes with BPM. Our discussion is not meant to cover
every topic, but to provide the foundation for designing a workflow process with
BPM.

We cover the following topics:

򐂰 Basic concepts in process design:
– Tasks, participants, and roles
– Workflow properties (workflow groups and attachments)
– Steps
– Routing
– Milestones
– Deadlines
– Task reassignment
– Process voting
򐂰 BPM supported data types and system functions
򐂰 Inter-related processes
򐂰 Interaction with external systems

© Copyright IBM Corp. 2008. All rights reserved. 53

 Notes:
򐂰 The example presented in the current chapter derives from the sample
solution we described in Chapter 2, “Solution example” on page 23. We
recommend reading that chapter before proceeding with this one.
򐂰 After absorbing the basic concepts from the current chapter, read
Chapter 5, “Using Process Designer” on page 67 to learn how to use
Process Designer to design your business processes.
򐂰 Then, in Chapter 6, “Implementing business processes: Case study” on
page 127, we provide step-by-step instructions for actually implementing
business processes with a case study.

54 Introducing IBM FileNet Business Process Manager

4.1 Basic concepts in process design
IBM FileNet Business Process Manager comes with the Process Designer tool
that enables you to design business processes for your business needs. Before
using the tool, it is important to understand the key elements in a business
process:
򐂰 Tasks, participants, and roles
򐂰 Steps
򐂰 Routing
򐂰 Deadlines
򐂰 Task reassignment
򐂰 Process voting

4.1.1 Tasks, participants, and roles

A task is a unit of work, inside or outside of an organization, which has a defined
start and finish. Tasks are performed by people, departments, or systems, and
are a part of business processes. A task can be something small and atomic,
such as a person making a phone call; or it can be much more complex, such as
renting a car.

People who perform tasks are participants for the defined tasks. In a business
process, participants are assigned specific work that must be completed at one
or more steps in the workflow. Participants can be people, a group of people, a
department, or a system.

Participants who perform similar or related tasks can be grouped together in

roles. A role is a collection of tasks assigned to, or expected of, a person or
group. When a role is defined, it identifies all of the work that the participants will
do, as well as the tasks that are associated with the step in the process.

In Chapter 2, “Solution example” on page 23, we described a sample solution for

XYZ Corporation, where we automate their auto insurance claim approval
process using IBM FileNet Business Process Manager. Let us revisit this
example to better understand tasks, participants, and roles in a business
process.

From the auto claim approval process example, we identify the following tasks:
򐂰 Interact with policy holders.
򐂰 Create and submit claim packet.
򐂰 Review submitted claims packets.
򐂰 Assign field agents to claims.

Chapter 4. Process design with BPM 55

 򐂰 Gather supporting documents for claims including police reports, witness
interviews, and photographs.
򐂰 Review and process claims.
򐂰 Send claim packets to adjustors.
򐂰 Approve or rejects claims.
򐂰 Issue funds for claim settlement to policy holders.
򐂰 Classify claims received in the back office.

From the auto claim approval process example, we also identify the following
participants:
򐂰 Agents
򐂰 Adjustors
򐂰 Field agents
򐂰 Accounting department
򐂰 Administration department
򐂰 Finance department
򐂰 Mail room

Roles tie tasks to participants. From our example, we identify the following roles
in the claim approval process:
򐂰 Agents who interact with policy holders and create and submit claim packets
򐂰 Adjustors who review the claims packets and assign field agents to claims
򐂰 Field agents who gather supporting documents for claims
򐂰 Accounting department that processes claims and issues claim settlement
funds to policy holders
򐂰 Administration department that performs administrative tasks such as
classifying claims that are received in the back office, and sends initial claim
packet to the adjustors
򐂰 Finance department that approves or rejects claims
򐂰 Mail room that classifies and processes claim packets that are received from
the field agents

So in the XYZ Corporation example, employees and departments have tasks

associated with them. In addition, internal or external systems might also have to
perform tasks. We discuss that topic later.

As we stated before, a task is a unit of work, inside or outside of the organization,

with a defined start and a finish, performed by people, departments, or systems.
Part of designing a business process is to identify all the tasks, participants, and
their roles in the process.

56 Introducing IBM FileNet Business Process Manager

4.1.2 Workflow properties (workflow groups and attachments)
Workflow properties specify the subject and other general properties of a
workflow. Some of the workflow properties include workflow groups,
attachments, and data fields that can apply to any step in a workflow. The
workflow properties apply to the entire workflow definition.

Workflow group
A workflow group represents a collection of one or more users or groups that
can be assigned as workflow participants. The members of a workflow group
typically perform a particular task or a set of tasks in a business process, for
example, claims adjustors or supervisors.

For flexibility in defining a workflow, you can either assign one or more specific
participants to a workflow group, or allow the participants to be assigned later,
either as part of the launch process or at a particular step when the business
process is running. This technique of using unspecified workflow groups in a
workflow definition is useful when the participant for a particular task is likely to
change each time the business process runs.

System workflow groups

BPM provides system workflow groups. F_Originator is a special workflow group
that represents the name of the user who launched the business process. At
launch time, the system assigns the name of the launch user to that workflow
group.

All assigned members of the F_Trackers group become workflow trackers, with
access to the Process Tracker application. Although it is not required, every
workflow definition should have one or more users or groups assigned as
trackers to monitor events and help resolve problems when a process is running.

Attachments
An attachment is a link to information that a participant uses to complete a step in
a workflow. The specific item that an attachment links to is referred to as a target
in the Process Engine documentation. The most common target is a document
located in an object store. However, document arrays, stored searches, folders,
URLs, or files located on a shared disk can also be targets.

You define attachments as part of the workflow definition properties, and then
indicate which attachments will be used at each step.

An attachment can be designated as an initiating attachment. This designation

means that the document used to launch the workflow will be assigned to the
selected attachment. A workflow definition can have only one initiating
attachment.

Chapter 4. Process design with BPM 57

4.1.3 Steps
Steps represent specific business activities or system activities. In each step, you
can assign participants (individuals, groups, and systems) to perform tasks.

A business process consists of a series of steps to be completed. How you

actually create the specific steps (from the tasks and participants) in a business
process depends on what is logical, necessary, and convenient for your business
process application.

Using our claim approval example, we can design a matching step for each
identified task in the claim approval process. If a task is complicated, you can
break the task into multiple subtasks, thus multiple steps.

4.1.4 Routing
Routing specifies the sequence in which steps are executed in a business
process. A route represents the path between two steps and defines the order in
which steps will be processed. Aside from the first step and the last step in a
business process, all other steps can have one to many inbound routes and one
to many outbound routes.

Routing options and responses can be defined in a business process. The

routing options are evaluated when the step is completed and they determine the
next step in the business process. A list of available responses can be defined in
each step so that the participants can select them as part of the work completed
at the step. The response that a participant selects can also determine the next
step in the business process. For example, if a participant approves a claim (that
is, selects the approve response), the business process can take the approval
route to complete the remainder of the work. If the participant rejects the claim,
then a different route will be used.

There are three ways to incorporate routing options and responses in a business
process:
򐂰 Single (linear) routing
򐂰 Conditional routing
򐂰 Parallel routing

Single routing describes a route in a business process where work can only
continue along a single path. At most, one step in the workflow is active. There
can be one or more outbound routes defined for a step, but only one step can be
active at a time.

58 Introducing IBM FileNet Business Process Manager

Conditional routing allows you to specify a condition on a route. A condition can
be based on a response from a participant or evaluation of a data expression. If
an expression evaluates to true, then that route is taken. An expression can also
be a combination of response and data fields.

Multiple steps can simultaneously be split out to be processed in parallel during a

business process. If steps do not depend on each other, then they can be
processed in parallel. A step can have multiple outbound parallel routes, which
results in multiple parallel steps. When all parallel steps are completed, the
outbound routes from these steps are merged together in a collector step and the
business process continues from there. The collector step is a special type of
step that collects the parallel routes into one step. If data values vary from
processing along different paths, then different merge options determine which
values result in the collection step.

Let us see how different routes can be used in the claim approval example. We
have the following two steps:
򐂰 Step A: An agent documents a loss due to an auto accident by a policy
holder, and submits a claim packet to the back office.
򐂰 Step B: The mail room clerk logs and adds a timestamp on the claim packet
in the back office.

The routing between these two steps is straightforward, single routing. There is a
single active path from one step to the next. The task in the first step must be
completed by the agent before the task in the second step can be completed by
the mail room clerk.

Continuing with the same example, with some deviation for the purpose of
explanation, we assume that after the mail room clerk logs the claim in the
system (Step B), we have the following steps:
򐂰 Step C: A field agent needs to assess and report property damage.
򐂰 Step D: A medical reviewer needs to assess and report injuries resulting from
the accident.
򐂰 Step E: When the property damage report and the injury report are filed, then
the adjustor decides the claim amount.

Since Step C and D are two steps that are independent of each other, we can
design them to work in parallel. Thus, we can use parallel outbound routing from
Step B, to go to Step C and Step D in parallel. Then we can have a collector step
to ensure that both finish before it routes to Step E.

Routes ensure that work is sent to the right participants to be processed. The
combination of defined steps and routes becomes a template that is used each
time the business process runs.

Chapter 4. Process design with BPM 59

4.1.5 Milestones
Milestones are defined as key notification points in a business process. These
are points at which participants need to receive a message describing an
important task or tasks that are pending or completed.

Each milestone can be used in one or more steps. The status message can be
either specific text, or it can include the values of one or more data fields. When
the workflow process reaches a milestone, the specified message can be
displayed for participants and users who launched the workflow.

For our claim approval example, we do not use any milestones.

4.1.6 Deadlines
Deadlines are time-based constraints that can be applied to steps in a business
process. A step with a deadline implies that the participant must complete the
step within a specified amount of time. The specified time (the deadline) is
relative to the time that the step starts (when the participant receives the work
packet).

A participant that is routed a task with a deadline can receive a reminder of the
pending deadline via e-mail. When a deadline is reached, e-mail messages are
sent to the step participants (with the uncompleted step) and the workflow
trackers. Users can update their e-mail notification preferences to turn off
different types of e-mail notification. When a deadline expires, it can also be set
to trigger escalation such that the work packet is routed to a special escalation
step to be performed by a supervisor or some other participants. Using a
deadline provides added control and an escalation mechanism for a business
process.

Using our claim approval example, we have the following step that might require
implementation of a deadline:
򐂰 Step: The adjustor assigns a field agent.

The adjustor assigns a field agent. If this task does not happen in a timely
fashion, it impacts the entire claim approval process. By adding a deadline on
this step, we can add some time management to the step to ensure that the
adjustor is not delaying the claim approval process. We can also set the
adjustor's supervisor as the tracker for the workflow (by adding the supervisor to
the F_Trackers group for the workflow). When a deadline or a reminder is
reached, both the adjustor and the supervisor are notified. Again, receivers of the
reminder have options to turn off different types of e-mail notifications.

60 Introducing IBM FileNet Business Process Manager

 When a deadline expires on a step and a submap is specified, the work is routed
to the first step in the submap.

Of course, we can also automate the field agent assigning step such that the
adjustor does not manually assign a field agent to the claim; rather, the system
automatically assigns a field agent based on the place of the incident. In this
scenario, a deadline is not required.

The task of gathering supporting documents for a claim might be a good step in
which to apply a deadline. Although a participant might not always be able to
complete a task in a timely manner due to factors outside of their control,
deadline mechanisms (including reminder e-mails and automatic escalations)
provide effective time management and control to ensure more timely completion
of a step.

4.1.7 Task reassignment

For flexibility and control in a business process, you can design a step to enable
the original participant to dynamically reassign the tasks to other participants in
the step. Specifically, a participant can assign another participant to take over the
task or delegate the work to another participant. In the IBM FileNet Business
Process Manager context, when reassigning tasks within a step, two options are
available, delegation and abdication:
򐂰 When using the delegation option, a participant reassigns the task to another
participant. After the task is complete, the work packet comes back to the
original participant for review before the work packet moves on to the next
step in the business process.
򐂰 When using the abdication option, a participant reassigns the task to another
participant. After the task is complete, the work packet is directly sent to the
next step in the business process.

Reassignment can be disabled on any user step in the workflow.

4.1.8 Process voting

There are instances when the system requires input from multiple participants to
determine the next step in the business process. This calls for process voting.
When a process voting is implemented, each participant will vote and the results
of the voting will determine the next step in the business process.

You can implement the routing decision based on one of the following conditions:
򐂰 When all participants select the same response (for example, when all
approve)

Chapter 4. Process design with BPM 61

 򐂰 As long as one participant responds a certain way (for example: when at least
one person rejects).
򐂰 When none of the participants select a particular response (for example:
no one rejects).
򐂰 Simple majority, when most participants select a certain response (for
example: most participants approve rather than reject).

Let us see how this can be applied to the XYZ Corporation’s claim approval
business process example (we will deviate somewhat from the actual example
for the purpose of explanation).

We assume that, in the claim approval process, there is a step when the claim
approval decision task is escalated to a supervisor if the claim is determined to
be a high risk (for example, the claim amount is extremely high). According to the
business rules in place, multiple risk managers make the approval decision.
The supervisor thus routes the decision task to the risk managers who are
experienced in dealing with assessing risks. The risk managers review the claim
and then each either approves or rejects the claim. Per the business rule in the
example, as long as the majority of the risk managers approve, the claim is
deemed approved. In this case, we would implement the simple majority
condition for processing voting.

The process voting function allows organizations to implement collaborative

decision making while still keeping within a structured business process.

4.2 BPM supported data types and system functions

IBM FileNet Business Process Manager (BPM) supports a set of data types and
provides a set of powerful system functions that you can use for your process
design.

4.2.1 Supported data types

BPM supports literals and variables of the following simple and multiple (arrays)
data types:
򐂰 Boolean
򐂰 Float
򐂰 Integer
򐂰 String
򐂰 Time

62 Introducing IBM FileNet Business Process Manager

4.2.2 System functions
IBM FileNet Business Process Manager provides a set of functions to use in any
expression. Some example functions include data type conversion functions,
system interrogation functions, general functions, and time functions.

Data type conversion functions

BPM supports data type conversion. Conversion must be explicit in string
variables. The data conversion functions include:
򐂰 convert(source_exp, type_name)
Converts an expression from one supported data type to another.
򐂰 numbertostring(num_exp{, num_mask}opt)
Converts an integer or float to a string, using a specified mask.
򐂰 timetostring(time_exp{, time_mask}opt)
Converts a time to a string, using a specified mask.

System interrogation functions

System interrogation functions include the following function:
򐂰 systemtime()
Returns current server time.

General functions
General system functions include the following functions:
򐂰 if(bool_expr, expr2, expr3)
Conditional expression returns expr2 if true or expr3 if false.
򐂰 sizeof(array_id)
Returns the size of the array.

Time functions
Time functions include the following function:
򐂰 adddays()
Adds a specified number of days. Functions are also available for minutes,
hours, and so forth.

Queue operations
An operation is a function within a step that performs a specific task associated
with a particular queue. When an operation and its parameters are being defined,

Chapter 4. Process design with BPM 63

 the process designer, system administrator, and application developer typically
work together to determine its names and usage.

Operations can be created in work queues, component queues, and user

queues. For information on work queues, component queues, and user queues,
refer to “Queues” on page 51.

4.3 Inter-related processes

The business processes in an organization can be very complex. There are
occasions when one business process relies on another business process to
complete its work. The other process might even start this business process. The
concept of inter-related processes involves the notion of one business process
launching another to carry out additional processing.

When designing business processes, understand which other business

processes are already available that you can incorporate and re-use. In addition,
you can decompose your business process into multiple logically separate
processes that can be shared and used by other processes.

Referring back to the claim approval example, there are areas where we can
leverage inter-related processes. Consider the following steps:
򐂰 Step A: An agent documents a loss due to an auto accident by a policy
holder, and submits a claim packet to the back office.
򐂰 Step B: The mail room clerk logs and adds a timestamp on the claim packet
in the back office.

If the claimant’s car is no longer operational due to the accident and the
claimant’s policy covers the cost of a rental car while the claim is being
processed and the car is being fixed, then we can add a few new steps in
between the above two steps to get the claimant a rental car immediately.

For the purpose of explanation only, using this example, we can add steps for a
rental car reservation. The resulting business process looks like this:
򐂰 Step A: An agent documents a loss due to an auto accident by a policy
holder, and submits a claim packet to the back office.
򐂰 Step A2 (new step): The agent performs an initial review on the claim.
򐂰 Step A3 (new step): If the claimant claims that the insured car is no longer in
operation due to the accident and the policy allows for a rental car in this type
of situation, then the agent calls another entity within the company to set up a
rental car for the claimant, gets the rental car confirmation number back from
the entity, and passes that confirmation information to the claimant.

64 Introducing IBM FileNet Business Process Manager

 򐂰 Step B: The mail room clerk logs and adds a timestamp on the claim packet
in the back office.

In Step A3, another group is involved in making a car reservation for the claimant
or any individual within the company. This is an independent business process
performed by the group. If it already exists for the company, then the claim
approval process can call the car rental process to perform the task. The agent,
involved in the claim approval process, does not have to understand the actual
process of reserving the rental car. All it requires is for the agent to pass the
necessary claimant information and retrieve the confirmation number later.

If, at the design time of the claim approval process, the car reservation process is
not created within the system, you can create this separate process and call it in
the claim approval process.

4.4 Interaction with external systems

IBM FileNet Business Process Manager provides an extensive collection of tools
for building and customizing solutions and integrating with external systems and
processes. There are several options for business processes to interact with
external systems: Component Integrator, Web services, and EAI connectors.

For the rental car reservation process that is mentioned in 4.3, “Inter-related
processes” on page 64, the reservation process can be a process performed by
the rental car company that is transparent to the agent. If such an automated
process exists in the rental car company, then the claim approval process can
interact directly with the external process to make the appropriate rental car
reservation. All the process has to do is to have the agent to approve the rental
car, after which it requests the external reservation process to reserve the car
and retrieves the reservation number automatically.

Which technology to use (either Component Integrator, Web services, or the EAI
connector) depends on how the external system works and what is available for
the claim approval process to interact with the external system.

For more information regarding direct interaction with the external systems, refer
to Chapter 7, “Integration with external systems and services with BPM” on
page 257.

Chapter 4. Process design with BPM 65

66 Introducing IBM FileNet Business Process Manager
 5

Chapter 5. Using Process Designer

In this chapter we show you how to use the Process Designer tool to create a
business process definition.

We cover the following topics:

򐂰 Process Designer user interface overview
򐂰 Creating workflow definition overview
򐂰 Setting up workflow properties
򐂰 Creating steps in a workflow map
򐂰 Creating routes with Process Designer
򐂰 Advanced topics:
– Workflow inheritance
– Default system submaps
– Expression Builder

For process design concepts, refer to the previous chapter. For creating a case
study business process using Process Designer with step-by-step instructions,
refer to the next chapter.

Note: The online ecm_help provides much detailed information on using

Process Designer to create workflow processes. For your convenience and
the completeness of this book, we extract some content from the online help in
this chapter. For more detailed information regarding using Process Designer,
refer to the online help.

© Copyright IBM Corp. 2008. All rights reserved. 67

5.1 Process Designer user interface overview
Process Designer is a graphical tool that you use to create and modify business
process definitions. Process Designer is one of several tools provided with IBM
FileNet Business Process Manager.

Using Process Designer, you can create the specifications for each step in a
business process, including which participant processes the work for the step;
what interface application (also known as a step processor) the participant uses
to perform a task; what attachments, if any, are required for the process; what
data is necessary to view, add, or edit; what responses the participant can
choose; and what routes to take within a business process. The electronic
representation of the business process designed in Process Designer is called a
workflow definition (also known as a process definition).

Note: In IBM FileNet Business Process Manager, a business process is more

than just a simple workflow. However, for convenience when writing for this
book and as well as consistency with the wording used in the Process
Designer application, we use workflow interchangeably with business process
from this point on.

Process Designer is launched from the IBM FileNet Workplace XT application by

selecting Tools → Advanced Tools → Process Designer. See Figure 5-1.

Figure 5-1 Launching Process Designer from Workplace XT

68 Introducing IBM FileNet Business Process Manager

Alternatively, you can launch it from IBM FileNet Workplace by selecting
Author → Advanced Tools → Process Designer from the navigation bar on
the left hand pane. See Figure 5-2.

Figure 5-2 Launching Process Designer from Workplace

Note: After launching the Process Designer application, do not close the IBM
FileNet Workplace Internet Explorer® window. Closing the Internet Explorer
window will result in the closing of the Process Designer application.

Figure 5-3 shows the Process Designer user interface. The window consists of
the Map pane, Properties pane, and Step Palette pane. There is also a set of
toolbar icons on top of the Map pane.

Chapter 5. Using Process Designer 69

 Toolbar icons Map pane (workflow map area) Properties pane

Step Palette pane

Figure 5-3 Process Designer user interface

Note: This book is written based on Version 4.0 of IBM FileNet Business
Process Manager. For release 4.5 or later, some of the user interface,
including menus and toolbar, might be changed.

5.1.1 Map pane

A business process consists of one or more workflow maps. You create the
maps in the Map pane of Process Designer. The Map pane is displayed on the
top left of the Process Designer window. See Figure 5-3. In this pane, you define
steps and routes. There is always one main map and optionally multiple
submaps.

70 Introducing IBM FileNet Business Process Manager

5.1.2 Properties pane
The Properties pane is displayed on the top right of the Process Designer
window. The Properties pane displays the properties for the selected step or
route in the Map pane. When you create an object in the map (a step or route),
you specify the properties for that object in the Properties pane. The configurable
properties vary depending on the type of step or route that is selected. For
example, if the selected object is a General step, you can specify a step
destination, required parameters, milestones, deadlines, and other options that
are specific to the General step. If the selected object is a route, you can specify
participant responses, field values, and routing conditions.

Figure 5-4 shows the Properties pane for the LaunchStep. For this step, you can
provide the user with instructions for how to complete the step in the General tab,
select parameters to read and write in the Parameters tab, modify parameter
values in the Assignment tab, and specify routing information in the Routing tab.

Figure 5-4 Properties pane

We discuss the various settings in the Properties pane for different steps and
routing in later sections.

5.1.3 Step Palettes pane

The Step Palettes pane at the bottom left of the Process Designer interface
contains collections of predefined steps that represent different types of activities
in a workflow.

Chapter 5. Using Process Designer 71

 Step Palettes include the following palettes:
򐂰 BPM Palette
򐂰 CheckPoint Palette
򐂰 General System Palette
򐂰 Timer Palette
򐂰 Web Services Palette
򐂰 My Palette

Figure 5-5 shows the Step Palette drop-down menu in Process Designer.

Figure 5-5 Step Palettes in Process Designer

BPM Palette
The BPM Palette (Figure 5-6) contains common types of steps you use for a
workflow.

The BPM Palette comes with the following types of steps:

򐂰 General step
򐂰 Submap step
򐂰 System step
򐂰 Component step

72 Introducing IBM FileNet Business Process Manager

Figure 5-6 BPM Palette

General steps are used to handle basic workflow activities (tasks) that are
performed by an individual participant or a group of participants. System steps
are used to call built-in system functions. Component steps are used to call
operations from an external application.

To simplify a workflow map, you can break a workflow into one main map and
several smaller submaps. To call the submap from a main map, you insert a
Submap step and specify which submap to call at the step. A submap can call
other submaps.

General System Palette

The General System Palette (Figure 5-7) contains the steps that call system
functions.

The General System Palette comes with the following types of steps:
򐂰 Assign
򐂰 Create
򐂰 DbExecute
򐂰 Delay
򐂰 Log
򐂰 Return
򐂰 TerminateBranch
򐂰 TerminateProcess
򐂰 WaitForCondition

Figure 5-7 General System Palette

All of these system functions can be called directly in the system step available
from the BPM Palette. The difference is visual. By using the system step from the
BPM Palette, you do not know what system function the step calls in the map
unless you look into the step properties (or you can provide a very descriptive
name for the step that shows on the map). If you use the step from the General
System Palette, you can tell what the step does by its name.

Chapter 5. Using Process Designer 73

 The advantage of using the system step in the BPM Palette is that it enables you
to call multiple system functions, one after another within the same step. Thus,
one system step can perform, for example, DbExecute(), Create(), and then
TerminateBranch(), and they all appear as one single step in the workflow map.
This simplifies the visual appearance of the entire workflow map.

Timer Palette
The Timer Palette (Figure 5-8) contains steps that allow timer functions to
interact with steps in the workflow definition. The timer functions are used to
specify processing time limitations for steps in the workflow. This is different than
the deadline function we discussed earlier in Chapter 1, “Overview” on page 1.
A deadline applies to a specific step. A timer usually applies to a series of steps.

The Timer Palette includes the following types of steps:

򐂰 BeginTimer
򐂰 SuspendTimer
򐂰 ResumeTimer
򐂰 EndTimer
򐂰 EndAllTimers

Figure 5-8 Timer Palette

CheckPoint Palette
The CheckPoint Palette (Figure 5-9) contains steps that allow the checkpoint
system functions to interact with steps in the workflow definition. The checkpoint
functions provide mechanisms that allow data field values of work items to be
rolled back to an earlier state. Checkpoint functions also make it possible to
resume processing for work items at a previous point in the workflow.

The CheckPoint Palette includes the following types of steps:

򐂰 BeginCheckpoint
򐂰 RollbackCheckpoint
򐂰 EndCheckpoint

Figure 5-9 CheckPoint Palette

74 Introducing IBM FileNet Business Process Manager

 Web Services Palette
The Web Services Palette (Figure 5-10) contains the steps that allow the
workflow definition to interact with Web services system functions. The steps
specify where in the workflow a Web service can be invoked or where in the
workflow a step can receive a reply from a Web service.

The Web Services Palette includes the following types of steps:

򐂰 Invoke
򐂰 Receive
򐂰 Reply

Figure 5-10 Web Services Palette

My Palette
This palette allows you to create your own collection of step types you use most
often in designing your workflow. For example, if you create a custom step that is
reusable in another workflow, you can add the custom step to My Palette (drag it
from the workflow map in the Map pane) and then use it later for a different map.
You can treat My Palette as a temporary holding place for commonly used or
reusable steps.

We cover what you can set for each step in later sections. For more detailed
information about what each step does, refer to ecm_help.

5.1.4 Toolbar icons

Below the menu in the Process Designer, there is a list of icons (Figure 5-11).
They are shortcuts to access various functions of Process Designer. Here, we list
and group them in the order they appear in Figure 5-11:
򐂰 New Workflow
򐂰 Open
򐂰 Save
--------------------------
򐂰 FileNet Open/Checkout
򐂰 FileNet Add New
򐂰 FileNet Checkin
򐂰 FileNet Cancel Checkout
򐂰 FileNet Save
--------------------------

Chapter 5. Using Process Designer 75

 򐂰 Workflow Properties
򐂰 Workflow Collections
--------------------------
򐂰 Cut
򐂰 Copy
򐂰 Paste
--------------------------
򐂰 Print
--------------------------
򐂰 Validate Workflow
򐂰 Launch Workflow
--------------------------

Figure 5-11 Toolbar icons

The Workflow Properties, Validate Workflow, and Launch Workflow icons are
commonly used while you create your workflow. We use these icons as we
create our case study for the book.

5.2 Creating workflow definition overview

Using Process Designer, you create a workflow definition (also known as a
process definition). As mentioned earlier, a workflow definition describes all
activities required to complete the business process.

A workflow definition contains workflow properties and workflow maps. Workflow

properties contain properties global to the entire workflow. Sample workflow
properties include all data fields that can be used in a workflow, all possible
attachment types that a workflow can have, and all participants that can process
work items in the workflow. Workflow maps include steps and routes.

For each step, you specify the actual participants that will process work items at
this step, as well as the data fields available at this step for reading and writing by
systems or participants. There are other configurations that you can set up in
each step, depending on the type of step. Figure 5-12 illustrates the elements
that comprise a workflow definition.

76 Introducing IBM FileNet Business Process Manager

 Workflow Definition (also known as Process Definition)

Fields (also known as workflow properties)

Map

Step
Route
Step Step
Route
Route Step

Figure 5-12 Workflow definition (also known as process definition)

To create a workflow definition, you must set up workflow properties and create
the appropriate workflow maps.

5.3 Setting up workflow properties

Each workflow has a set of workflow properties. It is different than the properties
shown in the Properties pane of the Process Designer window because the
workflow properties are global to the entire workflow, whereas the Properties
pane only shows the properties of the selected object on the workflow map.

To set or view the workflow properties, click the Workflow Properties icon from
the toolbar in the Process Designer, or select View → Workflow Properties
from the menu. See Figure 5-13.

Figure 5-13 Viewing and editing Workflow Properties

Chapter 5. Using Process Designer 77

 The following tabs are shown on the Workflow Properties dialog box:
򐂰 General tab
򐂰 Advanced tab (roster, event log, condition identifier, and e-mail notification)
򐂰 Data Fields tab
򐂰 Attachments tab
򐂰 Workflow Groups tab
򐂰 Maps tab
򐂰 Milestones tab
򐂰 Web Services tab

General tab
Figure 5-14 shows the General tab in the Workflow Properties dialog box.

Figure 5-14 Workflow Properties - General tab

The General tab allows you to specify general information for the workflow,
including base workflow, name, subject, description, and deadline.

From the General tab, set up the following fields:

򐂰 Base Workflow
Select a base workflow that the new workflow definition is based on. You can
base your new workflow definition on an existing workflow (the base
workflow) to inherit the properties and submaps defined in the base workflow.

78 Introducing IBM FileNet Business Process Manager

 To create a workflow definition based on default settings, leave the default
WorkObjectEX as the base workflow. To create a workflow definition based
on some previously defined workflow, click the Modify icon and select an
appropriate base workflow.
Refer to 5.6.1, “Workflow inheritance” on page 117 for more information on
how workflow inheritance works.
򐂰 Workflow Name
Enter a name for the workflow definition. It is displayed in the list of workflows
when the workflow definition is transferred to the Process Engine. The name
of a workflow definition must conform to the following rules:
– The name must begin with an alphabetic character. It can contain letters,
digits, underscores, or spaces.
– The maximum length for the name is 249 characters.
– The name must not contain F_ or two tilde (~~) characters as the first two
characters.
– The name must be unique among the workflow definitions in an isolated
region.
򐂰 Subject
Enter the title that describes the workflow. This is what participants see in
their inbox when the workflow is running and there are work items for them to
process from this workflow.
򐂰 Description
Enter a description for the workflow. This is an optional field.
򐂰 Deadline
Specify a time limit for the workflow to be completed in minutes, hours, days,
or weeks. This time starts when the workflow launches. You can also set a
reminder notification prior to the deadline. This is an optional field.
If the e-mail notification is set, then the workflow tracker receives the reminder
notification. To enable participants to be notified of the workflow deadlines,
you set a deadline in individual steps of a workflow.

Advanced tab
Figure 5-15 shows the Advanced tab in the Workflow Properties dialog box.

Chapter 5. Using Process Designer 79

 Figure 5-15 Workflow Properties - Advanced tab

The Advanced tab allows you to specify advanced features for the workflow,
including roster, event log, condition identifier, and e-mail notification.

Set up the following fields in the tab:

򐂰 Roster
Select a roster for the workflow. There is always a default roster selected.
Roster is a database table that stores the current location of the workflow
(which step it is at, and which working queue the work item is in) and other
information about workflows. Workflow rosters provide the Process Engine
with an efficient way to locate a specific workflow.
򐂰 Event log
Select an event log. There is always a default event log selected.
An event log is a database table that contains information about system-level
events related to work item processing.
򐂰 Condition identifier
Enter an expression that identifies a specific work item at runtime. This value
is used by the WaitForCondition system function to identify a specific
waited-for work item.
When a work item has to wait for another item to complete or meet a certain
condition, you can use the WaitForCondition system function. For example,
you have a main workflow map and multiple submaps. In the main workflow
map, it requires the completion of a submap to continue the main map’s
process. Setting a condition identifier enables the submap to communicate
with the main map about its completion.

80 Introducing IBM FileNet Business Process Manager

򐂰 Enable Email Notification
Enable or disable e-mail notification for the current workflow definition.

Data Fields tab

Figure 5-16 shows the Data Fields tab in the Workflow Properties dialog box.

Figure 5-16 Workflow Properties - Data Fields tab

The Data Fields tab allows you to define all data fields that can be used in
workflow steps. When you define each step, you specify which data fields are
parameters for that step, and what access permission the participants have to
modify the values. Enter the following information for each data field:
򐂰 Name
Enter the name for the data field.
򐂰 Type
Select one of the data types for the field: Boolean, Float, Integer, String,
Time. If the field has multiple values, then select the array version of the type
indicated by brackets [].
򐂰 Merge Type
Specify how values will be merged when necessary.
A step in a workflow can have multiple outgoing routes if multiple tasks
(following this step) can be performed in parallel. When they merge back to
one step, this field specifies how values from multiple sources of the same
data type are to be merged into one. The default for Merge Type is Override.
򐂰 Expression
Specify an initial value for the field. String fields can be blank, but all other
types require an initial value. The value must match the data type and meet
the requirements for expressions. From our example in Figure 5-16, we use
systemtime() for several date fields.

Chapter 5. Using Process Designer 81

 򐂰 Description
Provide a description of the data field. This is an optional field.

Attachments tab
Figure 5-17 shows the Attachments tab in the Workflow Properties dialog box.

Figure 5-17 Workflow Properties - Attachments tab

The Attachment tab allows you to specify attachments that participants use in
workflow steps. Attachments include documents and URLs. For each
attachment, specify the following information:
򐂰 Name
Enter the name for the new attachment.
򐂰 Array
Specify whether there are multiple attachments. If the Array checkbox is
selected, it represents multiple attachments. For our example, as shown in
Figure 5-17, we select the Array checkbox for SupportingDocs. This is
because we expect the workflow to have multiple supporting documents.
򐂰 Value
Specify a value for the attachment. If it is a file or folder in an object store or
library, you can double-click the field and select one from a list. Alternatively,
you can leave it blank. This allows the value to be assigned at the workflow
launch time or by a participant during workflow processing.
򐂰 Description
Provide a description for the attachment. This is an optional field.

82 Introducing IBM FileNet Business Process Manager

Workflow Groups tab
Figure 5-18 shows the Workflow Groups tab in the Workflow Properties dialog
box.

Figure 5-18 Workflow Properties - Workflow Groups tab

The Workflow Groups tab allows you to specify an individual user or a group of
users who can be assigned as participants in the workflow. For each workflow
group, specify the following information:
򐂰 Name
Enter a name for the workflow group.
򐂰 Participants
Select users for the workflow group. Click Modify to select users.
If a step is processed by a workflow group, then the participant or the group
must be defined in the workflow group here before the step can use it in its
setup.
򐂰 Descriptions
Provide a description for the workflow group. This is an optional field.

Maps tab
Figure 5-19 shows the Maps tab in the Workflow Properties dialog box.

Chapter 5. Using Process Designer 83

 Figure 5-19 Workflow Properties - Maps tab

The Maps tab allows you to specify the main workflow map and the submaps that
you use in your workflow. You can create more flexible and modular workflow
definitions by creating submaps. In addition, you can redefine the behavior of
maps that are inherited from the base workflow. For example, by default, you get
Terminate and Malfunction submaps. Using submaps and the ability to redefine
inherited maps enable you to create custom actions and processing for your
workflow process.

For each map, define the following information:

򐂰 Name
Enter a name for the main workflow map or submap. Indicate whether it is the
main workflow map that launches the workflow. The main map that drives the
workflow process is denoted with a special pencil next to the down-arrow
icon.
򐂰 Description
Provide a description of the map. This is an optional field.

If there is a down-arrow icon next to the map, it means that the map is inherited
from another map. If you create a new map from scratch, there will not be an icon
next to the map.

Figure 5-20 shows the Maps tab with a newly created submap called Claim
Verification. Clicking the Map Usage icon on the upper right corner reveals the
map that calls the Claim Verification submap. See Figure 5-21.

84 Introducing IBM FileNet Business Process Manager

Figure 5-20 Claim Verification submap

Figure 5-21 Caller map for the Claim Verification submap

Milestones tab
Figure 5-22 shows the Milestones tab in the Workflow Properties dialog box.

Figure 5-22 Workflow Properties - Milestones tab

Chapter 5. Using Process Designer 85

 The Milestones tab allows you to define milestones in a workflow. Milestones
help track the progress of a workflow process. When the workflow reaches a
milestone, a milestone message is written to a log file. Depending on the
milestone message level, the message is displayed for workflow participants,
trackers, and users who launched the workflow. The message can be text or it
can include the value of one or more data fields. The maximum display length for
a milestone is 250 characters. For each milestone, define the following
information:
򐂰 Name
Enter a name for the milestone.
򐂰 Message level
Specify a message level for the milestone from 1 to 99.
򐂰 Message
The milestone message that will be recorded in the log when this milestone is
reached. The maximum runtime display length is 250 characters. Characters
beyond 250 are truncated.

Web Services tab

Figure 5-23 shows the Web Services tab in the Workflow Properties dialog box.

Figure 5-23 Workflow Properties - Web Services tab

The Web Services tab allows you to specify Web Services to be used in the
workflow. Set the following information for the Web Services:
򐂰 General tab
Enter general settings for invoking or providing Web services.

86 Introducing IBM FileNet Business Process Manager

 – Incoming Web Services Attachment Folder
If the workflow uses Web services operations that include incoming
attachments, specify a folder in the object store where these files will be
stored.
– Verify messages at runtime
When an XML message is received either as incoming data for a Receive
system instruction or as a reply message from a Reply system instruction,
specify whether to use the appropriate schema to validate the message at
runtime as it is received.
– Finalize existing Web services operations
Select this option to prevent changes in the input and output parameters of
existing operations in Receive and Reply functions in this workflow
definition.
򐂰 Partner Links tab
Specify the Web services partner links that are used in this workflow. Partner
links consist of a name and the WSDL URL for Invoke functions, or a name
and port definition for Receive functions. See Figure 5-24. These are required
before you can create Invoke, Receive, or Reply system functions.

Figure 5-24 Workflow Properties - Web Services -> Partner Links tab

򐂰 XML Schema tab

Specify an XML schema for complex data in a Receive or Reply function. See
Figure 5-25.

Chapter 5. Using Process Designer 87

 Figure 5-25 Workflow Properties - Web Services -> XML Schema tab

򐂰 XML Data Fields tab

Used for Web services operations that use XML mode instead of parameter
mode. It can also be used to specify XML fields for uses such as messages.
See Figure 5-26.

Figure 5-26 Workflow Properties - Web Services -> XML Data Fields tab

We discuss Web services and how to set one up in later chapters of the book.

5.4 Creating steps in a workflow map

A workflow definition consists of workflow properties and workflow maps. A
workflow map consists of two or more steps. To create a workflow map, you
define the steps that are involved in a workflow process and the routes that
specify the sequence of the steps.

88 Introducing IBM FileNet Business Process Manager

Each step in a workflow represents a unique task or activity to be completed by a
participant, a group of participants, or a system. Process Designer contains
many predefined steps (or to be more precise, step templates) to perform
different types of activities. These steps are contained in the Step Palettes and
can be added to workflow maps as needed. Recall that the BPM Palette contains
the most common steps used in the workflow:
򐂰 General step
򐂰 Submap step
򐂰 System step
򐂰 Component step

There are two other steps that can be found on a workflow map:
򐂰 Launch step (the default first step in a main workflow map)
򐂰 Start step (the default first step in any submap)

Figure 5-27 illustrates what some of the steps look like in a workflow map. Notice
that for each type of step, there is a different icon symbol representation.

Launch step Submap step Component step

Main workflow map

General steps

Figure 5-27 Main workflow map with different types of steps

Chapter 5. Using Process Designer 89

 Every workflow map has a default launch step. This is the first step when the
workflow launches. The submap step, shown in Figure 5-27, with the down arrow
icon symbol (it looks like it is drilling down to the next level) calls a submap to
perform activities. The component step with the electric plug icon symbol (it looks
like it is plugging into something) calls external systems or services to perform
activities. The general steps with the people icon symbols (they look like they are
something to be processed by humans) require participants to perform activities
at the step. There are other icon symbols for the general steps, depending on
who or what processes the activities at the step. We discuss configuring these
steps in more detail later in this section.

Adding a step
To add any step to a workflow map:
1. Use the Step Palette drop-down menu and select the appropriate palette.
Usually, this is the BPM Palette.
2. Select the step icon from the palette and drag it to the Map pane.

The user can also add a step to the map using the map's context menu.

Configuring a step
To configure a step that you added to a workflow map:
1. Select the step icon in the workflow map.
2. Edit its properties in the Properties pane.

Depending on the type of the step, you can configure different settings for it.
We cover that in more detail later in this section.

Deleting a step
To delete a step from a workflow map:
1. Right-click the step icon in the workflow map.
2. Select Delete from the context menu.

You cannot delete the default steps (launch step and start step) that are always
included in a main workflow map and its submaps.

90 Introducing IBM FileNet Business Process Manager

5.4.1 General step
A general step is the most frequently used type of step in a workflow map. At
each general step, you assign work (a task) for a participant to complete. When
the workflow moves to this step (you can also think of when the work items are
moved to this step) and a participant is assigned the task, the task shows up in
the participant’s inbox.

You can assign a task to a work queue in the general step. Anyone who has
access to the work queue can perform the task. Assigning work to a work queue
offers more flexibility than assigning a task to an individual. How you assign tasks
depends on your business practice.

Many of the configuration settings for the general step apply to other types of
steps. Use the setting discussed here as a reference for configuring similar
properties for other types of steps.

To configure a step, provide a name in the Name field of the Properties pane. We
recommend providing a meaningful, descriptive name. The name appears in the
workflow map below each step icon and it provides information about what the
step does.

Other properties of a general step can be configured through the following tabs:
򐂰 General tab
򐂰 Deadline tab
򐂰 Parameters tab
򐂰 Assignments tab
򐂰 Routing tab

If you configure your workflow to use a rules engine, there is also a Rules tab. It
is beyond the scope of this book to cover rules engines. We do not discuss the
configuration of rules engines in this book.

General tab
Figure 5-28 shows the General tab in the Properties pane for a general step.

Chapter 5. Using Process Designer 91

 Figure 5-28 Properties pane - General tab

The General tab allows you to configure general information about the step, such
as the participant’s instructions for the step, the step destination, and the step
processor.

From the General tab, set up the following fields:

򐂰 Instruction
Enter the instruction that displays on the user interface when the participant
processes a task at this step. For our example in Figure 5-28, we have an
Adjustor Review step. When the adjustor processes work at this step, we
provide the following instruction as shown in Figure 5-28:
Please Review the attached claim and related documents and Approve
or Reject the claim.
򐂰 Step Destination
Select whether the step is assigned to a participant or a work queue.
As mentioned earlier, depending on your business practice, you can assign
work to a participant or to a work queue where anyone who has access to the
queue can process the work.

92 Introducing IBM FileNet Business Process Manager

 To assign work to a work queue, select Work Queue, and select an existing
queue from the drop-down list. If an automated process has access to the
queue, the automated process can also process work at this step.
To assign work to participant, select Participants. To add the participants,
click the Modify icon (denoted as a pencil symbol) on the upper right corner
of the Participants entry box. The Participant Selection dialog box appears.
You can select workflow groups, users, or groups for the participant. If you do
not see the workflow groups that you want in the selection list, make sure to
add them in the workflow properties first.
Figure 5-29 shows three icons. When the step destination is assigned to a
work queue, participant, or unassigned, it has a different icon representation
for ease of reading the workflow map.

Step destination: Step destination: Step destination:

work queue participants not assigned

Figure 5-29 Icons representing general steps

For our example in Figure 5-28, we assign work to a participant, specifically a

workflow group called Adjustor.
򐂰 Ignore Missing Participants
If Participant is selected for the step destination, specify whether to ignore
missing participants if they are missing from a workflow group. If this flag is
not set and there is no valid participant who can process the work, the
workflow will stop at this step and produce an error. For our example in
Figure 5-28, we want the work to stop if there is a missing participant;
therefore, we do not select the check box.
򐂰 Participant Privileges
If Participant is selected for the step destination, specify whether the
participant can reassign work to someone else, view the status of the
workflow (through the Tracker tool), or view the workflow history. For our
example in Figure 5-28, the adjustor has all the privileges.

Chapter 5. Using Process Designer 93

 򐂰 Step Processor
Step processor is the user interface that a participant uses when processing a
task at a step. Select from a list of the step processors that can be used for
the participant to process the work.

Deadline tab
Figure 5-30 shows the Deadline tab in the Properties pane for a general step.

Figure 5-30 Properties pane - Deadline tab

For time management, you can set a deadline for when the task should be
completed at this step. Reminders can also be set to notify participants of an
approaching deadline. The deadline set in the workflow properties applies to the
entire workflow process. The deadline set here applies to this step only.

94 Introducing IBM FileNet Business Process Manager

As mentioned in the previous chapter, you cannot force a participant to finish the
task before the deadline. However, setting a deadline helps to provide some time
management. If the participant does not finish the task on time, an icon indicating
the expired deadline will display next to the task that the participant has to
complete in the inbox. Also, you can set up escalation points here to notify a
supervisor or another control group to manage the work.

From the Deadline tab, set up the following fields:

򐂰 Complete within
Specify when the task should be completed at this step. Use the first option to
specify a fixed number of time period for the deadline. Use the second option
to enter an expression that determines when the deadline should be.
For our example, in the Adjustor Review step, we set a deadline of 1 week.
This means that the adjustor has one week to complete the task from the time
the task is assigned to the adjustor at this step.
򐂰 Send Reminder
Specify whether there should be a reminder sent to the participant within a
certain period before the deadline.
For our example in Figure 5-30, we configure a reminder to be sent to the
adjustor 2 days before the task’s deadline.
򐂰 Deadline Submap
If the deadline expires, you can activate a deadline submap to handle the
situation.
The deadline submap can be any submap you design to handle the situation
when the deadline expires.

Setting up a deadline for a step is optional. You can choose not to set deadlines,
or selectively set up different fields. For example, you can set up a Complete
within field, without setting up the Send Reminder and Deadline Submap fields.

Chapter 5. Using Process Designer 95

 Parameters tab
Figure 5-31 shows the Parameters tab in the Properties pane for a general step.

Figure 5-31 Properties pane - Parameters tab

You can specify the operations and parameters that can be used for read and
write at this step. Parameters include data fields, attachments, workflow groups,
and XML fields. You can use the parameter values to help you perform the task
at this step or update their values as part of the task. There is a list of available
parameters for you to choose from. If the one you want does not appear in the
Available Parameters list, go back to the workflow properties and add it there.
Also, the checking of any of the four icons above the list determines what
parameters appear on the available list.

From the Parameters tab, set up the following fields:

򐂰 Operation
If the work queue assigned as the destination step has operations defined,
you can select the operation to be executed here.

96 Introducing IBM FileNet Business Process Manager

򐂰 Select Parameters
Select the parameters that you want to be used at this step. For our example,
some of the parameters are AccidentDescription, Agency, AgencyPhone,
AgentName, CaseFolder, and ClaimID. This information is important for the
adjustor to use at this step.
򐂰 Access Rights
Indicate whether the participants have read, write, or both read and write
permission for the selected parameters. For our example in Figure 5-31, the
adjustors can read the CaseFolder and ClaimID, but they cannot update these
fields. They can however, read and write (update) many other fields.
򐂰 Prompt
Enter a descriptive prompt for participants when they work with a parameter.
For example, when the adjustor is working with the AccidentDescription field,
the prompt for the adjustor is:
Description of the accident

Assignments tab
Figure 5-31 shows the Assignments tab in the Properties pane for a general step.

Figure 5-32 Properties pane - Assignments tab

Chapter 5. Using Process Designer 97

 In the Assignments tab, you can set the milestone information, assign values to
fields using expressions, and set up attributes before the execution of the step
(in the Before Execution subtab) or after the step is completed (in the After
Completion subtab).

From the Assignments tab, set up the following fields:

򐂰 Milestone
Assign a milestone before or after the step is completed.
򐂰 Field Assignments
Assign a value to a field. You can enter an expression as the value for the
field.
򐂰 Attributes
Specify name, type, and value for the attributes. Attributes exist only in the
step for which they are set. They are like the temporary parameters specific
for the step.

Setting up assignments for a step is optional. You can choose not to set them or
selectively set up different fields. For example, you can set up Milestone without
setting up Field Assignments and Attributes information.

Routing tab
Figure 5-31 shows the Routing tab in the Properties pane for a general step.

Figure 5-33 Properties pane - Routing tab

98 Introducing IBM FileNet Business Process Manager

In the Routing tab, you configure routing-related settings such as routing
responses, as well as incoming and outgoing routing settings.

From the Routing tab, set up the following fields:

򐂰 Responses
Specify a list of possible responses participants can select at this step. The
response determines the routing from the current step to the next step. For
our example in Figure 5-33, we show the Adjustor Review step’s Routing tab.
At this step, the adjustor can either deny the claim, approve the claim, or
escalate to a supervisor for review. We therefore enter three responses for
the step: Approve, Escalate, and Reject. You can rearrange the order of the
responses by using the upper and lower arrows on the upper right corner of
the Responses entry.
򐂰 Incoming Routing Information
If you used parallel routing earlier, at some point, you must have a collector
step to combine the parallel routes into one place. If the current step is the
collector step, check the Collector Step option. Otherwise, leave it
unchecked.
򐂰 Outgoing Routing Information
If there are multiple routes leaving from this step to other steps, you specify
how the route is determined in this field. You can select one of the following
options:
– All true conditions: All specified conditions must be true before the route
can be taken.
– First true conditions: As soon as one condition becomes true, the route is
taken.

There are three options listed under the Outgoing Routing information:
򐂰 Approve
򐂰 Escalate
򐂰 Reject

These show the possible outgoing routes from this step. They are configured
when you set up the outgoing routes for the step (which we discuss later in the
route configuration section).

Notice that, although you cannot change the outgoing routes in this step, you can
specify the order in which the route conditions are evaluated. Select any one of
the outgoing routes and click the up and down arrows next to it to rearrange its
order in the list. Depending on how you set up the condition, sometimes the order
of the condition might become important.

Chapter 5. Using Process Designer 99

 Clicking Details on the bottom of the Routing tab shows the routing condition to
the next step. Figure 5-34 shows the outgoing routing conditions from the
Adjustor Review step. The conditions are set when you configure individual
outgoing route from the current step.

Figure 5-34 Properties pane - Routing conditions

Figure 5-35 shows the main workflow map with the Adjustor Review step.

Figure 5-35 Claim approval process main workflow map

As long as any participant selects the Approve response from the Adjustor
Review step, the next step goes to the Generate Letter step. If any participant
selects the Escalate response, then the next step goes to the Supervisor Review
step. If the response is Reject, then the next step goes to the Claim Rejected
step.

If multiple tasks can be performed in parallel after this step, you can select the
option All true conditions.

100 Introducing IBM FileNet Business Process Manager

5.4.2 Submap step
A submap step calls a submap to perform activities. Submaps can help organize
complex workflows by breaking them into more manageable pieces. In addition, if
part of a process is executed multiple times (with multiple steps), you can create
a submap to encapsulate that part of the repeated process and have the submap
steps calling the submap whenever needed in a workflow map. A main workflow
map can have multiple submaps. A submap can also have other submaps.

Figure 5-35 on page 100 shows our example, the main workflow map. In the
map, there is a submap step called Claim Setup. This submap step calls the
Claim Setup submap as shown in Figure 5-36. The submap encapsulates all the
tasks of setting up the claim, allowing the main workflow map to hide the details.

Figure 5-36 Claim Setup submap

You can configure a submap step through the following tabs:

򐂰 General tab
򐂰 Attributes tab
򐂰 Routing tab

If you configure your workflow to use a rules engine, there is also a Rules tab. It
is beyond the scope of this book to cover rules engines. We do not discuss the
configuration of rules engines in this book.

Chapter 5. Using Process Designer 101

 General tab
Figure 5-37 shows the General tab in the Properties pane for a submap step.

Figure 5-37 Properties pane - General tab for submap step

The General tab allows you to configure general information about the step, such
as the submap to call and the description.

From the General tab, set up the following fields:

򐂰 Map
Select the submap to be executed when workflow reaches this step. For our
example in Figure 5-37, we show the properties of the Claim Setup submap
step. This submap step calls Claim Setup Subprocess submap. You create
the submap in similar ways as you create a main workflow map, with steps
and routes.
򐂰 Description
A description of the submap.

102 Introducing IBM FileNet Business Process Manager

Attributes tab
Figure 5-38 shows the Attributes tab in the Properties pane for a submap step.

Figure 5-38 Properties pane - Attributes tab for submap step

You can set up attributes to be used in the submap step for special processing.
Attributes are valid only within this step. Unlike parameters, which can be used
throughout the entire workflow, attributes are created and used only at the
particular step.

From the Attributes tab, for each attribute that you want to use, set up the
following information:
򐂰 Name
Enter a name for this attribute.
򐂰 Type
Select the attribute type, including boolean, float, integer, and string. Attribute
type also supports arrays.
򐂰 Value
Enter a value for the attribute. Note that you cannot use Expression Builder
for an attribute value.

Setting up attributes for a submap step is optional. For our example, we do not
set up any.

Chapter 5. Using Process Designer 103

 Routing tab
Figure 5-39 shows the Routing tab in the Properties pane for a submap step.

Figure 5-39 Properties pane - Routing tab for submap step

In the Routing tab, you configure how the step handles the incoming routes and
how it processes the outgoing routes.

From the Routing tab, set up the following fields:

򐂰 Incoming Routing Information
If you used parallel routing earlier, at some point, you must have a collector
step to combine the parallel routes into one place. If the current step is the
collector step, check the Collector Step option. Otherwise, leave it
unchecked.
򐂰 Outgoing Routing Information
If there are multiple routes leaving from this step to other steps, you specify
how the route is determined in this field. You can select one of the following
options:
– All true conditions: All specified conditions must be true before the route
can be taken.
– First true conditions: As soon as one condition becomes true, the route is
taken.
For our example as shown in Figure 5-39, we select the second option,
First true conditions.

104 Introducing IBM FileNet Business Process Manager

 There are two options listed under the Outgoing Routing information:
򐂰 > 3K
򐂰 <= 3K

Clicking Details on the bottom of the Routing tab shows the routing condition
that is set from this step to the next step. Figure 5-40 shows the outgoing routing
conditions from the Claim Setup step. If the VehicleEstimationAmount is greater
than 3000, then it goes to the Adjustor Review step. If the estimated amount is
less than or equal to 3000, then it goes directly to the Generate Letter step.

Figure 5-40 Properties pane - Routing condition for submap step

5.4.3 System step

A system step calls system functions in a workflow definition. System functions
perform specific functions. They can be grouped as follows:
򐂰 General system functions: Provide general system functions. See details
below.
򐂰 Checkpoint system functions: Enable rolling back of the workflow process
to a specific point in time.
򐂰 Timer: Provide timer constraint on parts of a workflow process that must be
completed within a certain time frame. This includes setting a beginning timer,
ending time, and the ability to suspend and resume the timer.
򐂰 Web services system functions: Provide functions to interact with Web
services, including calling and receiving the Web services, and replying back
to the caller.

Chapter 5. Using Process Designer 105

 General system functions include the following functions:
򐂰 Assign: Assign values to data fields.
򐂰 DbExecute: Executes a stored procedure in any supported database.
򐂰 Create: Launches a new workflow.
򐂰 Call: Call a submap.
򐂰 Return: Exit from the current submap and return to the calling map.
򐂰 Delay: Delay the workflow for a certain period of time.
򐂰 Log: Log the event in the event log.
򐂰 TerminateBranch: Terminate the current branch of workflow processing.
򐂰 TerminateProcess: Terminate the entire workflow processing.
򐂰 WaitForCondition: Suspend existing processing until a certain condition
becomes true.

Checkpoint system functions include the following functions:

򐂰 Checkpoint - BeginCheckpoint
򐂰 Checkpoint - EndCheckpoint
򐂰 Checkpoint - RollbackCheckpoint

Timer system functions include the following functions:

򐂰 Timer - BeginTimer
򐂰 Timer - EndAllTimer
򐂰 Timer - EndTimer
򐂰 Timer - ResumeTimer
򐂰 Timer - SuspendTimer

Web services system functions include the following functions:

򐂰 Invoke
򐂰 Receive
򐂰 Reply

There are several tabs available where you can configure a system step:
򐂰 General tab
򐂰 Attributes tab
򐂰 Routing tab

The configuration of the last two tabs is similar to that of a general step and a
submap step. We discuss only the general tab here.

106 Introducing IBM FileNet Business Process Manager

General tab
Figure 5-41 shows the general tab in the Properties pane for a system step.

Figure 5-41 Properties pane - General tab for system step

From the general tab, provide a description for the system step. For each system
function that you want the system step to perform, follow these procedures:
1. Select the system function from the Available Functions list.
2. Click the right arrow icon.
3. After the system function appears in the Selected Functions list, double-click
it.
4. Depending on the type of the system function you added, enter appropriate
configuration information.

The best way to learn how to use various system functions is to try them out in
Process Designer. The settings for the system functions are fairly intuitive. We
do not cover the configuration of each of them in detail. For detailed information,
refer to ecm_help.

Chapter 5. Using Process Designer 107

5.4.4 Component step
Component steps enable a workflow to interact with external applications,
systems, and processes. You can import custom components and use the
functions they provide in your workflow.

Figure 5-42 shows the Claim Setup submap with the component steps.

Component steps

Claim Setup submap

Figure 5-42 Claim Setup submap with component steps

There are several tabs available where you can configure a component step:
򐂰 General tab
򐂰 Attributes tab
򐂰 Routing tab

The configuration of the last two tabs are similar to that of a general step, a
submap step, or a system step. We discuss only the general tab here.

General tab
Figure 5-43 shows the general tab in the Properties pane for a component step.

108 Introducing IBM FileNet Business Process Manager

Figure 5-43 Properties pane - General tab for component step

The general tab allows you to set up the component and the operations that you
use at this step.

From the general tab, set up the following information:

򐂰 Description
Enter a description for the component step.
򐂰 Operations
Add one or more operations to be performed by components at this step.
The order of the operations listed here is the order of operation execution. To
rearrange the order, select the operation and click the up or down arrow icon
on the upper right corner of the Operations entry box. To delete, click the X
icon. To add, click the document icon on the upper right corner.
The default component provided by the system is the CE_Operations. You
can also add your own component operations there.
For our example, we show the component step, Create Case. In this step, the
following operations are executed:
– createFolder(): Create a folder for the claim case.
– file(): File all the claim related files in the claim folder.
– searchForOne(): Searches for claim related files.

Chapter 5. Using Process Designer 109

 Notice that the first and the last operations are available through the
More_Operations component. The file() operation is available through the
CE_Operations component.
To add an operation, follow these procedures:
a. Click the Add (document) icon on the upper right corner of the Operation
entry box. The Operation Selection dialog box appears (see Figure 5-44).
b. Select a component from the Component drop-down list.
c. Select one or more operations in the list and click OK.

Figure 5-44 Operation selection for the component step

򐂰 Operation Parameters
Enter the necessary parameter information (their names, types, and values in
the expression column) required for the operations in this step. Not all
operations require parameters. For our example, as shown in Figure 5-43,
only the file() operation requires two parameters:
– sourceDocument
– destFolder
Both of these parameters are of attachment type. The file() operation files the
claim related documents (sourceDocument) to the claim folder (destFolder).

5.4.5 Launch step

A launch step represents the first step in a business workflow process.
Configuring a launch step is similar to configuring a general step, with a few
differences.

110 Introducing IBM FileNet Business Process Manager

 In the general tab, there is no step destination. In addition, you cannot define a
deadline in a launch step. In the Assignments tab, there is only the After
Completion subtab, but no Before Execution subtab. This is because the launch
step is the first step in a workflow process. In the Routing tab, there is no
Incoming Routing Information setting. Again, this is because the launch step is
the first step and there are no incoming routes for the step.

Refer to 5.4.1, “General step” on page 91 for configuration details.

5.4.6 Start step

The start step is the first step on a submap. It is similar to a general step except
that you cannot set a deadline for the step.

Refer to 5.4.1, “General step” on page 91 for configuration details.

5.5 Creating routes with Process Designer

Routes define how work moves from one step to the next in a business process.

Adding a route
To add a route between two steps in a workflow map, do the following procedure:
1. Move the mouse over the starting step until the route symbol appears.
2. Drag the route from the starting step to the ending step.

Figure 5-45 illustrates these steps.

Step 1: Move the mouse over to

the starting step.

Step 2: Drag the route from the starting

step to the ending step.

Resulting route between the steps

Figure 5-45 Drawing routes between steps

Chapter 5. Using Process Designer 111

 Configuring a route
To configure a route that you added in a workflow map:
1. Select the route in the workflow map.
2. Edit its properties in the Properties pane.

Deleting a route
To delete a route from a workflow map:
1. Right-click the route in the workflow map.
2. Select Delete from the context menu.

5.5.1 Setting conditions for routes

You can set routing conditions to either Always true or Condition.

If there is only one outgoing route from the current step to the next step, that is, if
the workflow process always goes from the current step to the next step, then
select Always true. Figure 5-46 shows routes in the main workflow for our
example. From the Launch step to the Claim Setup step, there is only one route.
Thus, for the example, we select Always true in the Properties pane. Notice that
when Always true is selected, you cannot configure anything else for the route.

If there are multiple routes coming from the current step to other steps, you
select Condition for the route. You can configure the routing condition based on
responses that the participants select in the current step (where this route comes
from) or based on values of the data fields in the workflow process. You can also
use the combination of both responses and data fields to set up the routing
condition.

The data fields used for conditional routes, such as data fields, attachments, and
workflow groups, must be defined in the workflow properties.

With multiple outgoing routes, configure the condition for each route so the
system understands under which condition a specific route is taken.

112 Introducing IBM FileNet Business Process Manager

Figure 5-46 Properties pane - Route configuration: Always true

For a route that is set as Condition and not as Always true, configure it with the
following tabs:
򐂰 Responses tab
򐂰 Data fields tab

Response tab
When a participant completes a task at a step, you can set up a list of responses
that a participant can select. Each response corresponds to an outgoing route
from the step.

To configure the route based on the responses, configure the following fields in
the Response tab:
򐂰 Condition
Select one of the condition options:
– ALL: Return true if all participants select a particular response.
– NONE: Return true if none of the participants selects a particular
response.

Chapter 5. Using Process Designer 113

 – ANY: Return true if any participant selects a particular response.
– COUNT: Return true if the total number of participants selecting a
particular response is either equal to, not equal to, greater than, greater
than or equal to, less than, or less than or equal to a specific value.
– Operator: If COUNT is selected, then it specifies the operator used to
compare the count with a particular value. Operator options include equal,
not equal, great than, greater than or equal, less than, and less than or
equal to that value.
– Value: If COUNT is selected, then it specifies the value that corresponds
to the number of participants who selected a particular response.
򐂰 Response
Select the particular response that works with the specified condition above.

After setting the condition, click the Insert button to add the expression on the
condition field. You can manually edit the condition expression also. After you
insert the condition for a route on the bottom entry box, it does not matter what
values are selected or shown in the Condition, Response, and other fields
mentioned previously. Always make sure that the condition appears at the
bottom before continuing. See Figure 5-47.

Condition for the route

Figure 5-47 Properties pane - Conditional Routing for the Approve route

114 Introducing IBM FileNet Business Process Manager

The following list shows conditional routing examples based on our Claim
Approval main workflow map:
򐂰 The Approve route that connects the Adjustor Review step to the Generate
Letter step has the following condition:
ANY(Approve)
This means that as long as one participant selects the Approve response at
the Adjustor Review step, the Approve route’s condition returns true, and the
next step in the workflow goes to the Generate Letter step.
򐂰 The Escalate route that connects the Adjustor Review step to the Supervisor
Review step has the following condition:
ANY(Escalate)
This means that as long as one participant selects the Escalate response at
the Adjustor Review step, the Escalate route’s condition returns true and the
next step in the workflow goes to the Supervisor Review step.
򐂰 Using the same logic, the Reject route has the following condition:
ANY(Reject)
If any one participant selects the Reject response, then the next step goes to
the Claim Rejected step.
򐂰 Let us assume that we have the following condition for a route:
COUNT(Approve) >= 2
In this case, as soon as two or more participants select the Approve
response, the condition returns true and the route is taken.

Remember that you must define a list of available responses at the step from
which the routes originate. At each route leading from the step, you set the
condition for when each route should be taken.

Note: If there is only one participant processing the task at the step, then the
condition ANY and ALL would be the same.

Data Fields tab

Instead of using responses selected by participants as the condition for a route,
you can also use the values in the data fields to create a condition for evaluation.

Chapter 5. Using Process Designer 115

 Figure 5-48 shows the condition set for a route.

Figure 5-48 Properties pane - Conditional Routing Data Fields for a route

To set up the condition, use the following fields to construct an expression for the
condition and then click Insert:
򐂰 Field: Select a data field to be used in the condition expression.
򐂰 Operator: Select the operator to evaluate the condition.
򐂰 Value: Select a value to evaluate the condition.

Remember that, after you select and set the field values, click Insert to add the
condition for the route. You can manually edit the condition using AND, OR, and
parentheses to create complex expressions.

For the example shown in Figure 5-48, we have the following condition:
EstimateVehicleAmount>3000

This means that the route is taken if the estimated repair cost for the vehicle is
less than 3000.

116 Introducing IBM FileNet Business Process Manager

 Combining both responses and data fields
You can set the route condition to be based on both participant responses and
data field values. To accomplish this, manually edit the condition expression. The
following expression is an example of a condition based on both a data value and
the participant's response:
EstimateVehicleAmount>3000 and ANY(Approve)

Note: Set up condition routing carefully. If you have multiple outgoing routes
from a step, yet none of the conditions in the routes returns true, the workflow
will stop and produce an error.

5.6 Advanced topics

In this section, we discuss topics related to creating workflow definitions using
Process Designer that are beyond the basic concepts we have covered so far.
Specifically, we discuss workflow inheritance and how it works, default system
submaps and how to override them, and using the Expression Builder to build
various expressions for route conditioning or parameter assignment.

5.6.1 Workflow inheritance

When workflow inheritance is enabled, you can choose an existing workflow
definition as the base workflow from which a new workflow definition inherits.
This allows you to leverage existing definitions at a higher level and pass them
along to new workflow definitions.

A new workflow can inherit the following items from a base workflow:
򐂰 Main workflow map
򐂰 Submaps
򐂰 Data field, attachment, and workflow group definitions
򐂰 Workflow deadline and reminder
򐂰 Milestones
򐂰 Event log
򐂰 Roster
򐂰 Condition Identifier
򐂰 Partner link and XML schema
򐂰 XML data field
򐂰 Incoming Web services attachment folder
򐂰 Rule set names
򐂰 E-mail notification preference

Chapter 5. Using Process Designer 117

 To override the inherited items, you can redefine them in the new workflow
definition.

Let us look at some examples. Figure 5-49 shows two workflows, Workflow A
and Workflow M, and their local and inherited workflow properties.

Figure 5-49 Workflow inheritance flow hierarchy for Workflow A and M

Workflow A inherits from the default workflow definition, WorkObjectEx.

Workflow A contains the following items:
򐂰 Main workflow map A
򐂰 Submaps a1 and a2
򐂰 Data fields a1 and a2

Workflow M inherits some of the items from Workflow A and overrides others.
Workflow M contains the following items:
򐂰 Main workflow map M (which overrides the one from Workflow A)
򐂰 Submap a1* (which overrides the one from Workflow A)
򐂰 Submap a2 (which is inherited from Workflow A)
򐂰 Submap m1 (which is new for Workflow M)
򐂰 Data fields a1 and a2 (which is inherited from Workflow A)
򐂰 Data field m1 (which is new for Workflow M)

Now adding a new Workflow R, we have the workflow inheritance flow hierarchy
for these three workflows as shown in Figure 5-50.

Figure 5-50 Workflow inheritance flow hierarchy for Workflow A, M, and R

118 Introducing IBM FileNet Business Process Manager

Workflow R inherits from Workflow M. It contains the following items:
򐂰 Main workflow map M (which is inherited from Workflow M)
򐂰 Submaps a1* and m1 (which are inherited from Workflow M)
򐂰 Submap a2 (which is inherited from Workflow A)
򐂰 Submap rm1 (which is new for Workflow R)
򐂰 Data fields a1 and a2 (which are inherited from Workflow A)
򐂰 Data field m1* (which is inherited from Workflow M)
򐂰 Data field r1 (which is new for Workflow R)

If inheritance is disabled, then Workflow R contains only the following items (see
Figure 5-51):
򐂰 Main workflow map R (which is new for Workflow R)
򐂰 Submap rm1 (which is new for Workflow R)
򐂰 Data field r1 (which is new for Workflow R)

All other inherited main maps, submaps, and data fields are no longer available.

Figure 5-51 Disable workflow inheritance and its affect on Workflow R

Now, let us look at an example of enabled inheritance, but switching base

workflow. Again, we have Workflow A, M, and R as discussed earlier. We
introduce a new Workflow N that inherits from Workflow A with a new submap
and a new data field. Now, if we switch the base workflow for Workflow R from
Workflow M to Workflow N — in other words, Workflow R now inherits from
Workflow N — let us see what happens. Figure 5-52 shows the new resulting
Workflow R.

Chapter 5. Using Process Designer 119

 Figure 5-52 Switching base workflow from Workflow M to Workflow N

From Figure 5-52, Workflow N, which inherits from Workflow A, contains the
following items:
򐂰 Main workflow map N (which is new for Workflow N)
򐂰 Submaps a1 and a2 (which are inherited from workflow A)
򐂰 Submap n1 (which is new for Workflow N)
򐂰 Data fields a1 and a2 (which are inherited from Workflow A)
򐂰 Data field n1 (which is new for Workflow N)

When inheritance is enabled and we switch the base workflow for Workflow R
from Workflow M to Workflow N, Workflow R now contains the following items:
򐂰 Main workflow map N (which is inherited from Workflow N)
򐂰 Submaps a1 and a2 (which are inherited from Workflow A through N)
򐂰 Submap n1 (which is inherited from Workflow N)
򐂰 Data fields a1 and a2 (which are inherited from Workflow A through N)
򐂰 Data field n1 (which is inherited from Workflow N)
򐂰 Submap rm1 (which is new for Workflow R)
򐂰 Data field r1 (which is new for Workflow R)

By comparing the resulting main workflow map, submaps, and data fields of
Workflow R now with the one earlier (which inherits from Workflow M), you can
see how inheritance works.

120 Introducing IBM FileNet Business Process Manager

5.6.2 Default system submaps
When defining a simple, stand alone workflow process, it might be sufficient to
put all the steps on the main workflow map. However, if the workflow definition
becomes more complex, it might be more practical to break up the workflow into
smaller workflows (submaps) that can then be called by the main workflow map.
Submaps make the workflow definition more modular. When using submaps, the
main workflow map acts as a container for the submaps.

IBM FileNet Business Process Manager comes with a set of default system
submaps that you can use for your workflow: Malfunction and Terminate maps.
You can customize them for your business requirements.

Malfunction map
The malfunction map executes whenever an error occurs during workflow
processing. When an error occurs, the malfunction map moves the work item to a
system queue to be reviewed by an administrator. After the work item is reviewed
and the error condition is resolved, the work item can be returned to either the
calling step where the error condition occurred or it can be routed to the next step
in the business process.

Figure 5-53 shows the Malfunction map in Process Designer.

Figure 5-53 Malfunction map in Process Designer

Chapter 5. Using Process Designer 121

 In most cases, there is no need to work with the malfunction map. The system
executes it only when required. By default, the Malfunction map is read-only, but
you can override it to meet your business requirements. Use caution when
changing the default behavior of the Malfunction map. Test thoroughly to avoid
unexpected results during workflow processing.

Terminate map
When a process is complete, the work item is routed to the Terminate map.
There are no steps associated with the Terminate map and there is no need to
explicitly call it from the main workflow. Each workflow map automatically calls
the Terminate map at the end of the workflow.

Figure 5-54 shows the Terminate map in Process Designer.

Figure 5-54 Terminate map in Process Designer

In most cases, there is no need to work with the Terminate map. The system
executes it automatically. By default, the Terminate map is read-only, but you
can override it to embed common terminating steps for your workflow process.
Use caution when changing the default behavior of the Terminate map. Test
thoroughly to avoid unexpected results during workflow processing.

5.6.3 Expression Builder

The Expression Builder dialog box is used to build expressions that can be
evaluated by the workflow. The expressions are built using components that
exist in the workflow definition such as data field names, attachments, workflow
groups, XML data fields, system fields and partner links. In addition to workflow
components, system functions can also be used in expressions. The Expression
Builder dialog box can be launched in any field that permits expressions.

122 Introducing IBM FileNet Business Process Manager

To build an expression, specify the following items in the Expression Builder:
򐂰 Choose a category from the category drop-down menu. This determines
which items are available to be added to the expression. Categories include
Data Fields, Attachments, Workflow Groups, XML Data Fields, System
Fields, Partner Links, and Functions. See Figure 5-55 for the Expression
Builder category drop-down menu.

Figure 5-55 Expression Builder Category drop-down menu

򐂰 Choose the desired category items from the pick list to be added to the
expression.
Figure 5-56 shows the data fields available for the workflow.

Figure 5-56 Expression Builder: Data Fields

Chapter 5. Using Process Designer 123

 Figure 5-57 shows the system fields that can be used in an expression for the
workflow.

Figure 5-57 Expression Builder: System Fields

Figure 5-58 shows the functions that are available to be used in an expression for
the workflow.

Figure 5-58 Expression Builder: Functions

򐂰 Choose appropriate operators (such as +, -, *, /, is equal, not equal, is greater

than, is greater than or equal), conjunctions (AND, OR), parentheses to be
added to the expression.
򐂰 Click the Insert button to add the selected function, operator, or other
selected items to the expression area at the bottom of the Expression Builder
dialog.

124 Introducing IBM FileNet Business Process Manager

򐂰 Modify the expression in at the bottom of the dialog as necessary.

Note: Click the Clear button to remove the content from the expression area.

You can use a combination of data fields, system fields, functions, Workflow
Groups, and other categories to build the expression. If you manually edit the
expression, make sure the syntax is correct.

To type literals into the expression area, enter values in the appropriate format:
򐂰 Strings must be enclosed in quotes.
򐂰 Array values must be separated by commas (,) and surrounded by braces {}.

For more details, refer to ecm_help.

Chapter 5. Using Process Designer 125

126 Introducing IBM FileNet Business Process Manager
 6

Chapter 6. Implementing business

processes: Case study
In this chapter we provide step-by-step instructions for implementing business
processes with IBM FileNet Business Process Manager. We use the case study,
a claim approval application of a fictitious insurance company that is introduced
earlier in the book, to demonstrate how to implement the business processes.

We cover the following topics:

򐂰 Building business processes overview
򐂰 Defining the basic process object model
򐂰 Starting a new workflow process
򐂰 Building the process workflow map
򐂰 Finishing the step properties definition
򐂰 Building a submap
򐂰 Creating a workflow subscription
򐂰 Testing the basic process
򐂰 Creating and calling a Web service

Note: This book is written based on Version 4.0 of IBM FileNet Business
Process Manager. For release 4.5 or later, some of the user interface,
including menus and toolbar, might be changed.

© Copyright IBM Corp. 2008. All rights reserved. 127

6.1 Building business processes overview
There are many ways to build a business process using IBM FileNet Business
Process Manager. We recommend following these steps:
1. Define the basic process object model.
2. Start a new workflow process:
a. Define the workflow properties.
b. Save the workflow process.
Always perform this step whenever you make any major changes to
ensure that your work is properly saved.
3. Build the main workflow process map:
a. Add steps.
b. Add routes.
c. Define conditional routes if any.
d. Validate the workflow process map.
Always perform this step whenever you make any major changes to
ensure that there is no error in the workflow process map.
4. Define the individual steps in the main workflow process map.
– Define Web services steps, if any.
– Define Component Integrator steps, if any.
5. Create and define any submap used in the main workflow process.
6. Create subscriptions to launch the workflow process.
This is an optional step. You can invoke your workflow process in multiple
ways. We show you one way of creating a workflow subscription in a later
section of the chapter.

Not all the steps have to be created in the sequence we describe here. For
example, it is common to create a simple workflow map, with all the steps and
routes first drawn in Process Designer, before going into more detail of defining
each step. This is a top-down approach where you define all the elements (steps
and routes) on the map graphically and then define each element in detail.
Another approach is to define each element in detail as you build the entire map.
In addition, you can decide when to build the submap and when to build the Web
services or Component Integrator steps. Based on your preferences, you decide
how you want to build your workflow.

In the following sections, we describe the detailed steps required to build

business processes.

128 Introducing IBM FileNet Business Process Manager

6.2 Defining the basic process object model
The claims approval process integrates content (claims and supporting
documents) with a business process. Content is permanently stored as objects in
the IBM FileNet Content Engine. The workflow process runs in the Process
Engine and has its own temporary storage in the Process Engine while the
process is active. To permanently store and update information, the workflow
process can use and update the data persistently stored in the Content Engine.

To store content in Content Engine, you must first define the object model using
the IBM FileNet Enterprise Manager. It is beyond the scope of the book to show
you how to define the object model step-by-step. However, for your convenience,
we provide the XML data files that you can import into the Content Engine so that
you can follow the steps we show here for our case study.

For our case study, we have to define the following components:

򐂰 Document object model for supporting documents
򐂰 Document entry template for supporting documents

Document object model for supporting documents

The claim approval process requires a document object to be created in the IBM
FileNet P8 Content Engine. Create the document object from the IBM FileNet
Enterprise Manager tool using the configuration outlined in Table 6-1 through
Table 6-5:
򐂰 Table 6-1 and Table 6-2 show the Supporting Claim Document class
definition, including the class properties information.
򐂰 Table 6-3 and Table 6-4 show the choice list for different supported document
types.
򐂰 Table 6-5 shows the storage policy information.

Table 6-1 Supporting Claim Document class object

Name Symbolic Name Description

Supporting Claim SupportingClaimDocument Supporting Claim Document

Document

Table 6-2 Supporting Claim Documents class properties

Name Symbolic name Description Choice list Data type

ClaimID N

Last Name N

DocType Y

Chapter 6. Implementing business processes: Case study 129

 Table 6-3 Choice list
Name Data Type

DocType String

Table 6-4 Choice List elements

Name Value

Accident Photo Accident Photo

Medical Report Medical Report

Notice of Payment Notice of Payment

Police Report Police Report

Misc Misc

Witness Report Witness Report

Table 6-5 Storage Policy

Storage Policy objectstore

Document entry template for supporting documents

The claims approval process requires the creation of a Document Entry
Template for supporting documents. Using the Advanced Tools in the IBM
FileNet Workplace, create a Document Entry Template with the following
parameters. For more information on creating Document Entry Templates, refer
to the IBM FileNet Workplace Administration Guide.

Define an entry template. Set the folder, properties and security as shown in
Table 6-6.

Table 6-6 Document Entry Template Folder properties

Object Store P8Demo

Folder Incoming

Hide Select Folder step selected

Constrain to folder selected or it sub-folders unchecked

130 Introducing IBM FileNet Business Process Manager

 Modify the Supporting Claim Document Class with the properties shown in
Table 6-7.

Table 6-7 Supporting Claim Document Class

Class Supporting Claim Document

Show Set Properties step selected

Yes, auto classify the content on entry and use unchecked

base Document Class by default

Yes, use entry template for check-in checked

Set the Document Entry Template properties as shown in Table 6-8. Set the
Entry Template security settings as shown in Table 6-9.

Table 6-8 Document Entry Template properties

Required Property Default Value Access Level

unchecked Document Title {FILENAME} Hide

unchecked ClaimID Editable

unchecked Last Name Editable

unchecked DocType Editable

Add as major Yes Hide

version:

Table 6-9 Document Entry Template Security

Title Owner Promote Modify Modify View View Publish Remove
Control Version Content Props Content Props

#AUTHENTICA
TED-USERS

#CREATOR-O checked checked checked checked checked checked checked unchecked

WNER

Administrator

Chapter 6. Implementing business processes: Case study 131

 Set the Security step display option to: Hide Set Security step. See Table 6-10.

Table 6-10 Set security step

Hide Set Security step selected

Under option 6, select Folder and Document Title. See Table 6-11.

Table 6-11 Setting a folder for the Document Entry Template

Setting under option 6 Value

Select Folder ***BPM Lab → Template

Document Title Claim Supporting Document Entry

Template

6.3 Starting a new workflow process

To create a workflow process, use the Process Designer application. Follow
these steps to launch Process Designer and start a new workflow process:
1. Launch Internet Explorer and enter the following URL:
http://<host name>:<port>/Workplace
Where:
– <host name> is the name of the machine where Application Engine is
installed.
– <port> is the port number where the Workplace application resides.
For our case study, we enter:
http://queens:9080/Workplace
The Workplace logon window appears.
2. Log in using an appropriate user ID and password.

132 Introducing IBM FileNet Business Process Manager

3. From the left-hand pane, select Author → Advanced Tools → Process
Designer as shown in Figure 6-1.

Figure 6-1 Launching Process Designer

Chapter 6. Implementing business processes: Case study 133

 This launches Process Designer and automatically starts a new workflow
process. See Figure 6-2.

Figure 6-2 Process Designer: start a new workflow process

6.3.1 Defining workflow properties

Workflow properties contain global variables and property values that are
applicable to the entire workflow process. You can define them in the beginning
of the workflow creation process. Alternatively, you can modify or add
information later.

To define workflow properties, follow these steps:

1. From Process Designer, select View → Workflow Properties. See
Figure 6-3.

Tip: Roll your mouse over the tool bar below the main menu to locate the
shortcuts to frequently used functions, such as the Workflow Properties
icon.

134 Introducing IBM FileNet Business Process Manager

Figure 6-3 Workflow Properties: Define

The Workflow Properties dialog box opens as shown in Figure 6-4.

Figure 6-4 Workflow Properties: General tab

2. From the Workflow Properties dialog box, on the General tab:

– Enter the Workflow Name, Subject, and Description.
For our case study, we enter XYZ Basic Process for the Workflow Name.
We leave the other fields blank.
– Enter Deadline information. You can specify within how long (in minutes,
for example) the workflow must complete and within what time frame (in
minutes, for example) before the deadline comes to send a reminder.
For our case study, we do not use the Deadline feature. We leave these
fields blank.

Chapter 6. Implementing business processes: Case study 135

 3. From the Workflow Properties dialog box, click the Advanced tab. The panel
shown in Figure 6-5 opens. In this tab:
– Make selections for Roster, Event Log, and Condition Identifier.
For our case study, we keep the default values as-is.
– Select to enable or disable e-mail notification.
Enable this feature if you want to notify users via e-mails. Whether you
use it or not depends on your business requirements.
For our case study, we do not use e-mail notification. Clear the box.

Figure 6-5 Workflow Properties: Advanced tab

4. From the Workflow Properties dialog box, click the Data Fields tab. This is
where you specify all the data fields (parameters) that are used for the
workflow process. In this tab:
– Add the data fields used for your workflow process. For each data field,
double-click the Name cell to enter its value and press Enter. Use the
drop-down box under the Type to select the data type. Also, enter
information for Merge Type, Expression, and Description. If you decide
later that you need additional fields, you can come back to this tab and
add more fields.
For our case study, we enter the data fields as shown in Figure 6-6.

Tip: Before adding a number of data fields of the same type, change
the data field type to the one you use the most. Each subsequent data
field entry automatically uses this type. For our case study, we use
String frequently, so we change the Type to String, and then add the
rest of the data fields.

There is always one extra line in the end. Ignore that line.

136 Introducing IBM FileNet Business Process Manager

Figure 6-6 Workflow Properties: Data Fields tab

5. From the Workflow Properties dialog box, click the Attachments tab. This is
where you specify all the attachments that can be added during the workflow
process. Attachments can include folders, IBM FileNet P8 documents and
objects, and external documents that are a part of the work item for the
workflow process. In this tab:
– Add attachments used for the workflow process. For each attachment,
enter Name, Value, and Description.
For our case study, we enter attachment information as shown in
Figure 6-7.
– If there are multiple attachments of the type specified, select the Array
check box. If there is only one attachment of the specified type, leave the
check box blank.
For our case study, we might have multiple SupportingDocs. We select
the Array check box for that attachment type. For other types of the
attachments, such as ClaimForm and PolicyDoc, we leave that check box
blank.

Chapter 6. Implementing business processes: Case study 137

 – If a particular attachment initiates the workflow process, select the
attachment entry and click the blue Initiating Attachment icon on the
toolbar. This results an arrow appearing next to the entry.
For our case study, the creation of a ClaimForm initiates the workflow. We
therefore set it up as the Initiation Attachment. See Figure 6-7.

Figure 6-7 Workflow Properties: Attachments tab

– To enter a value for the attachment, double-click the value cell and select
an item from an existing object store.
For our case study, the workflow process uses a search template to
search for claims during one of its tasks. We need to add this search
template as an attachment for the workflow. We call this attachment
ClaimSearchTemplate and it uses a search template that is already
created in the system as its value. The existing search template, called
Search for Claim ID, searches from the system and returns the claims with
a given Claim ID. To specify the existing search template (Search for
Claim ID) as the value for the attachment (ClaimSearchTemplate), use the
following steps:
i. Select the attachment, for our case study, ClaimSearchTemplate.
ii. Double-click the corresponding Value entry. This displays a Browse for
item window as shown in Figure 6-8.

138 Introducing IBM FileNet Business Process Manager

Figure 6-8 Browse for attachment item in the object store

iii. Navigate to the location where the search template (Search for Claim
ID) is stored. In our scenario, double-click objectstore → Case Study
Searches, and select Search for Claim ID. See Figure 6-9.

Note: Our case study is built in an object store called objectstore.

The existing search template is in the Case Study Searches folder.
You can create your search template in any folder within your object
store and navigate to that location. If you import the object store
data provided by this book, use this folder.

Figure 6-9 Browse attachment template through object store

iv. Click Select.

Note: This attaches a reference to the Search Template process

parameter.

For our case study, we also select values for other attachments as shown in
Table 6-12.

Chapter 6. Implementing business processes: Case study 139

 Table 6-12 Attachments and values
Name Browse path Description

ParentFolder objectstore → Claims This is the root folder where all

claims will be stored.

SearchSupport objectstore → Case Study This is a search template that

ingDocs Searches → SupportingDoc via searches all supporting
LastName documents via a last name.

SearchPolicy objectstore → Case Study This is a search template that

Searches → Search for Policy searches for related policies.

WordTemplate objectstore → Case Study This is a Microsoft Word

Templates → Claims Processed template that is used to create a
Letter Template standard Claims Processed
letter.

6. From the Workflow Properties dialog box, click the Workflow Groups tab.
This is where you specify the workflow groups that are used in a workflow
process. A workflow process has multiple steps. Each step can have one or
more participants performing tasks. The participant can be an individual or a
group of people. If it is a group of people, you define the group of people as
the workflow group in this tab.
Enter the workflow group name by selecting the Name cell and typing in the
workflow group name or use the pencil icon to add or modify the workflow
group:
a. Click the Modify (pencil) icon. See Figure 6-10.

Figure 6-10 Workflow Groups tab

140 Introducing IBM FileNet Business Process Manager

 b. Select Users or Groups. Type the first few characters for the user name
or the group name in the Starts with field and click the Search (magnifying
glass) icon. See Figure 6-11, where we typed a in the Starts with field.

Figure 6-11 Search for a user or group

c. A list of users or groups appears that matches the characters you typed in
the Starts with field. Select the users or workflow groups and click the Add
button. See Figure 6-12.

Figure 6-12 Add a user or a group to the workflow participant list

d. Click OK.
For our case study, we add Adjustor, Field_Agent, and CreatedBy as the
workflow groups as shown in Figure 6-13.

Chapter 6. Implementing business processes: Case study 141

 Figure 6-13 Added participants for sample workflow

7. From the Workflow Properties dialog box, there are other tabs including the
Maps, Milestones, and Web services tabs. In these tabs, you can specify
other workflow maps that you use, any milestones you want to establish, and
any Web services that you want to use within the workflow process.
For simplicity of the first basic workflow, we do not define anything in these
tabs at this point in time for our case study.
8. When everything is defined in the Workflow Properties dialog box, click
Close.

6.3.2 Saving the workflow process

Now is a good time to save what you have created and defined so far. There are
three methods of saving a workflow process:
򐂰 Option 1: Save the workflow process definition locally on your workstation or
a share drive using File → Save As.
򐂰 Option 2: Check in the workflow process definition into Content Engine and
check out again for update using File → IBM FileNet Add New.
򐂰 Option 3: Hybrid of the above two methods using File → IBM FileNet Save.

Option 1: File -> Save As

The first option, saving your workflow process locally, allows you to develop and
test the process to completion before checking it into Content Engine. To save
the workflow process locally, from Process Designer, select File → Save As or
File → Save.

142 Introducing IBM FileNet Business Process Manager

 Note: The functions of Save and Save As are the same as what you would
expect in any Windows® environment.

Option 2: IBM FileNet Add New

The second option is to check the workflow process into Content Engine and
create a new version when the workflow process is checked in. This allows
Content Engine to keep track of the changes with a version number. To check in
the workflow process for the first time, follow these steps:
1. From Process designer, select File → IBM FileNet Add New. See
Figure 6-14.

Note: Use IBM FileNet Add New for the first time check-in only. After the
initial check-in, use IBM FileNet Open/Checkout, IBM FileNet Checkin,
and IBM FileNet Cancel Checkout functions to manage the versioning of
the workflow process.

Figure 6-14 Saving workflow process: IBM FileNet Add New

2. The Save the workflow definition to an object store dialog box displays as
shown in Figure 6-15. Click Browse and navigate to a folder where you want
to store the workflow process. To navigate, select a folder and click Open or
double-click the folder.

Chapter 6. Implementing business processes: Case study 143

 Figure 6-15 Saving workflow process: Object store browse window

For our case study, we browse to the Case Study Process folder as shown
in Figure 6-16 and click Select.

Figure 6-16 Saving workflow process: Select folder

3. After the folder is selected (see Figure 6-17), click Next.

Figure 6-17 Saving workflow process: Next

4. The dialog box prompts you to enter a document title for the workflow
process. See Figure 6-18. Enter a descriptive name and click Finish.

144 Introducing IBM FileNet Business Process Manager

Figure 6-18 Saving workflow process: Finish

Option 3: File → IBM FileNet Save

A third option is somewhat of a hybrid of the first two. Using File → IBM FileNet
Save, you save the changes to Content Engine without checking in the process.
When you are ready to check in as a version, you can then use option 2 to check
in the workflow process.

Tip: To test your workflow process, you can click the Launch icon on the
toolbar (the icon on the far right).

To keep Process Designer from adding or checking in a process (as a new

version) whenever you launch the workflow process, change your preference
as follows:
1. From Process Designer, select Tools → Preferences. See Figure 6-19.
2. Click the Workflow tab.
3. Clear the Add/Check In Workflow Before Transfer/Launch option.

Notice that from the preference dialog shown in the figure, you can also set
whether you want the system to automatically validate your workflow before
you transfer or launch the workflow.

Figure 6-19 Process Designer: Preference setting

Chapter 6. Implementing business processes: Case study 145

6.4 Building the process workflow map
A workflow process consists of multiple steps and routes. In this section, we
show you how to build a process workflow map.

6.4.1 Adding steps

At the bottom left of the Process Designer window, the step icons are arranged in
palettes based on their functions. Figure 6-20 shows some of the icons.

Figure 6-20 New map view

The default palette is the BPM Palette. To see steps that are available for a
different palette, click the drop-down arrow next to the palette name and select a
different palette. Provided palettes include CheckPoint Palette, General System
Palette, Timer Palette, and Web services Palette. You can also create your own
palette for convenience of workflow design.

146 Introducing IBM FileNet Business Process Manager

To add steps to a map, follow these procedures:
1. Add a new step to a map using one of the following methods:
– Select from the pop-up menu.
Place your cursor anywhere on the map. Right-click to display the pop-up
menu. Select one of the available steps from the pop-up menu. See
Figure 6-21.
– Use drag-and-drop.
From the palette on the bottom of the window, drag and drop the step to
the map above.

Note: The pop-up menu from the first method displays only a list of steps
that are available from the BPM palette. To add steps other than what are
available from the BPM palette, use the drag-and-drop method.

Figure 6-21 Add step with right mouse click

For our case study, we right-click at the map area and select New Submap
Step from the pop-up menu.

Chapter 6. Implementing business processes: Case study 147

 2. Enter a name for the step in the Properties pane’s Name field located at the
right side of the Process Designer menu. See Figure 6-22.
Each step is given a default name. You can modify this name at a later time.
For our case study, we type Claim Setup.

Figure 6-22 Define step name

3. Add the rest of the steps for your map.

For our case study, we add six steps as listed in Table 6-13. We arrange the
steps as shown in Figure 6-23.

Table 6-13 Steps created for our case study

Step name Step type

Claim Setup Submap

Adjustor Review General

Supervisor Review General

Claim Processed General

Claim Rejected General

Generate Letter Component

148 Introducing IBM FileNet Business Process Manager

 Figure 6-23 Steps added to the main map

Deleting steps
If you want to delete a step from the map, right-click the step in the map and
select Delete from the pop-up menu.

6.4.2 Adding routes

After adding steps to the map, you can draw routes to connect the steps
together. This can be done while you add steps to the map, or you can do it after
you add all the steps to the map.

To draw a route from one step (the from step) to another step (the target step):
1. Place the mouse cursor at the edge of the from step so that the cursor
changes to the one shown in Figure 6-24.
2. Left-click, hold, and drag the cursor to the target step and release the mouse
button.

Chapter 6. Implementing business processes: Case study 149

 Figure 6-24 Add route cursor

3. Enter a name for the route in the Properties pane Name field located at the
right side of the Process Designer. The route name becomes the label of the
route on the map. Naming a route is optional.

Tip: If there are multiple routes leading from one step to other steps, we
recommend providing a descriptive name for each route. This makes the
map easier to read and understand.

4. Add routes between all the steps as designed for your workflow process.

For our case study, we create routes as listed in Table 6-14. Figure 6-24 shows
the setup of our case study at this point.

150 Introducing IBM FileNet Business Process Manager

Table 6-14 Routes created for the case study
From Step Target Step Route name (Label)

LaunchStep Claim Setup

Claim Setup Adjustor Review > 3K

Claim Setup Generate Letter <= 3K

Adjustor Review Generate Letter Approve

Adjustor Review Supervisor Review Escalate

Adjustor Review Claim Rejected Reject

Supervisor review Generate Letter Approve

Supervisor review Claim Rejected Reject

Generate Letter Claim Processed

Figure 6-25 Completed routes

Chapter 6. Implementing business processes: Case study 151

 Notice that if a step has multiple routes, the route is automatically numbered
based on the order in which you draw the route on the map.

Deleting routes
If you want to delete a route from the map, right-click the route and select Delete
from the pop-up menu.

6.4.3 Setting up conditional routing

A step can have multiple routes (outgoing routes) connect from it to other steps.
An example of such a step is the Adjustor Review step from our case study,
where there are three outgoing routes, Approve, Escalate, and Reject.

If a step has multiple outgoing routes, you can specify the conditions with which
the routings happen, as follows:
1. Define the step’s routing responses if user responses are involved.
2. Configure the conditional routing to be based either on user responses or on
data fields.

Defining a step’s routing responses

If a user decides what to do with a work item at a particular step, then the user’s
decision at this step controls where the work item is going to be routed to. This
means that you have to define a list of acceptable user responses at this step.
These responses have a one-to-one correspondence with the outgoing routes
from this step to other steps in the process.

To define the step’s responses used for routing:

1. Select the step.
For our case study, select the Adjustor Review step.
2. In the Properties pane on the right side, click the Routing tab.
3. Double-click the first cell under the Name column in the shaded Responses
section and enter your response. Then press Enter.
For our case study, enter Approve in the Name column.
4. Continue to enter other responses, one for each route that connects from this
step to other step.

Note: Remember, when entering responses in the Name column, to press

Enter after each entry to ensure that the value is registered.

152 Introducing IBM FileNet Business Process Manager

For our case study, we configure the user responses for two steps as shown in
Table 6-15. Figure 6-26 shows the setup of our case study at this point.

Table 6-15 Configure step responses

Step Responses

Adjustor Review Approve

Escalate
Reject

Supervisor Review Approve

Reject

Figure 6-26 Configure step responses

Chapter 6. Implementing business processes: Case study 153

 Configuring conditional routing
If a step has only one outgoing route, then when a person or a system completes
the step, the work continues to the next step automatically. If a step has multiple
outgoing routes, we have to place conditions on the routes to allow the work from
this step to pass onto the correct step.

Routing of work can be based on data field values used in the process or based
on user responses.

Conditional routing based on user responses

As mentioned earlier, the system can route the work item to the appropriate step
based on the user’s response. For example, at the Adjustor Review step in our
case study, if a user decides to accept the claim (which means that the user
selects Accept at the step), the work is routed to the Generate Letter step. If the
user decides to escalate the claim, then the work is routed to the Supervisor
Review step. If the user decides to reject the claim, then the work is routed to the
Claim Rejected step.

To configure conditional routing based on a user’s response:

1. Select the route you want to configure.
For our case study, click the route between the Adjustor Review step and the
Generate Letter step.
2. In the Properties pane, click the Condition radio button.
3. Select the Responses tab and set up the following conditions:
a. From the Condition drop-down list, select one of these conditions:
ALL, NONE, ANY, COUNT.
For our case study, select ANY.
b. From the Response drop-down list, select a response you defined earlier.
For our case study, select Approve.

Note: The drop-down list should provide a list of the responses you
defined earlier for the step where this route comes from. If a response
you expected is missing from the list, probably you did not press Enter
earlier to register the response. Go back to the step and add the
missing response.

c. Click Insert.
If you make a mistake, you can click Clear to clear the routing condition
and restart again. Always remember to click Insert or else the condition is
not registered in the system.

154 Introducing IBM FileNet Business Process Manager

Figure 6-27 shows the setup of our case study at this point.

Figure 6-27 Configure conditional routing based on user responses

For our case study, we set up conditional routings based on user responses as
listed in Table 6-16.

Table 6-16 Configure conditional routing based on user responses

From step Target step Condition Response

Adjustor Review Generate Letter ANY Approve

Adjustor Review Supervisor review ANY Escalate

Adjustor Review Claim Rejected ANY Reject

Supervisor review Generate Letter ANY Approve

Supervisor review Claim Rejected ANY Reject

Chapter 6. Implementing business processes: Case study 155

 Conditional routing based on data field values
In addition to routing a work item based on a user’s responses, you can also set
up the routing based on data field values. For example, at the Claim Setup step
in our case study, if the estimated claim amount (VehicleEstimateAmount) is
more than $3000, the system routes the work to the Adjustor Review step, which
enables an adjustor to manually approve, reject, or escalate the claim to a
different step based on the adjustor’s review. Otherwise, if the estimated claim
amount is less than or equal to $3000, the system automatically routes the claim
to the Generate Letter step.

To configure conditional routing based on data field values:

1. Click the route you want to configure.
For our case study, click the route between the Claim Setup step and the
Adjustor Review step.
2. In the Properties pane, click the Condition radio button.
3. Click the Data Fields tab and set up the following conditions:
a. From the Field drop-down list, select the data field you use to route the
work to other steps.
b. From the Operator drop-down list, select the operator you want to use.
c. Enter a value in the Value box.
d. Click Insert.
For our case study, we want the conditional routing to happen when
VehicleEstimateAmount > 3000. We set up our case study as follows:
– Select VehicleEstimateAmount from the Field drop-down list.
– Select is greater than from the Operator drop-down list.
– Enter 3000 for the Value box.

Note: The Field drop-down list shows a list of the data fields you set up
earlier in the Workflow Properties dialog box. If a data field you require is
not on the list, go back to the Workflow Properties and add the data field
there.

For our case study, we set up conditional routing based on data field values as
listed in Table 6-17. Figure 6-28 shows the setup of our case study at this point.

156 Introducing IBM FileNet Business Process Manager

 Table 6-17 Configure conditional routing based on data field values
From Step Target Step Field Operator Value

Claim Setup Adjustor Review VehicleEstimateMaount Is greater than 3000

Claim Setup Generate Letter VehicleEstimateMaount Is less than or 3000

equal

Figure 6-28 Configure conditional routing based on data field values

6.4.4 Validating the workflow process map

Validating your map allows you to catch any errors in the workflow process map.

Note: We also recommend validating your map after you make any significant
change to the map.

Chapter 6. Implementing business processes: Case study 157

 There are two methods to validate a workflow process map from Process
Designer:
򐂰 Use the menu selection.
򐂰 Use the toolbar icon.

To validate a workflow process map using the menu selection, select File →
Validate as shown in Figure 6-29.

Figure 6-29 Workflow validation via menu

Or, you can use the second method to validate a workflow process map, by
clicking the Validate icon on the toolbar, as shown in Figure 6-30.

Figure 6-30 Workflow validation via toolbar icon

158 Introducing IBM FileNet Business Process Manager

 If you built your workflow process map according to our case study, you should
receive three errors. One error is caused because you have not defined the
submap yet. The other errors are caused because you have not yet added the
participants to the Adjustor Review and Supervisor Review steps. Figure 6-31
shows a sample error dialog box.

Figure 6-31 Validation error

If you receive more than the three errors mentioned, use the information in the
error dialog box to correct the problem before you continue with the rest of the
chapter.

6.5 Finishing the step properties definition

After building the workflow process map with steps and routes, you get an image
of what the process looks like visually. Now, you need to complete the definitions
of each step involved.

Note: You can define the properties of steps at the same time when you build
the workflow process map. From our experience, it is best to create the map
first and go back to define the steps at a separate time. The sequence of
building the complete workflow process depends on your preference.

You can define the properties of each step at the Properties pane on the right
side of the Process Designer window.

Chapter 6. Implementing business processes: Case study 159

 Depending on the type of a step, you define different information. For example:
򐂰 For the General step, there are five tabs where you can enter information:
– Parameters: Specify parameters to be used in the step and the operation
for the step if any.
– Assignments: Specify data field assignments, attributes, and milestone
before and after the step execution.
– Routing: Specify routing responses, and incoming and outgoing routing
information.
– General: Specify the instruction provided to users at the step, the step
destination (either participants or working queue), participant privileges,
and step processor.
– Deadline: Specify the deadline information.
򐂰 For the Submap step, there are three tabs:
– General: Specify the associated submap name and description.
– Attributes: Specify the attributes used for the submap.
– Routing: Specify the incoming and outgoing routing information.

To define each step:

1. Select the step from the workflow process map.
2. From the Properties pane on the right side, go through each tab.
3. Enter or select the necessary information.

For our case study, we set up properties for the following steps:
򐂰 LaunchStep
򐂰 Adjustor Review
򐂰 Supervisor Review
򐂰 Claim Rejected
򐂰 Claim Processed
򐂰 Generate Letter - A component step

160 Introducing IBM FileNet Business Process Manager

Setting up LaunchStep
To set up Launch step in our case study:
1. Select LaunchStep from the workflow process map.
2. From the Properties pane on the right side, select the Parameters tab.
Select the parameters that this workflow process uses from the Available
Parameters list and click Add. The parameters should show up on the
Selected Parameters on the right side. You can select multiple parameters by
holding down the Ctrl button and clicking each parameter or use the Shift
button.
For our case study, we select the parameters for LaunchStep as shown in
Table 6-18. Figure 6-32 shows the setup of our case study visually.

Table 6-18 Parameters for LaunchStep

Available parameters Available parameters Available parameters

AccidentDate DebitAccountNumber LastName

AccidentDescription EffectiveDate LossLocation

Agency Email PolicyNumber

AgencyPhone ExpirationDate ReportedDate

AgentName FirstName VehicleEstimateAmount

ClaimForm Injury VehicleReplacement

ClaimSettlementType InsuredNameAddress VehicleVIN

Company InsuredSSN

Chapter 6. Implementing business processes: Case study 161

 Figure 6-32 LaunchStep parameters

3. Select the Assignments tab. This is where you set the milestone for this step
and assign fields and attributes with values and expressions.
For our case study, after the completion of LaunchStep, we want to set the
field F_Subject to be "New Claim : Claim ID "+ClaimID with quotes. To set
this, we follow this procedure:
a. Under the shaded Field Assignments section, double-click the empty cell
under Name and enter the field name, F_Subject.

162 Introducing IBM FileNet Business Process Manager

 b. Under Expression, click the input box and then double-click the
Expression Builder button on the right of the input box (see Figure 6-33).

Figure 6-33 F_Subject assignment

c. The Expression Builder window opens as shown in Figure 6-34. Enter the
following expression with quotes on the bottom pane and click OK:
"New Claim: Claim ID "+ClaimID

Figure 6-34 Building expression for F_Subject field

Chapter 6. Implementing business processes: Case study 163

 d. Enter the remaining three expressions using the data from Table 6-19.

Table 6-19 LaunchStep - After completion field assignments

Name Expression

F_Subject "New Claim: Claim ID "+ClaimID

LossCity substr(LossLocation, 1, strloc(LossLocation, ",") - 1)

LossState substr(LossLocation, strloc(LossLocation, ",") + 1)

CreatedBy {Creator}

Setting up the Adjustor Review step

To set up the Adjustor Review step in our case study:
1. Click the Adjustor Review step on the main map.
2. From the Properties pane on the right side, click the General tab. In this tab,
you can enter the following information:
– Instruction: Provide instructions in the user interface to participants of the
step to guide them and help them in completing the step.
– Step Destination: Specify at which destination the task in the step can be
performed. Step destination includes Work Queue or participants.
Participants can include users, user groups, or workflow groups defined in
the workflow properties.
– Participant Privileges: Specify the privileges the participants have when
performing this step. The available option includes Reassign, View Status,
and View History.
– Step Processor: Specify the type of the user interface to use when
participants performing this step.

164 Introducing IBM FileNet Business Process Manager

 For our case study:
– Under Instruction, enter the following text as shown in Figure 6-35:
Please review the attached claim and related documents and
approve or reject the claim.

Figure 6-35 Adjustor Review - General tab, Instructions

– This step can be accessed by users under the CreatedBy workflow group.
To set this up, under the Step Destination:
i. Select Participants and click the Modify (pencil) icon. The participant
Selection dialog box opens (see Figure 6-36).
ii. Select Workflow Groups.
iii. Select CreatedBy from the Available Participants, click Add to add it to
the Selected Participants list.
iv. Click OK.

Note: In reality, only participants belong to the Adjustor workflow group

should be able to work on this step. For simplicity of the case study, we
select CreatedBy here.

Chapter 6. Implementing business processes: Case study 165

 Figure 6-36 Adjustor review participants

e. The participants in this step do not have rights to reassign the task, or to
view the workflow status or history. To set this up, under Participant
Privileges, clear Reassign, View Status, and View History.
3. From the Properties pane on the right side, click the Parameter tab. For our
case study, we select the parameters for the Adjustor Review step as listed in
Table 6-20.

Table 6-20 Adjustor review parameters

Available parameters Available parameters Available parameters

AccidentDescription EffectiveDate LossCity

Agency Email PolicyNumber

AgencyPhone ExpirationDate ReportedDate

AgentName FirstName VehicleEstimateAmount

CaseFolder Injury VehicleReplacement

ClaimSettlementType InsuredNameAddress VehicleVIN

Company InsuredSSN

DebitAccountNumber LastName

166 Introducing IBM FileNet Business Process Manager

 Note: When the claim is filed, the loss location information, which includes
the loss city and state, is entered. The system parsed the location
information from the Loss Location field to the Loss City and Loss State
fields. At this step, we only need the Loss City information to process.
Therefore, the parameters required at this step include only Loss City and
not Loss Location or Loss State.

Setting up the Supervisor Review step

For our case study, we set up the Supervisor Review step as follows:
򐂰 The General tab:
– Instruction: Please review the attached claim.
– Participants: Workflow Groups, CreatedBy
– Participant Privileges: Clear all
򐂰 Parameters (see Table 6-21).

Table 6-21 Parameters for the Supervisor Review step

Available parameters Available parameters Available parameters

AgencyPhone EffectiveDate LossLocation

AgentName Email PolicyDoc

CaseFolder ExpirationDate ReportedDate

ClaimID FirstName VehicleEstimatAmount

ClaimSearchTemplate Injury VehicleReplacement

ClaimSettlementType InsuredNameAddress VihicleVIN

Company InsuredSSN

DebitAccountNumber LastName

Chapter 6. Implementing business processes: Case study 167

 Setting up the Claim Rejected step
For our case study, we set up the Claim Rejected step as follows:
򐂰 The General tab:
– Instruction: Claim has been rejected.
– Participants: Workflow Groups, CreatedBy
– Participant Privileges: Clear all
򐂰 Parameters:
Agency ExpirationDate
AgencyPhone LastName
EffectiveDate PolicyNumber

Setting up the Claim Processed step

For our case study, we set up the Claim Processed step as follows:
򐂰 The General tab:
– Instruction: Claim has been processed.
– Participants: Workflow Groups, CreatedBy
– Participant Privileges: Clear all
򐂰 Parameters (see Table 6-22).

Table 6-22 Parameters for the Claim Processed step

Available parameters Available parameters Available parameters

Agency EffectiveDate LastName

AgencyPhone ExpirationDate PolicyNumber

168 Introducing IBM FileNet Business Process Manager

 Setting up the Generate Letter step
The Generate Letter step is a component step. See the next section on how to
set up a component step.

6.5.1 Setting up a component step

A component step calls Java methods that are available for use by the workflow
process to complete a specific task. For our cases study, the Generate Letter
step calls a Java method that copies a word document and another file to the
claim folder of the current claim that is being processed.

To set up a component step:

1. Select the component step.
For our case study, we select the Generate Letter step.
2. In the Properties pane on the right side, in the shaded Operations section,
click Add (see Figure 6-37).

Figure 6-37 Generate Letter - Define a component step

3. From the component drop-down list, select the component to use.

For our case study, we click More_Operations.

Note: More_Operations is a component that has been made available for

our case study. It is an unsupported IBM code at the time of writing.

For complete steps in setting up a Component Integrator before you use it

in your workflow process, refer to 8.2, “Implementing Component
Integrator” on page 274.

Chapter 6. Implementing business processes: Case study 169

 4. From the Operation list, select one or more operations that are associated
with the selected component and click OK.
For our case study, we select copy as shown in Figure 6-38.

Figure 6-38 Copy function from the More_Operations component

5. For each operation, under the shaded Operation Parameters list, enter values
for the parameters that are required for the operation.
For our case study, the following parameters are required for the copy
operation:
– Object: A container for the document to be copied. In the Expression
drop-down list, select WordTemplate as shown in Figure 6-39.

170 Introducing IBM FileNet Business Process Manager

Figure 6-39 Copy function parameter set up: Object

– File: Specifies whether a copy of the document is to be created in the

same folder as the old one. For the case study, set it to false by
double-clicking the Expression box and entering false.
– Copy: The original document to be copied. From the Expression
drop-down list, select ClaimNotice.

Chapter 6. Implementing business processes: Case study 171

 6. Table 6-23 shows the component set up for the Generate Letter step. Using
the table and set up the remaining CE_Operation component and its function.

Table 6-23 Component and function setup for the Generate Letter step
Component name Operation names Parameter names Expression

More_Operations Copy Object WordTemplate

File false

Copy ClaimNotice

CE_Operation file sourceDocument ClaimNotice

destFolder CaseFolder

6.6 Building a submap

In this section, we show you how to build a submap. Submaps help you to
simplify complex workflow processes. You can group related tasks and activities
into logical modules and create corresponding submaps to accomplish these set
of tasks and activities. You can reuse the submaps within your workflow or copy
them to other workflow processes. Submaps can also call other submaps to
further modularize works.

Building a submap is similar to building a normal workflow process. It consists of

the following steps:
1. Create a new submap using Process Designer.
2. Add steps and routes as a normal workflow map.
3. Call this submap from other workflow process.

A submap cannot execute by itself. It must be called by the main workflow map
or other submaps.

For our case study, we create the Claim Setup submap.

6.6.1 Creating the Claim Setup submap

For our case study, the Claim Setup submap handles the following tasks:
1. Create a claim folder based on a system-generated Claim ID.
2. Call the Get Field Agent step to assign a field agent to the claim.

172 Introducing IBM FileNet Business Process Manager

3. Query Content Engine for all documents related to the claimant’s last name.
We assume, to simplify the case study, that all of these documents with the
claimant’s last name are related to the claim. In reality, we should use more
fine-tuned criteria to search for related documents and add to the folder. This
step actually does not add the documents to the folder. At the time of writing,
we decide to search for the documents without adding them into the folder.
4. Wait until the assigned field agent adds one or more supporting documents to
the claim folder. To simplify the case study, we assume that when at least one
supporting document is added to the claim folder, the field agent’s job is
done. In reality, we probably want to make sure the right documents (such as
multiple vehicle repair estimates and the police report) have been received for
the claim (that is added to the claim folder) before continuing to the next step.
5. Call the Get Adjustor step to assign the appropriate adjustor for the claim.
6. Complete its process and returns to the calling workflow map.

Creating new submap

To create the new Claim Setup submap:
1. From Process Designer, click the Create Map icon as shown in Figure 6-40.

Figure 6-40 Create submap: Claim Setup

Chapter 6. Implementing business processes: Case study 173

 2. A Create Map dialog box opens. Select Create New Map.
3. Enter a descriptive name for the Name field of the new map. For our case
study, we enter Claim Setup Submap.
4. Click OK, see Figure 6-41.

Figure 6-41 Create Map dialog box

Adding steps
For our case study, Table 6-24 lists the steps in the Claim Setup submap. Add
these steps as shown in Figure 6-42.

Table 6-24 Claim Setup submap steps

Step name Step type

Create Case Component

Retrieve Related Docs Component

Get Field Agent System

Field Agent General

Get Adjustor System

Check for Attachments Component

File Supporting Docs Component

174 Introducing IBM FileNet Business Process Manager

Figure 6-42 Claim Setup submap steps

Adding routes
Add routes to connect the steps in the Claim Setup submap as shown in
Figure 6-43.

Figure 6-43 Claim Setup submap steps with routings

Chapter 6. Implementing business processes: Case study 175

 Defining conditional routing
In the case study, when a field agent attaches a document to the claim folder in
the Field Agent step, the workflow property AttachmentCheck is set to true. The
numbered routes, [1] from the Check for Attachments step to the Field Agent
step, and [2] from the Check for Attachments step to the File Supporting Docs
step, are conditional routes based on the value of AttachmentCheck.

At the Check Attachments step, we want to see if there is any attachment

associated with the case study. If AttachmentCheck is set to false — that is, the
field agent has not attached any document to the claim folder — then the
workflow process goes back to the Field Agent step and waits for the field agent
to attach related documents to the claim folder. If AttachmentCheck is set to true,
that is there is at least one attachment associated with the claim folder, then the
conditional routing goes to the File Supporting Docs step.

In real life, the conditional routing would be more complicated than what we show
here. As mentioned earlier, depending on your business process requirements,
you can set the Check for Attachments step to check whether a policy report has
been filed and whether the vehicle repair estimate from an approved auto-repair
shop has been submitted.

To set up the conditional routings for Claim Setup submap:

1. Select the route [1] from Check for Attachments to Field Agent.
2. From the Properties pane on the right side, click Condition.
3. Select the Data Fields tab.
4. Select AttachmentCheck (Boolean) from Field drop-down list.
5. Select is equal for Operator.
6. Enter false for Value.
7. Click the Insert button. See Figure 6-44.

176 Introducing IBM FileNet Business Process Manager

Figure 6-44 Conditional routing to route from Check for Attachment to Field Agent

8. For route [2] from Check for Attachments to File Supporting Docs, click
Always true in the route’s Properties pane.

Table 6-25 shows the conditional routing set up in the Claim Setup submap.
Notice that the routing [1] from Check for Attachment to Field Agent is always
checked first. If the condition is not true, the second routing [2] always executes.

Table 6-25 Conditional routing in Claim Setup map

Source step Target step Conditional Data fields
routing type

Check for Field Agent Condition AttachmentCheck

Attachment = false

Check for File Supporting Always true

Attachment Docs

Setting up the Create Case step

In the Claim Setup submap, there is a component step called Create Case.
Table 6-26 shows the setup for the Create Case step.

Chapter 6. Implementing business processes: Case study 177

 Table 6-26 Component and function setup for the Create Case step
Component name Function Parameter Expression
name Name

More_Operations createFolder objectStoreName This should be set to the

object store you are
using on your system.
For our case study, it is
“objectstore”

parentFolder ParentFolder

className "Folder"

properties {"FolderName",
"STRING", ClaimID}

folder CaseFolder

More_Operations searchForOne searchObject SearchPolicy

objectType "document"

itemIds {1}

values {LastName}

object PolicyDoc

CE_Operations file sourceDocument ClaimForm

destFolder CaseFolder

Note: It is important that the functions are added in the order listed. When
control is passed to a component step, the functions are executed in a serial
order. Also, remember the quotes around the expressions. In an expression, if
the quotation marks are used, then the value of the parameter is set to the
value within the quote. Finally, you must press Enter after each entry to
ensure that the values are registered.

For details of how to set up a component step, refer to 6.5.1, “Setting up a

component step” on page 169.

The descriptions of the parameters used in the More_Operations functions can

be found in the Installation Guide for the More_Operations class. The description
of the parameters for the CE_Operation class can be found in the product
manual.

178 Introducing IBM FileNet Business Process Manager

Setting up the Retrieve Related Docs step
The Retrieve Related Docs step is also a component step where we search for
documents related to the claim. Table 6-26 shows the setup for the Retrieve
Related Docs step.

Table 6-27 Component and function setup for the Retrieve Related Docs step
Component name Function Parameter Expression
name Name

More_Operations searchForMany searchObject SearchSupportingDocs

objectType "document"

itemIds {1}

values {LastName}

objects SupportingDocs

Note: This step only searches for documents related to the claim but it does
not do anything. To simplify the case study, we keep the step as is. In real life,
we would probably want to search for the related documents and then add to
the claim folder.

Setting up the Field Agent step

For our case study, we set up the Field Agent step as follows:
򐂰 The General tab:
– Instruction: Please include your damage estimate and file all
supporting documents before submitting.
– Participants: Field_Agent workgroup
– Participant Privileges: Clear all
򐂰 Parameters (see Table 6-28).

Table 6-28 Field Agent parameters

Available parameters Available parameters Available parameters

AccidentDescription Company LastName

AdjustorName DebitAccountNumber LossLocation
Agency EffectiveDate PolicyNumber
AgencyPhone Email ReportedDate

Chapter 6. Implementing business processes: Case study 179

 Available parameters Available parameters Available parameters

AgentName ExpirationDate SupportingDocs

CaseFolder FirstName VehicleEstimateAmount
ClaimID Injury VehicleReplacement
ClaimSettlementType InsuredNameAddress VehicleVIN

򐂰 Assignments.
To simplify the case study, before the execution of the step, we assign
SupportingDocs to an array of empty objects. See Table 6-29.
Table 6-29 Before execution tab
Name Expression

SupportingDocs {““}

After the execution of the step, we want to count the number of supporting
documents that have been added to the claim case:
– If there is no support document added to the claim folder, which is
elementcount(SupportingDocs) = 0, then:
• Set the AttachmentCheck field to false.
• Set F_Subject to the following text:
"Please add the supporting documents to this claim: "+ClaimID
– Otherwise:
• Set the AttachmentCheck field to true.
• Set F_Subject to the following text:
"New claim ClaimID: " + ClaimID
The AttachmentCheck field value (true or false) is used in the conditional
routing to determine whether the control should go back to the Field Agent
step again or move forward to the next step. If the controls are to go back to
the Field Agent step, we also set the F_Subject value to inform the field agent
at that step that supporting documents have to be added to the claim.

180 Introducing IBM FileNet Business Process Manager

 See Table 6-30 for the assignment set up after the step completion.
Table 6-30 After completion tab
Name Expression

AttachmentCheck if(elementcount(SupportingDocs) = 0, false, true)

F_Subject if(elementcount(SupportingDocs) = 0, "Please add
the supporting documents to this claim: " +
ClaimID, "New claim ClaimID: " + ClaimID)

Setting up the File Supporting Docs step

This is another component step where we use More_Operations to file all the
supporting documents to the claim folder. Table 6-26 shows the setup for the File
Supporting Docs step.

Table 6-31 Component and function setup for the File Supporting Docs step
Component name Function Parameter Expression
name Name

More_Operations fileMany folder CaseFolder

objects SupportingDocs

Setting up the Get Field Agent system step

The Get Field Agent step is a system step. Refer to the next section on how to
set up a system step.

Setting up the Get Adjustor system step

The Get Adjustor step is a system step. Refer to the next section on how to set
up the system step.

6.6.2 Setting up system steps

A system step enables you to call system functions provided by IBM FileNet
Business Process Manager. Some of the system functions include assigning a
data field in the workflow, launch another workflow process, and execute a
procedure in a database.

For our case study, in the Claim Setup submap, we use two system steps:
򐂰 Get Field Agent
򐂰 Get Adjustor

Chapter 6. Implementing business processes: Case study 181

 Setting up the Get Field Agent step
For the case study, if the loss city of the claim is Los Angeles, then the field agent
is assigned to Fran. Otherwise, the field agent is assigned to Fred. In real life,
you want to make the assignment to a workgroup rather than a specific field
agent name. In addition, you might want to obtain the field agent name from a
database according to the loss location. For simplicity of the case study, we
hard-code the condition.

To set up the Get Field Agent system step:

1. Make sure you select the Claim Setup submap from the drop-down list above
the map window.
2. Select the Get Field Agent system step from the Claim Setup submap.
3. From the Properties pane on the right side, select Assign from the Available
Functions list, add it to the Selected Functions list as shown in Figure 6-45.

Figure 6-45 Set up Get Field Agent system step: Add Assign function

4. Double-click Assign from the Selected functions and the Assign dialog box
opens.
5. Select FieldAgent from the drop-down list as shown in Figure 6-46.

182 Introducing IBM FileNet Business Process Manager

Figure 6-46 Expression builder for FieldAgent assignment

6. Click the Expression input box. Click the Expression icon and the Expression
Builder dialog box opens.
7. Enter the following expression on the lower input box and click OK:
if(LossCity = "Los Angeles", "Fran", "Fred")
8. The Assign dialog box should look similar to Figure 6-47. Click Close.

Note: Make sure to periodically save your work.

Figure 6-47 Set up Get Field Agent system step, Assign function, FieldAgent parameter

Chapter 6. Implementing business processes: Case study 183

 Setting up the Get Adjustor step
The Get Adjustor step is a system step. If the estimated vehicle repair amount
(VehicleEstimateAmount) is less than or equal to 7000, assign Annie as the
adjustor for the claim. Otherwise, assign Author as the adjustor.

This is the exact same setup as for the Get Field Agent step. Use the Assign
function, and assign Adjustor with the following expression:
if(VehicleEstimateAmount <= 7000, "Annie", "Author")

See Figure 6-48 for the field assignment.

Figure 6-48 Set up Get Adjustor system step, Assign function, Adjustor parameter

6.6.3 Linking the submap to the main map

To link a submap to the main workflow process map or calling map:
1. Select the calling map from the drop-down list as shown in Figure 6-49. For
our case study, we select Workflow (Main Map).

Figure 6-49 Select Workflow (Main Map)

184 Introducing IBM FileNet Business Process Manager

2. Select the submap step on the main map that you want to link to a submap. In
our case study, we select the Claim Setup step.
3. From the Properties pane on the right side, select the called submap from the
Map drop-down list for your submap step. The drop-down list contains a list of
submaps you created for this workflow process. For our case study, we select
Claim Setup Submap from the drop-down list. See Figure 6-50.

Figure 6-50 Linking the submap step with the called submap name

4. To test the connection, double-click the submap step and the called submap
should open. For our case study, click the Claim Setup step on the main map
and the Claim Setup submap opens.

Validating map
To validate your work, select File → Validate. Make sure you resolve any issues
before continue with the rest of the chapter.

Chapter 6. Implementing business processes: Case study 185

6.6.4 Checking in and transferring your workflow process map
In 6.3.2, “Saving the workflow process” on page 142, we mentioned that there
are several ways to save your workflow process. Before the workflow process
can be launched, you have to perform the following steps:
1. Check in the workflow process to Content Engine.
2. Transfer the workflow process from Content Engine to the Process Engine.

To check in, use File → IBM FileNet Add New from Process Designer. For
details, refer to “Option 2: IBM FileNet Add New” on page 143.

To transfer the workflow process, use File → Transfer.

6.7 Creating a workflow subscription

After you finish creating a workflow, you can create a workflow subscription that
specifies when to launch the workflow based on events occurred on documents,
folders, or custom objects. Events include creation and modification of
documents, folders, or custom objects. For our case study, the event that
launches the workflow is the creation of a Claim form. The Claim form is a
custom object class.

Note: Before you create a workflow subscription, make sure that the
associated workflow is saved, checked into Content Engine, and transferred
to the Process Engine.

186 Introducing IBM FileNet Business Process Manager

To create a workflow subscription, follow these steps:
1. From the workplace main menu, select Author → Advanced Tools → Add
Workflow Subscription as shown in Figure 6-51.

Figure 6-51 Add workflow subscription through workplace

Chapter 6. Implementing business processes: Case study 187

 2. Select the target object class that launches the workflow:
a. Navigate through the object store to where the target object will be stored.
For our case study, we select objectstore. See Figure 6-52.

Figure 6-52 Add workflow subscription: Select target object store

b. Select the target object class type. For our case study, the object type of
the Claim object class is a custom object. We select Custom Object. See
Figure 6-53.

Figure 6-53 Add workflow subscription: Select target object class type

188 Introducing IBM FileNet Business Process Manager

 c. Select the target object class and click Next. For our case study, the object
class is called Basic Process. We select Basic Process. See Figure 6-54.

Figure 6-54 Add workflow subscription: Select target object class

3. Select the workflow process that is launched by the subscription:

a. On the left side of the window, the arrow should be placed next to the
Select Workflow step. Click Browse/Search for Workflow Definition.
See Figure 6-55.

Figure 6-55 Add workflow subscription: Select workflow process, browse the folders

Chapter 6. Implementing business processes: Case study 189

 b. Navigate to the location where your workflow is stored in the object store.
For our case study, we select objectstore → Case Study Process. See
Figure 6-56.

Figure 6-56 Add workflow subscription: Select the workflow process

190 Introducing IBM FileNet Business Process Manager

 c. Click Select from Versions under your workflow process and select the
version of the workflow that you want the workflow subscription to launch.
For our case study, we click the link under XYZ Basic Process. See
Figure 6-57.

Figure 6-57 Add workflow subscription: Select the workflow version to launch

Chapter 6. Implementing business processes: Case study 191

 d. Find the workflow version that you want the workflow subscription to
launch and click Select under the version. For our case study, we select
the latest version, 8.0, of our workflow. See Figure 6-58.

Note: In most cases, you select the most current workflow version.
However, you have the option to select an older version of a workflow if
required.

Figure 6-58 Add workflow subscription: Select workflow version

e. Click Next after the workflow version has been selected.

4. Set the subscription properties:
a. Enter information for the following fields (see Figure 6-59):
• Name: Enter a name for the workflow subscription. For our case study,
we enter XYZ Basic Process Subscription.
• Description: Enter an appropriate description for the subscription.
• Initial State: Leave the check box selected.
• Enable Manual Launch: Change it to No.

192 Introducing IBM FileNet Business Process Manager

Figure 6-59 Setting properties

b. Click Subscribed Events.

c. Subscribe the events that launch a workflow. The available events for our
case study’s target object class include:
• Change Class Event
• Creation Event
• Deletion Event
• Lock Event
• Unlock Event
• Update Event
• Update Security Event
There are different events available for subscription based on the target
object class that you selected to launch a workflow.

Chapter 6. Implementing business processes: Case study 193

 To subscribe to events, select the entries from the Available Events list on
the left side of the windows, and click the right arrow key to move them to
the Selected Events list on the right side. For our case study, we select
Creation Event. This that means when our target object is created, the
workflow is launched. See Figure 6-60.

Figure 6-60 Select the creation event

5. Click Accept and Next to move to the next step. See Figure 6-61.

194 Introducing IBM FileNet Business Process Manager

Figure 6-61 Click next to move to set expression

6. Click Next to move on to the next pane. We do not build any special
expressions for the workflow subscription. Click Next again to continue to the
next step.
7. You should now be at the Set Property Map step. See the arrow on the left
side of the pane as shown in Figure 6-62.
In this step, you map the properties of your target object class to the data
fields used of the workflow to be launched:
a. From the left side, the Data Field Name drop-down box, select the data
field of the workflow that requires data information. For our case study, we
select Agency (String). See Figure 6-62.

Chapter 6. Implementing business processes: Case study 195

 Note: The data fields on the left side of the pane represent the data fields
you define for the workflow (which is stored in the Process Engine). The
properties on the right side of the pane represent the properties you define
for your target object class (which is stored in Content Engine).

When you launch a workflow based on the subscribed event that occurred
on a target object, the property information from the target object will be
mapped to the data fields of the launched workflow. This way, the workflow
gets the information it needs from the target object to do its work.

Figure 6-62 Set property map: Select the data field of the target object class

b. From the right side, the Property Name drop-down box, select the property
of the target object class that triggers the workflow. For our case study, we
select Agency (String). See Figure 6-63.
c. Click the plus sign icon to complete the mapping. See Figure 6-64.
d. Repeat the previous three steps for other property mappings. When all
data fields of the workflow are mapped by the properties of the target
object class, click Next.

196 Introducing IBM FileNet Business Process Manager

Figure 6-63 Set property map: Select the property name of the workflow to be launched

Figure 6-64 Set property map: Click the plus sign icon to complete the mapping

Chapter 6. Implementing business processes: Case study 197

 For our case study, we map the data fields of the target object class with the
properties of the workflow as shown in Table 6-32.

Note: If you follow through our case study example, make sure that you
map things precisely as described in the table.

Table 6-32 Property mapping for case study

Data Field Name Property Name
(of the workflow) (of the target object class)

AccidentDate Date

Agency Agency

AgentName AgentName

ClaimID ClaimID

Company Company

Creator Creator

EffectiveDate EffectiveDate

LastName LastName

LossLocation LossLocation

PolicyNumber PolicyNumber

VehicleEstimateAmount VehicleEstimateAmount

VehicleReplacement VehicleReplacement

The data fields of the workflow and the properties of the target object class
might not always have a one-to-one naming correspondence. For our case
study, the workflow uses a data field called AccidentDate. The object class
uses a property called Date. We map Date from the object class to
AccidentDate in the workflow.
8. The last step in the workflow subscription creation process includes setting
the security. In this step, you can specify who has owner control, modify
control, view control, and remove right of the workflow subscription. For our
case study, we use the default.

198 Introducing IBM FileNet Business Process Manager

 9. Click Finish to complete the workflow subscription creation procedure. See
Figure 6-65.

Figure 6-65 Add workflow subscription: Click Finish to complete the subscription

Note: You can also create a workflow subscription using IBM FileNet
Enterprise Manager if you want to create a workflow subscription for a
particular document, folder, or custom object. For simplicity of discussion, we
do not cover this topic in the book.

6.7.1 Exposing a system property and mapping it in a workflow

System properties cannot be mapped to the workflow’s data fields by default.
However, there are times when the system properties have to be exposed to the
launched workflow. For example, in our case study, we would like to capture the
person that starts the workflow. This is the person who creates a target object
that then triggers the workflow to be launched. The system property that contains
this person’s information is Creator. By default, it is not in the object class. This
property needs to be exposed in the object class that we have built and use to
launch the workflow.

If you imported the Content Engine objects from the downloadable materials
provided with this book, the Creator property has already been exposed in the
object class.

Chapter 6. Implementing business processes: Case study 199

 Because it is important to understand how to expose system properties in case
you are required to do so, we outline the steps to accomplish this task here:
1. Start the IBM FileNet Enterprise Manager application:
From Content Engine, select Start → Programs → IBM FileNet P8
Platform → Enterprise Manager SnapIn.
2. Navigate to the target object class.
For our case study, the target object class we use is a custom object, called
Basic Process, and it is stored in our object store called objectstore.
Therefore, we select Object Stores → objectstore → Other Classes →
Custom Object → Basic Process. See Figure 6-66.

Figure 6-66 Browse to the target object class - For our case study, it is Basic Process

200 Introducing IBM FileNet Business Process Manager

3. Open the object class properties dialog box:
Right-clicking your target object class and select Properties from the pop-up
menu. For our case study, we right-click Basic Process and select
Properties.
4. Expose the system property in the object class:
a. Select the Property Definitions tab. See Figure 6-67.

Figure 6-67 System properties definitions

b. Select the System Properties check box. This enables the system
properties to be exposed for property mapping when creating a workflow
subscription.
c. Select the system property you want to expose for property mapping. For
our case study, we select Creator.

Chapter 6. Implementing business processes: Case study 201

 d. Click Edit. The information for the system property displays (see
Figure 6-68).
e. Find the symbolic name for the system property (in the Symbolic Name
field). This is the name that you use for property mapping when creating a
workflow subscription. It might not be the same name as what you see
from the previous dialog box. Make a note of the symbolic name.

Figure 6-68 Symbolic name

f. Click OK to go back to the object class property dialog box.

5. Map the system property exposed earlier in the workflow subscription setting:
a. Select the Subscription tab (from the object class properties dialog box).
b. Select the workflow subscription you created earlier under the
Subscription Name section. For our case study, we select Basic Process
Subscription. See Figure 6-69.

202 Introducing IBM FileNet Business Process Manager

Figure 6-69 Basic process subscription properties

c. Click Properties to show the workflow subscription properties. See

Figure 6-70.

Figure 6-70 Basic process subscription properties workflow tab

Chapter 6. Implementing business processes: Case study 203

 d. Select the Workflow tab as shown in Figure 6-71.

Figure 6-71 Add system property to the subscription

e. Add the system property that you want to expose here:

i. Click Add. The Property Map dialog box displays (see Figure 6-72).
ii. Enter the data field name of the workflow (Workflow Property) to be
mapped to. For our case study, we enter Creator.
iii. Enter the system property name (Document Property) to be mapped.
For our case study, we enter Creator.
iv. Click Add.

Figure 6-72 Map the system to process property

204 Introducing IBM FileNet Business Process Manager

 6. Click OK in the object class properties dialog box. For our case study, see
Figure 6-73. Notice that the system property Creator is mapped in the
workflow subscription.

Figure 6-73 Complete exposing of the system property and its mapping to the workflow

6.8 Testing the basic process

Always test your workflow processes when you have completed them. Make
sure that you test the workflow subscription to ensure that the specified event
does indeed launch the workflow.

6.8.1 Monitoring workflow processes

To test your workflow processes, we first describe two application tools that you
can use to monitor your workflow processes:
򐂰 Process Tracker
򐂰 Process administrator

Chapter 6. Implementing business processes: Case study 205

 Process Tracker
Process Tracker provides the status of a workflow that is currently running in a
graphical view. From the Process Tracker graphic view window, you can tell what
steps have been completed in the workflow, when they were completed, and
which steps are currently active.

Enabling users to use Process Tracker

To enable users to use Process Tracker (other than the Administrator), users
must be added in the F_Trackers workflow group as follows:
1. Navigate to the workflow group setup:
Open the workflow and go to Workflow Properties → Workflow Groups.
See Figure 6-74.

Figure 6-74 Add workflow group in a workflow property: F_Trackers

2. Add participants to the F_Trackers workflow group. The participants of this

workflow group will be allowed to track the workflow process from Workplace.

Note: To save the time and effort of logging in and out as different users,
this process is set up so that testing can be done using the administrator’s
task box. The Administrator has not been added to the F_Trackers
workgroup, because the Administrator has access to the Process
Administrator, which can view Process Tracker. If you would like to add
Administrator to the F_Trackers workgroup now, you have to save and
transfer the new process and then modify your subscription to use the new
version of the process.

206 Introducing IBM FileNet Business Process Manager

Launching Process Tracker
To launch Process Tracker:
1. From Workplace, go to Tasks → Task Tracker (see Figure 6-75).
2. Select the workflow process you want to track.

Figure 6-75 Workplace task tracker

Using Process Tracker

Process Tracker displays a combination of information from the workflow
definition, the values in the active work items, and the event logs. The Process
Tracker’s main window is divided into the following areas:
򐂰 The workflow map that shows the workflow’s current status visually.
The map shows the steps and routes defined for the workflow. The currently
active work items are indicated with the blue arrow. Steps that have been
traveled are indicated by the completed check marks. See Figure 6-76.

Chapter 6. Implementing business processes: Case study 207

 Figure 6-76 Process tracker window workflow map

򐂰 The workflow properties that show the details of the current status.
On the right side of the window, you can see the details of the selected item of
a workflow, a step, or a route. See Figure 6-77.

Note: The cursor in the figure is pointing to a blue arrow. The blue arrow
and the one above it allow you to expand either the workflow map or the
properties pane to the full width of the window. These arrows can also be
found on the Process Designer window.

208 Introducing IBM FileNet Business Process Manager

Figure 6-77 Process tracker window properties pane

To show the progress of a process, click the Refresh icon. See Figure 6-78.

Figure 6-78 Refresh the view of the process

Chapter 6. Implementing business processes: Case study 209

 Note: The boxed number to the left of the Create Case step indicates the
number of components that have successfully completed within that step
(remember that there are three operations associated with this step in our
case study).

Also in this view, the step currently being executed is the Field Agent step,
indicated by the hour glass. This tells you that input from the field agent is
required. It is waiting for the field agent to add supporting documents.

Process Administrator
With Process Administrator, you can search for and view workflows, edit
workflow data and properties, and manage workflows.

Launching Process Administrator

To launch the Process Administrator application, go to Admin → Process
Administrator from Workplace. See Figure 6-79.

Figure 6-79 Launch Process Administrator application

210 Introducing IBM FileNet Business Process Manager

Using Process Administrator
Process Administrator provides a wide variety of options so you can focus your
search very precisely. You can enter your search criteria and click Find Now to
have the Process Administrator application find the workflow processes that you
are looking for. See Figure 6-80.

Figure 6-80 Process administrator search.

After you locate one or more workflows or work items (as shown in Figure 6-81,
for example), you can use Process Administrator to make various changes to the
work in progress, such as these:
򐂰 Complete a step and send it on to the next step.
򐂰 Modify workflow field values.
򐂰 Assign users to or remove users from a workflow group.
򐂰 Delete an entire workflow or one or more work items.

Chapter 6. Implementing business processes: Case study 211

 Figure 6-81 Search for currently running workflow processes

The highlighted task in the figure shows a workflow process currently running.
The F_Subject column shows the completed expression you built earlier and
assigned to F_Subject. The cursor is currently pointing to the Process Tracker
icon. By clicking this icon, you launch Process Tracker to view the current status
of the highlighted workflow process.

6.8.2 Testing the workflow process from beginning to end

To test the workflow process that you created, we recommend adding an object
and triggers that are associated with the workflow and exercise through the
entire workflow process from beginning to end. For our case study, we perform
the following tasks to test the workflow process:
1. Create a new claim.
2. Add the supporting documents.
3. Complete the claim process.
4. View the completed claim.

212 Introducing IBM FileNet Business Process Manager

Creating a new claim
For our case study, we test the workflow process by create a new claim:
1. From Workplace, navigate to your object store and the folder that contains
your target object. For our case study, we select Browse → objectstore →
Claims. See Figure 6-82.

Figure 6-82 Add a new claim: Go to the Claims folder

Chapter 6. Implementing business processes: Case study 213

 2. Click Add Custom Object (see Figure 6-83).

Figure 6-83 Add a new claim: Add a custom object

3. Click Change Class (see Figure 6-84).

Figure 6-84 Add a claim: Change target object class

214 Introducing IBM FileNet Business Process Manager

4. Select the target class. For our case study, we select Basic Process from the
class list (see Figure 6-85).

Figure 6-85 Add a claim: Select target object class

Chapter 6. Implementing business processes: Case study 215

 5. Fill in the target object information. For our case study, we enter claim
properties (see Figure 6-86). Click Next.

Note: The only required field is the Claim ID. It is required to create the
claim folder.

Figure 6-86 Add a claim: Enter claim information

216 Introducing IBM FileNet Business Process Manager

6. Accept the default security settings given (see Figure 6-87) and click Finish.

Figure 6-87 Add a claim: Use default security and complete the claim

7. Click OK on the Add Confirmation window (Figure 6-88).

Note: The long name listed in the confirmation window is the internal ID for
the object.

Figure 6-88 Add a claim: Add confirmation window

Chapter 6. Implementing business processes: Case study 217

 8. For our case study, when a claim is added, a new folder is created to hold the
documents related to the claim. The new documents are also stored in the
claim folder. See Figure 6-89.

Figure 6-89 New claim information icon

218 Introducing IBM FileNet Business Process Manager

9. Located to the right of the new claim is an information icon. Clicking the
information icon displays the data that is entered for the claim (see
Figure 6-90).

Figure 6-90 Claim data

10.Click Exit to close the claim data window.

Chapter 6. Implementing business processes: Case study 219

 Adding supporting documents for the new claim
After the claim is added to the system, a field agent must review and claim and
add any supporting documents to the claim folder. To test if this step works as
expected, we walk through the adding of the supporting documents step:
1. From Workplace, click Tasks (see Figure 6-91).

Note: For this test, we have been logged in as Administrator so we can

monitor the process with Process Tracker, and step the claim through the
workflow process.

Figure 6-91 Look at the administrators tasks

220 Introducing IBM FileNet Business Process Manager

2. Click on the claim we just entered (see Figure 6-92).

Note: If the new claim does not appear here, this might be for one of two
reasons:
򐂰 The step that moves the new claim to the field agent has not completed
yet. Wait a while for the step to be completed.
򐂰 There might be a problem with the workflow process definition. The
problem can be seen in Process Tracker. Navigate from Workplace →
Tasks → Task Tracker to see where the workflow process stopped.

Figure 6-92 Click the new claim

3. Click SupportingDocs (see Figure 6-93 on page 222).

4. Click Assign (see Figure 6-94 on page 223).

Note: Documents can be attached to the claim in many ways. In our case
study, we choose to attach the ones that reside in the content repository.
The Add New link enables you to add documents outside of the repository.

Chapter 6. Implementing business processes: Case study 221

 Figure 6-93 Add supporting documents

222 Introducing IBM FileNet Business Process Manager

Figure 6-94 Add document from repository

Chapter 6. Implementing business processes: Case study 223

 5. Click the repository that contains the document (see Figure 6-95).

Figure 6-95 Browse the object stores

6. Click the folder that contains document (see Figure 6-96).

Figure 6-96 Find the folder

224 Introducing IBM FileNet Business Process Manager

7. Click the supporting document (see Figure 6-97).

Figure 6-97 Click the supporting document

8. Click Complete to move the claim to the next step in the workflow process
(see Figure 6-98).

Chapter 6. Implementing business processes: Case study 225

 Figure 6-98 Click complete

226 Introducing IBM FileNet Business Process Manager

Completing the claim process
After completing the field agent review and adding the supporting documents to
the claim folder, it reappears in the field agent’s (administrator) task box. The
field agent has to complete the claim process to test the complete workflow
process.

To complete the claim process:

1. From Workplace (as the field agent), go to Tasks → My Inbox.
2. Select the processed claim to view (see Figure 6-99).

Figure 6-99 View claim in process

Chapter 6. Implementing business processes: Case study 227

 3. Click Complete to complete the claim (see Figure 6-100).

Figure 6-100 Complete the claim

Viewing the completed claim

After the claim is processed, view the completed claim to double check that the
entire workflow process works as expected:
1. From Workplace, navigate back to the claims folder to view the claim by
Browse → objectstore → claims (see Figure 6-101).

228 Introducing IBM FileNet Business Process Manager

Figure 6-101 Go to the claims folder

2. Select the new claim folder (see Figure 6-102).

Figure 6-102 Click the new claim folder

3. View the claim, the supporting documents, and the claim processed letter
(see Figure 6-103).

Chapter 6. Implementing business processes: Case study 229

 Figure 6-103 Claim documents

6.9 Creating and calling a Web service

Web services provide functions that can be used by other processes. There are
many ways to create Web services. In this section, we show you how to create a
Web service from a workflow process. In other words, we show you how to
expose a workflow process such that it can be used as a Web service.

For our case study, we create two new workflow processes (GetAdjustor, and
GetFieldAgent) and expose them as Web services. These new processes (Web
services) replace the existing steps (Get Adjustor and Get Field Agent) in the
basic workflow process.

Note: We cover the Web services concept in Chapter 7, “Integration with

external systems and services with BPM” on page 257 and describe its
complete implementation in Chapter 8, “Implementing Component Integrator
and Web services” on page 271. If you are not familiar with Web services, or if
you encounter problems while following the case study, we recommend
reading these chapters first before you continue with the rest of the case study
implementation.

230 Introducing IBM FileNet Business Process Manager

6.9.1 Creating a workflow process as a Web service
Creating a workflow process to be used as a Web service consists of the
following main steps:
1. Create a new workflow and define its workflow properties.
Define the workflow properties as a normal workflow, except in the following
two areas:
– Data fields: These are the data fields that the workflow requires to process
work. If this workflow process requires the calling process to pass in
information, the parameters that contain the input information from the
calling processes should be defined in the workflow properties definition.
– Web services name and type: You must specify the name with which this
workflow process is called by other processes. This name becomes the
Web service name. Because this process is to be exposed as a Web
service, it has to support Receive and Reply functions. Therefore, you
have to set this process as a Receive/Reply Web service type in the
workflow properties definition.
2. Create workflow steps and routes.
Create the workflow steps and routes as a normal workflow. In addition, in
order to enable this workflow process to be called as a Web service, the
workflow process must contain at least one Receive step and optionally
multiple Reply steps as follows:
– Receive step: You must specify at least one Receive step in the workflow
process. The operation parameters defined for the Receive step are the
parameters provided by the calling process. The operation parameters
should be defined in the data fields of the workflow properties.
– Reply step: You can optionally include multiple Reply steps in the
workflow process. The operation parameters and their values defined for
the Reply step will be returned to the calling process.
3. Validate and check in the workflow process.

For our case study, we create two workflow processes that are to be used as
Web services:
򐂰 GetFieldAgent
򐂰 GetAdjustor

Creating GetFieldAgent workflow process as a Web service

The GetFieldAgent workflow process duplicates the functions of the Get Field
Agent step in the basic workflow process we defined earlier, but it will be used as
the Web service that other processes call.

Chapter 6. Implementing business processes: Case study 231

 For simplicity of the workflow, we hard-coded the names of the field agents. The
GetFieldAgent workflow process assigns either Fred or Fran as the field agent
based on the location information provided by the calling process.

Defining workflow properties

To define workflow properties, follow these steps:
1. Start Process Designer:
From Workplace, select Author → Advanced Tools → Process Designer.
2. Open the workflow properties:
From the top menu, select View → Workflow Properties.
3. Select the General tab and enter the following information (see Figure 6-104):
– Workflow Name: Enter GetFieldAgent.
– Subject: Enter GetFieldAgent.

Figure 6-104 GetFieldAgent workflow property - General tab

4. Select the Data Fields tab and add the City and State fields as shown in
Figure 6-105. When the other workflow process calls this Web service, the
other workflow process should supply the City and State values.

Figure 6-105 GetFieldAgent workflow property - Data Fields tab

232 Introducing IBM FileNet Business Process Manager

5. Select the Web services tab and then select the Partner Links tab. See
Figure 6-106. This is where you define the Web service name and type.

Figure 6-106 GetFieldAgent workflow property - Web services -> Partner Links tab

6. Enter the information in the page as shown in Figure 6-107:

– Double-click the cell under name in the Partner Links section, and type in
GetFieldAgent.
– Click the Receive/Reply check box.
– Type in GetFieldAgent for the Process Port Type field.
7. Click Close to close the Workflow Properties window.

Figure 6-107 GetFieldAgent workflow property - Web services -> Partner Links data

Chapter 6. Implementing business processes: Case study 233

 Adding the steps
For our case study, add the steps of the workflow process as follows:
1. From the Process Designer map, select Web services Palette from the View
Palette on the bottom of the window. See Figure 6-108.

Figure 6-108 GetFieldAgent workflow map: Change to Web services Palette

2. Place the steps on the workflow map as shown in Figure 6-109.

The GetFieldAgent step is a Receive Web service step. The Fred and Fran
steps are Reply Web services steps.

Figure 6-109 GetFieldAgent workflow map: Draw the required steps

234 Introducing IBM FileNet Business Process Manager

Adding routes and setting up condition routing
Add the routes and set up the condition routings as follows:
1. Add routes as shown in Figure 6-110. Name the routes to Los Angeles and
Denver as shown in the figure.
2. For the Denver route, if the City parameter is equal to “Denver”, the route
goes to the Fran step. Set up the conditional routing for the Denver route as
follows:
a. Select Condition in the Properties pane.
b. Select the Data Fields tab.
c. Select City (String) from the Field drop-down list.
d. Select is equal from the Operator drop-down list.
e. Enter “Denver” in the Value field. Make sure to include the double quotes.
f. Click Insert. Without clicking this button, the condition will not be saved.

Figure 6-110 GetFieldAgent workflow map: Add routes

3. For the Los Angeles route, if the City parameter is not equal to “Denver”, the
route goes to the Fred step. Set up the conditional routing for the Los Angeles
route as follows:
a. Select Condition in the Properties pane.
b. Select the Data Fields tab.
c. Select City (String) from the Field drop-down list.
d. Select is not equal from the Operator drop-down list.
e. Enter “Denver” in the Value field. Make sure to include the double quotes.
f. Click Insert. Without clicking this button, the condition will not be saved.

Chapter 6. Implementing business processes: Case study 235

 Note: The system evaluates the routes in the order that they are created or in
the order that you set up in the source step. If there are only two possible
routes, for the route that will be evaluated at the last, you can select Always
true without doing any special conditional routing setup.

Finishing defining the steps

After we add the steps and routes to the workflow map, we have to finish defining
the details of each step:
1. Define the GetFieldAgent step:
a. Select the GetFieldAgent step from the map and select GetFieldAgent
from the Partner Link drop-down list as shown in Figure 6-111.

Figure 6-111 GetFieldAgent workflow map: GetFieldAgent step - Partner Link property

236 Introducing IBM FileNet Business Process Manager

 b. Type in GetFieldAgent in the Operation field. See Figure 6-112.

Figure 6-112 GetFieldAgent workflow map: GetFieldAgent step - Operation property

c. Under the shaded Operation Parameters, specify the input parameter

passed from the calling process (see Figure 6-113):
i. Click the first cell under Name, and type in City and press Enter.
ii. Select String from the Type drop-down list.
iii. Select City from the Field Name drop-down list. If there is nothing in
the drop-down list, type City and press Enter.
iv. Repeat the same steps for the State parameter.
This workflow process takes the City and State information from the
calling process and assigns a field agent based on the information. For
simplification, although we pass both fields into the workflow process, we
use just the City field when assigning a field agent.

Chapter 6. Implementing business processes: Case study 237

 Figure 6-113 GetFieldAgent workflow map: GetFieldAgent step - Operation parameters

2. Define the Fred step:

a. Select the General tab. Select GetFieldAgent from the Partner Link
drop-down list and select GetFieldAgent from the Operation drop-down
list (see Figure 6-114).

Figure 6-114 GetFieldAgent workflow map: Fred step - Operation property

238 Introducing IBM FileNet Business Process Manager

 b. Under the Operation Parameters, enter FieldAgent for the Name field,
select String from the Type field, and enter “Fred” in the Expression field.
Remember to include the double quotes around Fred. See Figure 6-115.

Figure 6-115 GetFieldAgent workflow map: Fred step - Operation parameters

3. Define the Fran step using the similar procedures as the Fred step. Instead of
“Fred”, type “Fran” in the Expression field under the Operation Parameters
section.

Validating workflow process map

Using one of the validation methods we described earlier, validate the
GetFieldAgent workflow process. There should be no errors. If you receive any
errors, use the error messages to find and correct the problem.

Reminder: To validate a workflow process using the menu commands, select

File → Validate.

Checking in the workflow process

After you validate that the workflow process has no error, transfer and check in
the workflow process. See Figure 6-116. The transfer operation makes the
process available for use.

Reminder: To transfer a workflow process using the menu commands, select

File → Transfer.

Chapter 6. Implementing business processes: Case study 239

 Figure 6-116 Validate, transfer, check in

Creating GetAdjustor workflow process as a Web service

The GetAdjustor workflow process duplicates the functions of Get Adjustor step
in the basic workflow process we defined earlier, but it is to be used as the Web
service that other processes call.

For simplicity of the workflow, we hard-coded the names of the adjustors. The
GetAdjustor workflow process assigns Author or Annie as the adjustor name
based on the vehicle’s estimated repair amount provided by the calling process.

Defining workflow properties

To define the workflow properties, follow these steps:
1. Select File → New Workflow.
2. Select View → Workflow Properties.
3. In the General tab, type GetAdjustor for both Workflow Name field and the
Subject field. See Figure 6-117.

240 Introducing IBM FileNet Business Process Manager

Figure 6-117 GetAdjustor workflow properties: General tab

4. In the Data Fields tab, add the data field Amount, with type as Float. See
Figure 6-118. When other workflow process calls this Web service, the other
workflow process should supply the Amount value for this Web service.

Figure 6-118 GetAdjustor workflow properties: Data Fields tab

5. In the Web services tab, go to the Partner Links tab. Enter the Partner Links
information as shown in Figure 6-119. Note that we set the Web service with
the name, GetAdjustor, of type Receive/Reply. This means that the Web
service contains both Receive and Reply functions.

Figure 6-119 GetAdjustor workflow properties: Web services tab

Adding the steps and their routes

For our case study, we add the steps and routes in the workflow map as shown
in Figure 6-120. The GetAdjustor step is the Receive step. The Author and Annie

Chapter 6. Implementing business processes: Case study 241

 steps are the Reply steps. The route that connects GetAdjustor to Author is
labeled >7K. the route that connects GetAdjustor to Annie is labeled as <=7K.

Figure 6-120 GetAdjustor workflow map: Steps and routes

Finish defining the steps and routes

For our case study, both the Author and Annie steps are Reply steps. Each step
sets the AdjustorName value. Based on the conditional routing, only one of the
steps is executed and returns the AdjustorName value to the calling process.

To finish defining the steps for our case study:

1. Define the Author step:
a. Select the Arthur step from the workflow map.
b. In the General tab, fill out the information as shown in Figure 6-121.

Figure 6-121 GetAdjustor workflow map: Author step properties

242 Introducing IBM FileNet Business Process Manager

2. Define the Annie step as shown in Figure 6-122.

Figure 6-122 GetAdjustor workflow map: Annie step properties

3. Define the >7K route as shown in Figure 6-123.

Figure 6-123 GetAdjustor workflow map: >7K route properties

Chapter 6. Implementing business processes: Case study 243

 4. Define the <-7K route as shown in Figure 6-124.

Figure 6-124 GetAdjustor workflow map: <=7K route properties

Validating the workflow process map

Using one of the validation methods we described earlier to validate the workflow
process map. If you get an error, use the error message to correct the problem.

Checking in the workflow process

After validating the workflow process, transfer and check in the workflow
process.

6.9.2 Calling the Web service in a workflow process

After you create a workflow process as a Web service, you can call the Web
service from other workflows.

Calling a Web service from a workflow process consists of the following main
steps:
1. Open the existing workflow and define its workflow properties:
– Web service name and type - You must specify the Web service that the
current workflow process will call (invoke) within its process. In addition,
the specified Web service must be set up as an Invoke Web service type.

244 Introducing IBM FileNet Business Process Manager

2. Add the Invoke step and routes:
– Outgoing Parameters: If the Web service that the workflow process calls
requires input values, the workflow process must provide the information
in the Outgoing Parameters field. This corresponds to the operation
parameters of the Receive step in the invoked Web service (workflow
process).
– Incoming Parameters: If the Web service that the workflow process calls
returns values, the workflow process can get the returned information from
these parameters. This corresponds to the operation parameters of the
Reply steps in the invoked Web service (workflow process).
– Routes and other steps: Create routes and other steps as necessary for
the workflow.
3. Validate and check in the workflow process.

For our case study, we go back to our basic workflow process and update it to
use the two new Web services we created in the previous section.

Defining workflow properties

Before using the existing Web services in a workflow map, you must define the
Web services that you can invoke in the workflow properties:
1. Open the Claim Setup submap that needs to call the Web services.
2. Go to View → Workflow Properties → Web services → Partner Links.
3. Type GetFieldAgent in the Name field. This is the Web service that the
current workflow process will call.
4. Check the Invoke box and click the ellipse button (see Figure 6-125).

Figure 6-125 Claim Setup submap: Web services, Partner Links

Chapter 6. Implementing business processes: Case study 245

 5. Fill in the information as shown in Figure 6-126:
a. Click the Web services workflows radio button.
b. Type get in the Name field.
c. Click Execute.
This gets a list of available Web services that the current workflow process
can call.

Figure 6-126 Claim Setup submap: Browse for the Web services

246 Introducing IBM FileNet Business Process Manager

6. A list of Web services workflow displays. Select the Web service that your
workflow invokes. For our case study, we select GetFieldAgent from the
choices. Click OK (see Figure 6-127).

Figure 6-127 Claim Setup submap: Select the GetFieldAgent service

Deleting the existing steps (optional)

If you have steps in your workflow that will be replaced by the Web services,
delete these steps first from the workflow. For the case study, we delete the Get
Field Agent and Get Adjustor steps.

Chapter 6. Implementing business processes: Case study 247

 To delete the obsolete steps in our case study:
1. From the workflow map, right-click the Get Field Agent step and select
Delete from the pop-up menu (see Figure 6-128).

Figure 6-128 Claim Setup submap: Delete obsoleted steps

2. Repeat the delete procedure for the Get Adjustor step.

248 Introducing IBM FileNet Business Process Manager

Adding Web services steps in the workflow process
To call the Web services in your workflow:
1. Change the View Palette to the Web services Palette (see Figure 6-129).

Figure 6-129 Claim Setup submap: Change to Web services Palette

2. Add the Invoke steps and routes in your workflow. For our case study, we add
two Invoke steps: Get Field Agent and Get Adjustor steps (see Figure 6-130).

Chapter 6. Implementing business processes: Case study 249

 Figure 6-130 Claim Setup submap: Calling Invoke steps

Finishing defining the steps and routes

To define the Invoke steps:
1. Define Get Field Agent step. The is an Invoke step that calls GetFieldAgent
Web service. It requires location information as input. The values come from
the workflow’s City and State data field. Define the step as follows:
a. Select the Get Field Agent step and click the General tab.
b. Select GetFieldAgent from the Partner Link drop-down list (see
Figure 6-131). This is where you specify what Web service you want to
invoke.

Figure 6-131 Calling GetFieldAgent service: Partner Link

250 Introducing IBM FileNet Business Process Manager

 c. Select GetFieldAgent from the Operation drop-down list (see
Figure 6-132). Some Web services have multiple operations. You need to
specify which operation to invoke. For the Web services that we created
earlier, there is only one corresponding operation per Web service.

Figure 6-132 Calling GetfFieldAgent service: Operation

Chapter 6. Implementing business processes: Case study 251

 d. Ensure that the Parameters radio button is checked (see Figure 6-133).
e. Under the Outgoing Parameters section, set the values (in the Expression
fields) for the parameters to be used when invoking the Web service. For
our case study, the Web service requires the input values of City and
State to determine which field agent to assign to the insurance claim case.
Set these values as follows:
i. For the City parameter, from the Expression drop-down list, select
LossCity (see Figure 6-133).
ii. For the State parameter, select LossState.

Figure 6-133 Calling GetFieldAgent service: Outgoing Parameters

252 Introducing IBM FileNet Business Process Manager

 f. Under the Incoming Parameters section, set the returning parameter
names and values from the invoked Web service. For our case study,
select FieldAgent for the Name and Field Name fields. See Figure 6-134.

Figure 6-134 Calling GetFieldAgent service: Incoming Parameters

2. Define the Get Adjustor step. The step is an Invoke step that calls the
GetAdjustor Web service. It requires the Amount parameter as input. Its value
comes from the workflow’s VehicleEstimateAmount data field. Define the step
as follows:
a. Select the Get Adjustor step.
b. In the General tab:
i. Select GetAdjustor from the Partner Link drop-down list (see
Figure 6-135).
ii. Select GetAdjustor from the Operation drop-down list.

Chapter 6. Implementing business processes: Case study 253

 Figure 6-135 Calling GetAdjustor service: Partner Link

c. Ensure that the Parameters radio button is checked. See Figure 6-136.
d. Under the shaded Outgoing Parameters section, for the Amount
parameter, from the Expression drop-down list, select
VehicleEstimateAmount. This is how you set the workflow’s data field
value (from VehicleEstimateAmount) to the input parameter’s value
(Amount) required by the Web service.

Figure 6-136 Calling GetAdjustor service: Outgoing Parameters

254 Introducing IBM FileNet Business Process Manager

 e. Under the shaded Incoming Parameters section, for the AdjustorName
parameter, from the Field Name drop-down list, select AdjustorName.
See Figure 6-137. This is where the invoked Web service returns its
values.

Figure 6-137 Calling GetAdjustor service: Incoming Parameters

Check in and transfer your process

Always check in the process before you can use it. For our case, we check in the
process as XYZ Basic Process WS.

Chapter 6. Implementing business processes: Case study 255

6.9.3 Create a subscription to launch your process
A new subscription needs to be created for the workflow process that you just
modified. Use the following data to create the new subscription:
򐂰 Class - Basic Process WS
򐂰 Process - XYZ Basic Process WS

Note: Remember to select the most current version of the process.

Exposing system data properties

The Creator system property needs to be exposed in the custom class that we
have built and uses to launch the workflow. For procedures on exposing the
system properties, refer to 6.7.1, “Exposing a system property and mapping it in
a workflow” on page 199.

6.9.4 Testing the process with Web services

Test the process as you did before. The only difference is that for this process,
the object class you add for your new claim is Basic Process WS.

256 Introducing IBM FileNet Business Process Manager

 7

Chapter 7. Integration with external

systems and services with
BPM
In this chapter we introduce, from a high level business perspective, how IBM
FileNet Business Process Manager system integrates with external systems and
services.

We cover the following topics:

򐂰 Component Integrator
򐂰 Web services
򐂰 Rules connectivity framework

Next, in Chapter 8, “Implementing Component Integrator and Web services” on

page 271, written from a technical perspective, we provide more information to
help you understand and implement the Component Integrator and Web services
that are used to integrate with external system and services.

For step-by-step instructions on how to implement these features for our case
study, refer back to Chapter 6, “Implementing business processes: Case study”
on page 127.

© Copyright IBM Corp. 2008. All rights reserved. 257

7.1 Component Integrator
Component Integrator enables the Process Engine to connect with external
systems. With Component Integrator, you can use custom Java codes or existing
JMS components, and make them available in a workflow.

JMS provides basic Enterprise Information System (EIS) integration. For JMS
information, see:
http://java.sun.com/products/jms

Component Integrator provides extensible Java and JMS framework to enable

IBM FileNet Business Process Manager to interact with any system. In addition,
there is a component, CE_Operations, that allows access to Content Engine
objects and properties from a workflow.

How it works
In Figure 7-1, a work item represents a workflow step waiting in the Process
Engine for processing. The work item makes a request of a component. The
component provides a response by performing some actions that might be
outside of the Process Engine. The response is used in the subsequent steps on
the workflow. You do not have to know the details of the component itself, it can
be some custom Java code that connects to a legacy system, database server,
or mail server. The advantage of Component Integrator is that it hides the
complexity of the outside component.

1. Request (2+2?)
Component

3. Response (4)
2. Action (log “2+2=4”)

Work
Item

Process Engine

Figure 7-1 How Component Integrator works

258 Introducing IBM FileNet Business Process Manager

 Component Integrator use case examples
In a typical scenario where Component Integrator might be used, a company has
an existing legacy system and wants to leverage the functions provided by the
system within a business process.

Following are some examples of how the Component Integrator can be used in
an IBM FileNet Business Process Manager application:
򐂰 A company has existing Java codes that query proprietary systems for
claimant policy information. Adjustors need this claimant policy information to
adjust the claim amount in an IBM FileNet Business Process Manager
workflow process. The Java codes can be made available through
Component Integrator in the workflow step.
򐂰 A company has a Java application that stores and retrieves customer profile
data from a legacy data store. In an IBM FileNet Business Process Manager
workflow process, there is a requirement to retrieve and update customer
profiles. Component Integrator can be used to utilize the existing Java codes
for this IBM FileNet Business Process Manager workflow process.
򐂰 A company has a supplier that accepts order requests as XML documents
through guaranteed messaging service using message queues, and wants to
automatically generate order requests as part of its business process.
򐂰 An accountant uses existing JMS queue to send payment data. This data is
used (also considered as being consumed) by an electronic payment system
to process and issue check to customers.
򐂰 A customer has existing Java code that update sales order records in a
proprietary sales order system, and wants to integrate these functions in an
IBM FileNet Business Process Manager business process.

7.1.1 Content Engine operations

To provide content management operations in an object store, IBM FileNet uses
the CE_Operations component to support Content Engine operations in workflow
component steps.

The CE_Operations component provides the following functions:

򐂰 File a specific document to a given folder in the object store.
򐂰 Unfile a specific document from a given folder in the object store.
򐂰 Get the property value of a given document or folder.
򐂰 Set the property value of a given document or folder.

Chapter 7. Integration with external systems and services with BPM 259
 See Figure 7-2. which shows some of the CE_Operations you can select within a
workflow process.

Figure 7-2 CE_Operations

Additional operations are offered through other components such as

More_Operations and BPF_Operations.

The More_Operations component offers a rich set of functions (see Figure 7-3),
some of which include:
򐂰 Create, delete, copy, and move documents, folders, or custom objects.
򐂰 Get and set multiple property values of a given document or folder.
򐂰 Check in and check out of a document.
򐂰 Search for a document or documents using search templates.
򐂰 Send e-mails with attachments, and optionally using templates.

Figure 7-3 More_Operations

260 Introducing IBM FileNet Business Process Manager

7.2 Web services
Web services provide universal inter-operability between heterogeneous
applications using Web standards.

The Process Engine uses Web services technology to enable interaction

between IBM FileNet Business Process Manager processes, as well as between
IBM FileNet Business Process Manager processes and other business
processes.

There are three ways that IBM FileNet Business Process Manager uses the Web
services technology:
򐂰 Invoke Web services from a BPM workflow process. This enables you to use
the Web services developed for other applications in your organization and
the ones that are available externally on the Internet.
򐂰 Provide a BPM workflow process as a Web service and enable other
applications to use this BPM workflow process to accomplish their tasks.
򐂰 Use the Web service API to write step processors (not covered in this book).

There are two message type options when invoking or creating a Web service:
򐂰 Parameters: This is an easy way to use Web services. You do not need to
understand XML or XML schemas.
򐂰 XML: This is for working with complex parameters that the Parameters
message type cannot handle. To use it, you need to know XML and XML
schema knowledge.

Web services standards

In order for you to understand how to use XML Web services within IBM FileNet
P8, you should understand some basic definition of XML Web services.

There is a set of protocols that define how a Web service can be published,
discovered, and used in a technology neutral, standard format. Among these
standards are:
򐂰 Web Services Description Language (WSDL): Interface definition used to
define Web services. A WSDL definition describes how to access a Web
service and what operations it performs. You can use WSDL to create Web
services and publish these services to UDDI.
򐂰 Universal Description, Discovery and Integration (UDDI): The directory that
provides the means for discovering and publishing Web services. It is
provided by a third party that makes these services available for inquiry. It is
the equivalent of “Yellow and White Pages” of Web services.

Chapter 7. Integration with external systems and services with BPM 261
 򐂰 Extensible Markup Language (XML): The base meta-language for many
Service Oriented Architecture related technologies and standards. In Web
services, the messages including embedded data are written in XML format.
It can be used to exchange data, so that data can go back and forth between
the services provider and the client in the standard recognizable format. You
can consider it as the payload (in relation to HTTP).
򐂰 Hypertext Transfer Protocol (HTTP): The vehicle that carries the XML data
between systems. You can consider this as the transport.

Web services use case examples

Figure 7-4 illustrates a basic example of a Web services use case with three
steps.

In the typical scenario, Company A wants to offer a service to other companies.

Company A publishes the service as a Web service in a public UDDI repository
(denoted as step 1). The UDDI registry entry includes the WSDL, which
describes the service and how to use it.

Company B requires a particular service. Company B searches the public UDDI

repository (denoted as step 2) and finds Company A's Web services offering,
which meets its needs. Company B obtains the business contact information and
WSDL from the UDDI registry entry, and uses (consumes) Company A's Web
service (denoted as step 3).

An example of a service that Company A offers to other companies is credit

score reporting (FICO Score). Company A generates revenue by offering this
service to other companies through Web services. Company B, on the other
hand, might be a financial institution that provide loans to customers. In order for
Company B to process customer loans, company B uses (consumes) Company
A’s credit score reporting Web service to help determine whether to approve or
reject the loan.

262 Introducing IBM FileNet Business Process Manager

 UDDI Repository 1

WSDL Publishes Web Services

Company A
(Web Services provider)
2
e
r vic
Se
th e
es
3 Us 3

Company B
(Web Services consumer)

Figure 7-4 Providing and using Web services by companies

7.2.1 Process orchestration

Process orchestration describes the dynamic execution of a business process
from a technical view1. In the IBM FileNet Business Process Manager
environment, it describes the ability for processes to use and provide (consume
and be consumed by) Web services. An executable process can use other
processes to accomplish its work. It can also provide itself as a Web service for
other processes to use. Process orchestration involves combining disparate
components into joined-up process flows that automates end-to-end business
functions.

Process orchestration is a key BPM market requirement because it:

򐂰 Enables end-to-end processes.
򐂰 Enables collaborative business process.
򐂰 Enables Business to Business (B2B) transactions.
򐂰 Enables Web services composition.
򐂰 Complements Web services.

1
Definition is extracted from http://www.serviceoriented.org/process_orchestration.html

Chapter 7. Integration with external systems and services with BPM 263
 IBM FileNet Business Process Manager processes can use (consume) existing
Web services and provide (be consumed as) Web services. The IBM FileNet
Business Process Manager processes can be published as Web services, which
then can be discovered through a UDDI registry.

IBM FileNet Business Process Manager offers the following Web services
system functions to achieve Process orchestration:
򐂰 Invoke: Requests a service from a selected partner link.
򐂰 Receive: Provides a service in response to a Web service request.
򐂰 Reply: Sends a response to a Web service associated with a previously
accepted receive system function.

Invoke function
You can use the Invoke system function anywhere in a workflow process. The
Invoke system function calls an application component that provides a Web
service. This enables a workflow process to interact with and use external Web
services.

Invoke is implemented as an operation on a Web services queue called

WSRequest. When the IBM FileNet Business Process Manager system
encounters an Invoke system function at a workflow step, the work item is routed
to the WSRequest queue.

Invoke example: XYZ Corporation Authorize Credit Web services

Figure 7-5 illustrates an example of using the Invoke system function. The
workflow process shown in Figure 7-5 uses an external Web services provided
by Visa.com to authorize credit card transaction.

In the workflow process, there is a workflow step called Authorize Credit Card. At
this step, an Invoke system function calls a Web service from a Web services
provider, Visa.com. The Web service provides a credit authorization service.
Authorization request information is sent to the Web services. The Web service
returns a response, either Valid or Not Valid, back to the service requester (the
Authorize Credit Card workflow step) in the workflow process. The returned
response is evaluated at the next step of the workflow process and then routed to
the appropriate steps to accept or deny the credit card transaction.

In this example, Visa.com is the Web services provider and XYZ Corporation’s
workflow process is the Web services consumer.

264 Introducing IBM FileNet Business Process Manager

 Visa.com
Web Service Provider
Auth. Credit
Web Service

Web Service Consumer

Figure 7-5 Web services

Receive/Reply function
Handling of inbound requests from external Web services is done using the
Receive and Reply system functions.

The Receive system function can be viewed as representing an entry point to the
workflow process. Incoming messages are processed by a listener called
WS-Listener, which is running as part of Component Integrator. If you put a
Receive step immediately after a Launch step in a workflow, an appropriate
incoming message received by the system can launch a workflow process.

The Reply system function provides the ability to pass back information to the
application that made the original request. There might be multiple Reply steps in
a workflow process for one Receive step. The relationship between Receive and
Reply is usually one to many.

Similar to the Invoke system function, the Reply system function is implemented
as an operation on the WSRequest queue. When the Process Engine encounters
a Reply system function in a workflow, the work item is routed to the WSRequest
queue. The WS-Adapter, running under the Component Integrator, queries the
queue and processes the work item.

Unlike the Invoke system function, there is no response received back from the
call.

Chapter 7. Integration with external systems and services with BPM 265
 Receive/Reply example: Publish XYZ Corporation statistics
An example of using Receive and Reply system functions is in the insurance
claims processing workflow that provides statistics lookup service to the
executives of the company.

The executives require high level statistics on insurance claim totals and
processing time displayed in their home page.

The IT department creates a workflow that collects current statistics from the
claims processing database on demand. The workflow contains a Receive
system step that gets the request for high level statistics. When the request is
received, it launches the workflow process. A DB Lookup system step is called
within the process that retrieves the current statistics from the Claims database,
and summarizes them to the high level that is required by the executives. The
workflow also contains a Reply system step that sends back the high level
statistics information to the requester.

The IT department publishes the workflow to a UDDI registry. It also modifies the
executives’ home page template to include the WSDL that associated with the
Web services. The executives can use the link provided in their home page to get
the high level statistics they need.

Figure 7-6 illustrates the example. Notice that, in this case, the workflow process
is the Web services provider, and the executive home page is the Web services
consumer.

266 Introducing IBM FileNet Business Process Manager

 Executive home page

Web Service Consumer

Web Service Provider

XYZ claims
Statistics DB

Figure 7-6 Web services Receive and Reply system functions usage example

7.3 Rules connectivity framework

The Rules Connectivity Framework (RCF) allows you to implement dynamic
business rules for workflows. These rules can be simple or complex and can
perform a variety of actions, such as performing if-and-then tests, setting values
in data fields, or sending a work item to a submap or exception map. When you
change the business rules, the changes are immediately applied to the existing
workflow process.

Note: IBM FileNet Business Process Manager by itself does not provide a
business rules engine. It integrates with the third-party rules engine software
of your choice in its system.

Workflow process designer and business analysts can create and add business
rules to individual steps of a workflow. Rules are used to separate business rules
from the process, making it easier for a business analyst to adjust a complex
process by modifying the rules rather than modifying a workflow process.

Chapter 7. Integration with external systems and services with BPM 267
 Rules can control the following types of operations in a workflow:
򐂰 Assign data to a field in the work item.
򐂰 Send a work item to a workflow map.
򐂰 Send a work item to a custom exception workflow map.
򐂰 Skip a workflow step.
򐂰 Repeat a workflow step.

To implement rules in a workflow process, the workflow designer and the

business analyst work together to determine how rules should be used in the
workflow, what business decisions should be controlled by the third-party rules
engine, what workflow data are required for the rules engine, the appropriate
names for the rule sets, and the steps in a workflow where the rules are
executed. Each step in a workflow can have multiple rules, and the rules can be
executed at different stages of the step, before execution, on save, or after
completion.

How it works
When a workflow encounters a rule, the Process Engine sends a request to the
Rules Listener (an interface provided by IBM FileNet), which passes the request
to the rules engine. The rules engine executes the rule-set and returns the
results to the Process Engine.

In IBM FileNet Business Process Manager environment, if the Rules Listener is

stopped when a workflow tries to run a rule, the Process Engine sends the work
item to a special queue called Instruction Sheet Interpreter (ISI). When the Rules
Listener is restarted, the ISI will resubmit the work item to the Rules Listener.

If the rules engine interface is disabled in Process Configuration Console, the

workflow will run and ignore any rules implemented in the workflow.

Asynchronous rules processing

Although the rules engine can execute most rule-sets in less than a second,
some rule-sets, such as database lookups, might require longer processing time.
When a Process Engine process (VWKs) is waiting for a response from the rules
engine, it is not available to process any other incoming requests. To improve
system performance, use asynchronous rules processing (executing rules in the
background). You can configure specific rule-sets to be processed this way.

268 Introducing IBM FileNet Business Process Manager

Overview of implementing rules
Business rules are assigned to workflows using Process Designer. To implement
rules in your IBM FileNet Business Process Manager system, the following steps
are required:
1. Install a Rules Listener and the third-party rules software on a server.
2. Enable rules in the Process Engine (through Process Configuration Console.)
3. Configure the connection between the Process Engine and the Rules Listener
(through Process Task Manager).
4. Start and stop the Rules Listener (through Process Task Manager).

It is beyond the scope of this book to cover detailed steps of setting up a rules
engine and using the rules. However, in the following two chapters, we provide
more details on understanding and implementing Component Integrator.

Chapter 7. Integration with external systems and services with BPM 269
270 Introducing IBM FileNet Business Process Manager
 8

Chapter 8. Implementing Component

Integrator and Web services
Previously, in Chapter 7, “Integration with external systems and services with
BPM” on page 257, we introduced, from a high level business perspective, how
IBM FileNet Business Process Manager integrates with external system and
services.

In the current chapter, from a technical perspective, we provide information

about understanding and implementing various integration options.

We cover the following topics:

򐂰 Understanding Component Integrator
򐂰 Implementing Component Integrator
򐂰 Understanding Web services
򐂰 Implementing a process to invoke a Web service
򐂰 Implementing a process as a Web service

For step-by-step instructions about how to implement Component Integrator and

Web services for our case study, refer back to Chapter 6, “Implementing
business processes: Case study” on page 127.

© Copyright IBM Corp. 2008. All rights reserved. 271

8.1 Understanding Component Integrator
The Component Integrator is an IBM FileNet Business Process Manager service
that connects a workflow with an external system. The Component Integrator
provides an easy way to use external components such as Java classes in a
workflow. You can use the components without programming. It is installed as
part of a typical Application Engine server installation.

The Component Integrator consists of the following components:

򐂰 Component adaptors in the Application Engine server
򐂰 Component queues in the Process Engine server
򐂰 Component Manager in the Application Engine server

Component adaptors are used to interact with different types of components from
a workflow step. You can use the Java adaptor to call Java components, and the
JMS adaptor to interact with message queues. You can also write your own
adaptors for different components and applications.

Component queues are created to service different types of component adaptors.

When a workflow process is executed, the Component Manager (Component
Integrator Service Manager) retrieves the request from the component queues
and invokes the components via the component adapters.

For each step in the workflow, the general sequence of events takes place as
shown in Figure 8-1. The diagram illustrates the runtime interaction of the
Component Integrator with Application Engine services (such as Component
Manager), component queues in the Process Engine server, and a custom
entity. The numbers in the diagram represent the steps in the operation of the
Component Integrator.

Figure 8-1 shows the following flow:

򐂰 Information is sent to the component (via the adapter).
򐂰 The component performs its work and interacts with the custom entity.
򐂰 The result of the work is saved in the step (in the workflow).
򐂰 The step (in the workflow) is completed.

272 Introducing IBM FileNet Business Process Manager

 Figure 8-1 Component Integrator runtime interaction example

8.1.1 Adapter and security

In Component Integrator operations, access to the installed components is
controlled through Java Authentication and Authorization Service (JAAS). It
allows fine-grained security control over who is running code, as opposed to
where the code comes from or who signed it.

JAAS can be used for two purposes:

򐂰 Authentication: To authenticate users to reliably and securely determine
who is currently executing the Java code, regardless of whether the code is
running as an application, an applet, a bean, or a servlet.
򐂰 Authorization: To ensure that users have the access control rights
(permissions) required to do the actions performed.

JAAS is used for the Process Engine and component authentication of users. It
supports a single sign-on capability.

For information about JAAS information, see:

http://java.sun.com/products/jaas

Chapter 8. Implementing Component Integrator and Web services 273

8.2 Implementing Component Integrator
The main steps to implement Component Integrator can be summarized as
follows:
1. Create and configure a component queue.
2. Import component queue operations.
3. Configure Component Manager.
4. Use components in a workflow.

Various tools and applications are used to implement Component Integrator:

򐂰 Process Configuration Console: This is used to register the components.
From the tool, you import service definitions for the component. You must
create a component queue for each imported class and make the component
queue operations available to use during a workflow design.
򐂰 Process Designer: This is used to define component steps in a workflow.
When defining a component step, you select the component, the component
operation to perform at the step, and provide workflow fields as parameters
for the operation.
򐂰 Process Task Manager: This is used to start, stop, and administer
Component Managers. Your component must be registered with a
Component Manager. This can be verified under the Required Libraries tab
for the Component Manager.
򐂰 Component Manager: This polls the component queues and retrieves work
items that are dispatched to them. Component Manager passes a given work
item to the adaptor that is assigned to the work item's queue.

There is also a configuration option to allow the system to automatically check for
new events. This option causes the Component Manager to respond to items in
a component queue as they arrive. This event-driven mechanism can be enabled
in addition to polling for work items.

If your IBM FileNet Business Process Manager system has been configured to
automatically check for new events, the Component Manager responds to both
the component's polling rate and to new events as they occur.

8.2.1 Creating and configuring a component queue

Using Process Configuration Console, you can create and configure a
component queue with a Java adaptor or a JMS adaptor. Use Add Component
Queue Wizard page to associate Java adaptor with a Java application, and JMS
queue with a message queue.

274 Introducing IBM FileNet Business Process Manager

 Note:

To launch Process Configuration Console from Workplace XT:

1. Launch WorkplaceXT:
http://<Application Engine server name>:9080/WorkplaceXT
2. Log in.
3. Select Tools → Administration → Process Configuration Console.

To launch Process Configuration Console from Workplace:

1. Launch Workplace:
http://<Application Engine server name>:9080/Workplace
2. Log in.
3. Select Admin → Process Configuration Console.

The user you used to log into the Workplace must have administrative rights
on your Process Engine server.

To create a new component queue (for example, More_Operations) in the

Process Configuration Console tool, follow these steps:
1. Launch the Process Configuration Console tool.
2. Connect the isolated region (right-click the isolated region and select
Connect from the context menu).
3. If there is no XML file provided with the new component, right-click
Component Queues and select New from the context menu. See Figure 8-2.
If an XML file is already provided, you can right-click Component Queues
and select Import from XML file. Select the XML file, import it to the system.
Afterwards, right-click the new component (for example, More_Operations)
and select Properties. Then jump to step number 5 on page 277. The XML
file might contain descriptions of the methods available in the Java class.
This helps you to select the right methods to use later on in your workflow.
Otherwise, you would have to either guess what they do, or manually add the
descriptions for each method.

Chapter 8. Implementing Component Integrator and Web services 275

 Figure 8-2 Create a new component queue

4. The Add Component Queue Wizard starts. Enter the following information
and click Next:
– Queue Name: The name of the component queue. This name will appear
in the list of component queues in Process Designer. For our example, we
enter More_Operations.
– Description: You can optionally enter a brief description of the queue. For
More_Operations, we leave it blank. See Figure 8-3.

Figure 8-3 Set component queue name and description

276 Introducing IBM FileNet Business Process Manager

 f. Select Java Component or JMS Component from the Adaptor
drop-down list, and click Configure.
5. Configure the component:
– For Java component, set information for the following fields and click OK:
• Jar File Path: Locate and select the Jar file containing your Java
component. For our More_Operations example, we select:
c:\temp\more_operations\More_Operations.jar

Important: To complete this step, the Jar file must be accessible

from your local computer. At this point, you import only the class
name and methods. For runtime, the Jar file must be located in the
Application Engine server where the Component Manager runs. Use
the Process Task Manager to specify the exact location of the Jar
file and to start the component.

• Java Class: Select the appropriate class from the drop-down list of the
available classes. After you select the Java class, a list of available
methods display in the Available Methods box. When you configure the
component queue operations, you have the opportunity to select one
or more of these methods. For our More_Operations example, we
select:
com/IBM FileNet/churnett/ciMore_Operations.class
Figure 8-4 shows the Jar File Path and Java Class set up for our example.

Figure 8-4 Configure Java component: Jar File Path and Java Class

Chapter 8. Implementing Component Integrator and Web services 277

 – For the JMS component, set information for the following fields and click
OK:
• Queue Connection Factory (use JNDI)
• Queue Name (JNDI)
Java Naming and Directory Interface (JNDI) is a standard extension to the
Java platform. It provides Java applications with a unified interface to
multiple naming and directory services in the enterprise.
6. Set the Adaptor Properties:
– Concurrent Threads: Number of concurrent threads to execute for the
component. For More_Operations, we use the default value 1.
– Polling Rate: The amount of time in milliseconds between polls of the
queue by this component to look for work. For More_Operations, we use
the default value 1000.
– Exception Submap: Exception handling submap used by the component.
For More_Operations, we select the default, Malfunction.
– Startup Mode: Automatic or manual startup. For automatic start, the
adaptor starts when the Component Manager starts. For
More_Operations, we select the default, Automatic.
7. Set the JAAS Credentials:
– User Name and Password: User name and password used for
identification and permissions on both the Process Engine and potentially
any external systems that will be accessed. For More_Operations, we use
the administrator user ID and password.
– Configuration Context: The context you enter must be defined in the
taskman.login.config file (found in …\IBM FileNet\Router). This is the
JAAS login configuration file that the CE_Operations component uses by
default. It contains various login context sections that components use for
authentication. For More_Operations, we enter CELogin.
See Figure 8-5, which shows these selections.

278 Introducing IBM FileNet Business Process Manager

 Figure 8-5 Java Component set up using the Add Component Queue Wizard

8. Click Finish.

8.2.2 Importing component queue operations

After you create a new component queue, you must import operations from the
selected Java class or specify JMS events.

To do so, follow these steps:

1. From the Process Configuration Console, right-click the new component
queue you just created and select Properties from the context menu.
2. Select the Operations tab and click the Import icon. See Figure 8-6.

Chapter 8. Implementing Component Integrator and Web services 279

 Figure 8-6 Import operations for Java component

3. Select either the Java methods or the JMS events to import, depending on
the adaptor type:
– For the Java component, select one or more methods from the Available
Methods drop-down list and click OK. The methods you select will be
converted into operations. The method name becomes the operation
name, and the method parameters become the operation parameters.
In the Process Designer, you can then specify the operation and its
parameters when designing the component steps of your workflow.
You can optionally rename parameters and enter descriptions for
operations and parameters. These can be helpful to the workflow designer
when specifying expressions for the parameters.

Important: If the imported methods depend on other Jar files, then you
should make them available to the Process Configuration Console
through the Register Additional Classes interface. See Figure 8-2 on
page 276.

Otherwise, the Process Configuration Console will fail to load the class
and thus does not display any methods.

These Jar files should also be added to the Required Libraries of

Component Manager as described in 8.2.3, “Configuring Component
Manager” on page 283.

280 Introducing IBM FileNet Business Process Manager

 See Figure 8-7, which shows these selections.

Figure 8-7 Select methods from Java class to import for the Java component queue

Note: You cannot import Java methods that use byte, char data types,
or any Java objects other than Short, Integer, Long, Float, Double or
Boolean. You can use methods that use VWAttachment and
VWParticipant.

– For JMS operations, enter the events for this component queue. For each
event, you enter an event name, and then specify the parameters for that
event. All parameters that you specify have read-only access. You can
optionally enter a description for each event and for each parameter.
The events will be converted into operations. In the Process Designer, you
specify the expressions for parameters in the component step of your
workflow.
The JMS adaptor only supports outgoing data to a JMS queue.

Important: Component Queue operations cannot be deleted later. After you

commit the component queue definition, and it is included in a transferred
workflow, you cannot delete the operations.

In addition, you must commit your changes and restart Component Manager
before the new queue operations or changes take effect. Otherwise, the
changes will be lost when you stop or refresh the Component Manager.

Chapter 8. Implementing Component Integrator and Web services 281

 4. After importing the operations, the Operations tab should look similar to
Figure 8-8. Click OK.
Before you proceed, you might want to add a description for each operation
so that later, when using them in a workflow design, it is easier to see which
ones you should use. If you imported the component with an XML file, the
description might already have been provided.

Figure 8-8 Component properties showing imported operations

5. Commit the changes by right-clicking the isolated region and selecting

Commit Changes from the context menu. See Figure 8-9.

Figure 8-9 Commit changes for the component queue definition

282 Introducing IBM FileNet Business Process Manager

8.2.3 Configuring Component Manager
In order to use the component queue that you created, you must configure the
Component Manager to use the component queue.

The Process Task Manager in the Application Engine enables you to configure,
start, and stop a Component Manager and its associated components. You can
run one or more Component Managers. Each Component Manager coordinates
one or more components which are responsible for delivering events from the
Process Engine to an external entity such as a Web service or a messaging
system. In the Process Task Manager, the Component Managers folder displays
all the Component Managers currently defined. Selecting a Component Manager
displays its associated components.

To configure a Component Manager to use the new component queue that you
created, perform the following steps:
1. Start the Process Task Manager from the Application Engine.
2. Stop the Component Manager where you will add the component queue. You
cannot update its information while it is running.
3. In the General tab, append the Queues field to include the new component
queue name. Remember to add a comma in between value and the new
name. There should be no space between them. For our More_Operations
example, we append the queue name, More_Operations, at the end. See
Figure 8-10.

Figure 8-10 Add new component queue using Component Manager

Chapter 8. Implementing Component Integrator and Web services 283

 Tip: To see what component queues are available, click Show queue
names on the right side of the Queues field. It shows a list of the
component queue names. Make sure you append the exact name at the
end of the Queues field.

4. In the Required LIbraries tab, specify the location of the Jar file that contains
the class associated with the component queue. The Process Task Manager
appends the Jar file location to the class path when starting the Component
Manager. You must add a required library entry for each component queue
that you create. Click Add to add the path. See Figure 8-11.

Note: The Jar file must exist on your Application Engine server.

Figure 8-11 Add Jar file path (for the component queue) in the Required Libraries tab

For our example, we placed the More_Operations.jar file in the Application

Engine, C:\P8Components directory. So, we add the following path in the
Required Libraries:
C:\P8Components\More_Operations.jar

284 Introducing IBM FileNet Business Process Manager

 5. Restart the Component Manager by right-click it and select Start from the
context menu to make the changes effective (see Figure 8-12).
The Component Managers node allows a user to start and stop each
Component Manager separately, as well as the individual component that has
been defined. Starting the Component Manager also starts any components
that have been configured to start automatically via the Process Configuration
Console.

Figure 8-12 Start the Component Manager

8.2.4 Using components in a workflow

To use the component operations, add a component step in your workflow using
the Process Designer. A component step represents an activity in the workflow
that is assigned to one or more operations in a component queue.

Follow these steps to add component operations in a workflow:

1. Open your workflow definition.
2. Add the Component step from the palette below to your workflow map.
3. Select the new Component step from the map.
4. From the right side panel, in the Operations section, click the Add icon.
5. Select the component from the drop-down box. For our example, we select
More_Operations.
6. A list of operations available for this component appears on the Operations
section. Select one or more operations from this list. For our case, we select
copy and move.
7. Select the operation one at a time to set its parameters. When you select an
operation, the associated parameters appear on the bottom of the panel with
their names, data types, and access rights.

Chapter 8. Implementing Component Integrator and Web services 285

 An icon next to the parameter type indicates what access right the parameter
has. Possible options include:
– Read: Reads from a work item.
– Write: Writes to a data field in the work item.
– Read/Write: Passes a value from the work item to the operation and then
optionally updates the value when the operation is over.
The Expression field contains a named expression for your use in specifying
the value for each parameter.
You can also specify the incoming and outgoing routing properties for the
component step, similar to other step types.

Note: For step-by-step instructions, we describe how to add a component

step in our case study; refer to 6.5.1, “Setting up a component step” on
page 169.

8.3 Understanding Web services

Web services provide universal inter-operability between heterogeneous
applications using Web standards. The Process Engine uses Web services
technology to enable interaction between IBM FileNet Business Process
Manager processes, as well as between IBM FileNet Business Process Manager
processes and other business processes and applications.

IBM FileNet Business Process Manager processes can use (consume) the
existing Web services. In addition, the processes can be published as Web
services, to provide services to (be consumed by) others. The published Web
services can be discovered through a UDDI registry. Using Process
Configuration Console, you can configure a limited UDDI registry list for an
isolated region or allow users to browse for published Web services.

8.3.1 Web services adaptor

Component Integrator handles the interaction between the Process Engine and
Web services (see Figure 8-13). This communication is transparent to both the
workflow administrator and users. The architecture supports both polling and
event driven modes for the Component Manager. You can configure the
Component Manager to automatically respond to new events as they occur or
poll for work.

286 Introducing IBM FileNet Business Process Manager

 WS-
WS-Adapter Outbound
PE Server Component
Integrator WS-
WS-Listener
Inbound

WSRequest WSPending
Queue Table

Figure 8-13 Web services system architecture

As shown in Figure 8-13, the outbound requests for Web services are handled
through WS-Adapter, running as part of the Component Integrator. The inbound
requests for Web services are handled by WS-Listener, running as part of the
Component Integrator. The Process Engine has a WSRequest queue and a
WSPending table, which are system-managed data structures used in
processing the requests for Web services.

All communications are accomplished through XML. When using Web services,
you have to set data field mapping. There is no requirement for you to
understand XML. The Process Designer tool automatically parses the WSDL for
input and output parameters. The data fields are updated with the values of the
incoming parameters at runtime.

Invoking Web services

To invoke a Web service in your workflow, you can create an Invoke step in the
workflow. The Invoke step calls the Invoke system function, which allows you to
call any application component that has been exposed as a Web service. This
enables your workflow to interact with external Web services.

The Invoke function is implemented as an operation on the WSRequest queue.

When the system encounters the Invoke call, the work item is routed to the
WSRequest queue. The WS-Adapter, running under the control of Component
Integrator, queries the queue and processes the work item. The work item is
placed in the WSPending Table waiting for coming responses from the Web
service. WS-Listener listens to any inbound request.

Chapter 8. Implementing Component Integrator and Web services 287

 Replying and receiving Web services
You can publish a process as a Web service to be used by other processes and
applications. To handle inbound requests for your published Web services, use
the Receive and Reply steps in your workflow.

The Receive step in a workflow calls the Receive system function. The Receive
function can be viewed as representing an entry point to the workflow process.
Incoming messages are processed by WS-Listener, which is running as part of
Component Integrator. When processing incoming messages, the WS-Listener
interacts with both the WSRequest queue and the WSPending table. If you put a
Receive step immediately after a Launch Step in a workflow, an appropriate
incoming message received by the system can launch a workflow process.

The Reply system function provides the ability to pass back information to the
application that made the original request. There might be multiple Reply steps in
a workflow process for one Receive step. The relationship between Receive and
Reply is usually one to many.

The Reply step in a workflow calls the Reply system function. Similar to the
Invoke function, the Reply function is implemented as an operation on the
WSRequest queue. When the Process Engine encounters a Reply system
function in a workflow, the work item is routed to the WSRequest queue. The
WS-Adapter, running under the Component Integrator, queries the queue and
processes the work item.

Unlike the Invoke function, there is no response received back from the call.

8.4 Implementing a process to invoke a Web service

The main steps to invoke a Web service can be summarized as follows:
1. Configure the UDDI registry list.
2. Define the Web service information in workflow properties.
3. Use an Invoke step in a workflow.

Note: For the step-by-step instructions that we used for our case study
(excluding the configuration of the UDDI registry list), refer to 6.9, “Creating
and calling a Web service” on page 230. There, we show you how to create a
workflow process as a Web service, call the Web service in a workflow
process, and test the process with Web services.

288 Introducing IBM FileNet Business Process Manager

8.4.1 Configuring the UDDI registry list
Use the Process Configuration Console to configure the UDDI registry list for an
isolated region. This registry list includes entries for Web services. You can
search through this list to find the Web service to use or publish a workflow as a
Web service in this registry.

Select or clear the check box option to allow or deny access from the Process
Designer to any arbitrary WSDL. This controls access to Web services.

To configure the UDDI registry list:

1. Launch the Process Configuration Console tool.
2. Connect the isolated region (right-click the isolated region and select
Connect from the context menu). See Figure 8-14.

Figure 8-14 Connect an isolated region in the Process Configuration Console

3. Right-click the isolated region and select Properties from the context menu.
4. In the Isolated Region Properties window, select the UDDI tab and then select
the UDDI Registry List tab.
5. Click the Add icon to add an entry to the list. See Figure 8-15.

Chapter 8. Implementing Component Integrator and Web services 289

 Figure 8-15 Add the UDDI registry list

6. Enter information for the new UDDI Registry entry:

– Name: The name that identifies this UDDI registry entry. It will be
displayed in the Process Designer when using the Web services Explorer.
– Inquiry URL: The URL for inquiries to the registry. To validate a URL,
select its row in the table, and click the Validate icon. This test confirms
that an Inquiry URL is valid (can connect).
– Publish check box: Select this option to enable publishing services to the
UDDI registry specified by the Publish URL. The Publish URL, Business
Name, Business Description, Business key, and User fields are optionally
used when publishing a service.
– Enable Process Designer to enter WSDL links without browsing for Web
services: Select or clear the check box option. When this option is
enabled, you can type in an arbitrary WSDL URL in the workflow
properties Partner Links in the Process Designer. When this option is not
enabled, you are restricted to search for WSDLs in the list of UDDI
registries specified for this isolated region.
7. Click OK to close the Isolated Region Properties window.
8. Commit the changes by right-clicking the isolated region and select Commit
Changes from the context menu.

290 Introducing IBM FileNet Business Process Manager

8.4.2 Defining Web services information in workflow properties
In the Process Designer, you must define the workflow properties for the Web
services that you want to invoke.

If a workflow invokes or receives requests for Web services, specify the Web
services on the Partner Links tab. If XML data fields and XML schemas are
required for invoke, receive, or reply tasks, define them in the XML Data Fields
tab and the XML Schemas tab. On the Web services General tab, you can
specify a folder in your object store or library where any incoming attachments
are stored.

To enable your workflow to invoke a Web service:

1. Start the Process Designer, open the workflow definition.
2. Open the Workflow Properties window.
3. Select the Web services tab and then select the Partner Links tab.
4. Select the Invoke check box and click the UDDI Explorer icon. See
Figure 8-16.

Figure 8-16 Set Invoke WSDL URL

Chapter 8. Implementing Component Integrator and Web services 291

 5. Select UDDI registries (see Figure 8-17). Open the desired registry in the left
pane. This displays any UDDI registries previously defined using the Process
Configuration Console. Locate the Web service you want to invoke in the
UDDI registries. The Web service can be either in public or private UDDI
registries, or in workflows in the current isolated region that contain a Receive
step. You can find the WSDL in the UDDI registries.
You can perform a keyword search or simply browse through the configured
registry entries. To perform a keyword search, select either Business list or
WSDL tModel list, then enter the first few letters of the business name or
WSDL tModel name and click Execute to display the matching items. If a
large number of items are returned, you can narrow the search by entering
the whole name, or by using the Case sensitive and Exact match options.

Figure 8-17 Select a Web service from UDDI registries

6. Select the Web services you want to invoke.

7. Click OK to extract the WSDL information to the Partner Links table.

292 Introducing IBM FileNet Business Process Manager

8.4.3 Defining an Invoke step in a workflow
To define an Invoke step in a workflow map:
1. Open the workflow definition.
2. Place an Invoke step on the workflow map using one of the following
methods:
– Use a System step:
i. Add a System step to the workflow map.
ii. Select the System step icon from the workflow map.
iii. From the right panel under the Functions section, select Invoke. Use
the right-arrow key to add it to the Selected Functions field. See
Figure 8-18.

Figure 8-18 Use the System step with the Invoke function in workflow map

Chapter 8. Implementing Component Integrator and Web services 293

 – Use a special Invoke step:
i. Select Web services Palette from the View Palette drop-down box on
the bottom of the Process Designer window.
ii. Add the Invoke step to the workflow map. See Figure 8-19.

Figure 8-19 Use Invoke System step from Web services Palette

3. Specify the general properties for the Invoke system function. If you use the
System step, double-click Invoke from the Selected Functions field. If you
use the special Invoke step, select the step from the map and select the
General tab on the right. Enter the following information:
– Partner Link: Select the Partner Link for the Web service. The Partner Link
list contains only entries already specified for Invoke Web services in the
workflow properties.
– Operation: Select the appropriate operation in the Web service.

294 Introducing IBM FileNet Business Process Manager

 – Message Type: The Process Designer automatically selects the message
type. Two message types are:
• Parameters: This uses the workflow data fields and attachments for
input and output. This option is available for simple parameters.
• XML: This uses XML messages for input and output. This option is for
more complex messages.
If required, you can override the Parameters selection and select XML
instead, but not vice versa.
– Outgoing Parameters: If you select the Parameter message type, and the
Web service that you invoke requires input, then the required fields are
displayed under this field. In the Expression field, specify a previously
defined data field or expression of the appropriate data type.
– Incoming Parameters: If you select the Parameter message type, and the
Web service will return data, then the name and data type are displayed
under this field. In the Field Name field, select a previously defined data
field of the appropriate type to hold the returned data. See Figure 8-20.

Figure 8-20 General tab for the Invoke step

4. Select the Advanced tab. In this tab, you can optionally set the time out
expression for the step and the time out map that would be executed if a time
out does occur. In addition, you can select the Use reliable messaging option
here.
5. Make the appropriate routing connections between this Invoke step and other
steps and save the workflow definition.

Chapter 8. Implementing Component Integrator and Web services 295

8.5 Implementing a process as a Web service
Before you turn a workflow process over to a Web service, make sure that the
workflow works correctly as it is.

The main steps to implement a workflow process as a Web service can be

summarized as follows:
1. Define the Web service information in workflow properties.
2. Use the Receive/Reply steps in a workflow.
3. Finalize the Web service operations and transfer the workflow definition.
4. Publish the workflow process as a Web service.

The workflow administrator uses Process Configuration Console to publish the

Web service to the UDDI Registry List for the isolated region. Then external
consumers can locate the Web service in the registry and call it.

Note: For the step-by-step instructions we used for our case study, refer to
6.9, “Creating and calling a Web service” on page 230 in which we show you
how to create a workflow process as a Web service, calling the Web service in
a workflow process, and testing the process with Web services.

8.5.1 Defining Web services information in workflow properties

In the Process Designer, define the workflow properties for the Web service.
Specify the Partner Link and port definition. The Partner Link information is used
when defining the Receive and Reply steps in a workflow.

To do so:
1. Start the Process Designer and open the workflow definition.
2. Open the Workflow Properties window.
3. Select the Data Fields tab and enter the data fields that will be used to hold
the reply messages.
4. Select the Web services tab and then select the Partner Links tab.
5. Select the Receive/Reply check box (see Figure 8-21). Enter the following
information:
– Name: Enter the name for the Web service on the left panel.
– Process Port Type: Enter a name for the port.
– Process Role

296 Introducing IBM FileNet Business Process Manager

 Figure 8-21 Workflow properties for Web services Receive/Reply

8.5.2 Using the Receive/Reply steps in a workflow

In order for a workflow process to act as a Web service, the workflow must
contain at least one Receive step that accepts requests and optionally a Reply
step that returns messages to the consumers. We highly recommend testing the
workflow to make sure that it works before turning it over to a Web service.

To use a Receive step in a workflow:

1. In the Process Designer, start the existing workflow.

Important: A workflow that is intended to be published as a Web service

must not contain any blanks in the workflow name.

1. Open the workflow definition.

2. Place a Reply step on the workflow map using one of the following methods:
– Use a System step:
i. Add a System step to the workflow map.
ii. Select the System step icon from the workflow map.

Chapter 8. Implementing Component Integrator and Web services 297

 iii. From the right panel under the Functions section, select Receive. Use
the right-arrow key to add it to the Selected Functions field. See
Figure 8-22.

Figure 8-22 Use the System step that calls the Receive function

– Use a special Receive step:

i. Select Web services Palette from the View Palette drop-down box.
ii. Add the Receive step to the workflow map. See Figure 8-23.

298 Introducing IBM FileNet Business Process Manager

Figure 8-23 Use Receive step from Web services Palette

Important: The Receive step must be the first step after the Launch step in
order to launch this workflow automatically in response to an invoke of the
Web service.

3. Specify the general properties for the Receive function. If you use the System
step, double-click Receive from the Selected Functions field. If you use the
special Receive step, select the step from the map and select the General
tab. Enter the following information:
– Partner Link: Select the Partner Link for the Web service. The Partner Link
list contains only entries already specified for Invoke Web services in the
workflow properties.
– Operation: Select the appropriate operation in the Web service.
– Message Type: The Process Designer automatically selects the message
type. Two message types are:
• Parameters: Uses the workflow data fields and attachments for input
and output. This option is available for simple parameters.

Chapter 8. Implementing Component Integrator and Web services 299

 • XML: Uses XML messages for input and output. This option is for more
complex messages.
If required, you can override the Parameters selection and select XML
instead, but not vice versa.
– Outgoing Parameters: Specify outgoing parameter information.

Note: Because this workflow automatically launches when the Web

service is invoked, do not assign F_Originator to any step in the workflow.
Otherwise, it would not be considered a valid user.

4. Select the Advanced tab. In this tab, you can optionally set the time out
expression for the step and the time out map that would be executed if a time
out does occur. In addition, you can select whether to authenticate user or not
and restrict the users (who can invoke this Web service) to specific users or
groups.
5. Make the appropriate routing connections between this Receive step and
other steps.
6. Optionally, add a Reply System step if required, to reply to the Web service
consumer with its information.
7. Save the workflow definition.

8.5.3 Finalizing Web services operations and transferring the

workflow definition
After you finish defining your workflow as a Web service, you must finalize the
Web service operations in the workflow and then transfer the workflow definition
to the Process Engine. Once you transfer a finalized workflow to the Process
Engine, you no longer can make changes in the input and output parameters of
the Web service operations.

To finalize the Web service operations and transfer the workflow definition:
1. In the Process Designer, open the workflow properties of the workflow.
2. Select the Web services tab and then select the General tab.
3. Select the Finalize existing web services operations option. See
Figure 8-24. The system prompts you with a message regarding the
implications of this action.

300 Introducing IBM FileNet Business Process Manager

 Figure 8-24 Finalize existing Web services operations

4. Click Close.
5. Save the workflow.
6. Transfer the workflow by select File → Transfer.

The Partner Link of the finalized Web service does not identify a specific version
of the workflow. Other workflow processes always invoke the latest version of the
transferred workflow.

Note: After you transfer the workflow definition to the Process Engine with the
finalized option enabled, you cannot disable this option in the workflow
definition.

8.5.4 Publishing the workflow process as a Web service

If a workflow process is to be consumed (used) as a Web service, you must
deploy the workflow process as a WSDL definition and publish it to a UDDI
registry. You can publish the workflow to either your private UDDI registry or in a
public UDDI registry.

Before you can publish a workflow to a UDDI registry, you must specify the
appropriate UDDI registry in the isolated region properties in Configuration
Console.

After creating the workflow definition and transferring it into the Process Engine,
use the Process Configuration Console to publish the workflow to a UDDI
registry.

Chapter 8. Implementing Component Integrator and Web services 301

 To publish a workflow process as a Web service:
1. In the Process Configuration Console, select the isolated region that contains
the workflow to be published. The workflow must contain the Receive system
function, and it must already have been transferred to the Process Engine.
2. Right-click the isolated region and select Publish to UDDI from the context
menu.
3. In the Publish to UDDI window, select the Publish Workflow tab.
4. Enter a workflow definition name in the Search Web services workflows box,
or leave the field blank and search for all workflows in the current isolated
region that contains a Receive system function.
5. Click the Search icon to display the list of workflow definitions. Select the
workflow to publish.
6. Select a previously specified UDDI registry. The list contains only those
registries from which a valid certificate can be obtained.
7. Click Publish.

If you are publishing a business entity for the first time, enter the business name
and description, and provide a user name and password for the UDDI registry.
Otherwise, you cannot change the business name or user name. The information
that you enter here displays in the UDDI Registry List for the isolated region.

302 Introducing IBM FileNet Business Process Manager

 9

Chapter 9. Planning and designing a BPM solution

In this chapter we cover the process of planning and designing a content centric
IBM FileNet Business Process Manager (BPM) application at a high level. This
includes gathering requirements, transforming them into a workflow process,
validating, and finally producing a high level architecture design.

We cover the following topics:

򐂰 Planning a BPM solution
򐂰 Functional design
򐂰 Detailed design

Note: The discussion we present here is intended for prototype application.

To implement a production system, much more has to be taken into
consideration. Contact your IBM FileNet lab services professional or
appropriate business partners for details.

© Copyright IBM Corp. 2008. All rights reserved. 303

9.1 Planning a BPM solution
Planning and designing an IBM FileNet Business Process Manager (BPM)
application is an iterative process. In this section we cover a walkthrough of the
planning process at a high level. The intention is not to go deep to make it a step
by step process. Rather, we want to provide you with an overall understanding of
information gathering techniques, and validating and transforming requirements
into a solution to a level where BPM process administrators would be able to use
it for designing a prototype model in Process Designer.

The core discipline of business process management is a continuous process

improvement. As the models that drive work or business transactions through the
company evolve constantly, there is a need to have an agile method to model the
process that IBM FileNet Business Process Manager offers.

Figure 9-1 shows the steps in modeling a process in an IBM Business Process
Manager system.

Figure 9-1 steps in modeling a process in a BPM system

304 Introducing IBM FileNet Business Process Manager

The inner circle of Figure 9-1 defines high level steps, and the outer circle
outlines the detailed steps or the resulting actions to achieve the high level steps:
򐂰 Discover:
– Action: Review
– Action: Understand
– Action: Evaluate
򐂰 Design:
– Action: Process flow
– Action: Integration
– Action: Metrics
– Action: Controls
򐂰 Deploy:
– Action: User Interface
򐂰 Monitor:
– Action: Analyze
– Action: Optimize
– Action: Simulate

Most outer circle actions require multiple iterations. Requirements gathering,

designing, building, and testing all require multiple iterations to create and fine
tune a process. The analyzing and optimizing phases involve several additional
iterations as business analysts and managers experiment with alternations.

There are five major areas in the outer circle to look for when designing a BPM
process: process flow, integration, user interface, roles and responsibilities, and
controls.

Process flow
This action entails creating process definitions and flow. In the initial iteration,
focus on the core functionality that delivers the bulk of the value.

Integration
This action focuses on integrating with and extracting information from internal
and external systems, getting the required results, and updating databases.

User interface
After you understand the roles and responsibilities of all users, the next action is
to ensure that the user interface windows deliver the required information and
provide the right interface for users to interact with the system to perform their
tasks.

Chapter 9. Planning and designing a BPM solution 305

 Roles and responsibility
This action focuses on exploring the management information that is deemed
necessary. Helpful questions to ask include: How the data is gathered, who
should have access to the data, and how the data is presented.

Controls
Business managers want ways to improve performance of the process, allowing
them to better control process execution. They require mechanisms that help
them cater to peak system performance on demand, or influence the way in
which business rules apply. This action focuses on providing the controls to the
business managers.

The reiterated steps in modeling a business process ensures that no surprises

emerge along the way, while delivering the flexibility to change as needed.

9.1.1 Discovering and understanding the process

Every organization has a different starting point with different requirements.
Some organizations already have a defined process, while others are not as well
developed. Some want to emphasize automation of the process, whereas others
require better traceability, visibility, and performance measurement. Either way,
the first objective is to understand the process. There are at least four
possibilities for a business process:
򐂰 It is a brand new process.
򐂰 It is an ad-hoc workflow process.
򐂰 It is currently a paper-based manual process.
򐂰 It is an existing IBM FileNet Image Services (IS) eProcess.

The complexity of the discovery of a process varies based on the existing

environment. You have to capture the process as it is. Assessing the existing
process by using a set of complementary modeling techniques allows you to
understand the process better. Ideal techniques for this phase include:
򐂰 Flow diagrams or flow charts to look at the order of activities
򐂰 Role activity diagrams, also called business context diagram, to focus
attention on role interactions and the desired behavior of various actors

The emphasis here is on understanding the process, not building models for
transformation into code or executable process definitions. This enables both the
business analyst and the business users to step outside of the business-as-usual
view and see the existing process differently.

306 Introducing IBM FileNet Business Process Manager

 The process of gathering required information about a business problem is
termed a walkthrough. Before you start this process, we recommend that you
prepare yourself and your audience for it.

9.1.2 Preparing for a walkthrough

As a good practice, we recommend preparing the audience for an interview or a
walkthrough.

Before starting the requirement gathering process, conduct the following

preparation steps:
򐂰 Prepare a list of topics and questions.
򐂰 Prepare the participants.
򐂰 Provide information about the process and products.
򐂰 Provide questions that require advance research.
򐂰 Plan for the expected amount of time and required degree of participation.

After you and your audience are ready, look for information in the following
specific areas of the business:
򐂰 Organization and business objectives.
򐂰 Industry and comparable initiatives.
򐂰 Within current processes:
– Departmental, inter-departmental, and external flows
– Roles
– Inefficiencies and associated costs
򐂰 Perceived gaps to desired business processes:
– Functional requirements
– Non-functional requirements
򐂰 Opportunities for improvement that have not been identified, such as:
– Application integration
– External process participants
– Interactive process monitoring

9.1.3 Requirements gathering

In this phase, start gathering information by asking the related questions. The
questioning technique differs in a great deal with the type and source of
information. There are four methods of information gathering:
򐂰 Job shadowing

Chapter 9. Planning and designing a BPM solution 307

 򐂰 Interviews
򐂰 Focus groups
򐂰 Questionnaires

Each of the information-gathering techniques work best in a particular scenario.

Pick and choose the technique that matches the target audience and the desired
results. You might use a combination of techniques. Typically, use job
shadowing and interviews with heads-down knowledge workers; use focus
groups with supervisors and managers; and use questionnaires with occasional
or external users.

Job shadowing
The job shadowing is primarily use for observing the current business process of
workers below the supervisor level and observing corporate culture.

You can use job shadowing when the as-is state is important input to the
requirements. Sometimes the requirements are better observed than described.
Another reason for job shadowing is that the participants might have insufficient
skills to communicate their requirements or current process, or are unwilling to
communicate such information completely, or are intimidated in an interview
environment.

While performing job shadowing, encourage demonstration of all business

scenarios. Pay attention to the processes that are time sensitive, for example,
specific to a certain time of the day, week, month, or quarter. Observe the
manual handling of paper or electronic files. Check out the multiple ways and
user interfaces used for the existing process. Also ask for spreadsheets, e-mail,
paper logs, and phone calls that are part of the process.

Interviews
Use one-on-one interviews when a few individuals have specific unique
knowledge or the requirements are better described than demonstrated, for
example, when job shadowing is not appropriate.

You can use an interview primarily for management and executives, and for
understanding high-level business objectives and long-term vision, the big
picture.

Focus group
Use focus groups when knowledge is distributed over a number of individuals.
The interaction between the participants is relevant to determine the
requirements. The participants are peers (or nearly so) to reduce power-based
domination.

308 Introducing IBM FileNet Business Process Manager

 You can use a focus group primarily for supervisors and senior workers who
work closely and cooperatively on a daily basis. You can also use a focus group
to reach consensus on shared requirements.

Questionnaires
Use questionnaires when there are many participants to survey or they are
geographically remote. You can also use questionnaires for external constituents
who might be inappropriate for internal meetings. Answers to specific
well-defined questions, such as how many and how often, can also be surveyed
in the form of questionnaires.

9.1.4 Interviews and questionnaires

When creating a list of questions, ask about how the current business processes
are working and where they should be performed.

For managers and knowledge workers, it is often easiest to start with questions
about the department, such as:
򐂰 What are the existing processes, including any process diagrams or
procedures that they have created?
򐂰 How do they interface with other departments?
򐂰 How do they interface with external third parties systems, services, and
customers?
򐂰 What are the existing and upcoming regulatory and customer service
standards that they must meet?

As interview and questionnaires progress, you can move on to more specifics

about their work, for example:
򐂰 What are the systems that they currently work with each day?
򐂰 What are the things that they find most difficult about the current processes,
and what do they think can be changed to make it better?
򐂰 Are there any new reporting requirements that need to be implemented?
򐂰 What are the main software applications that they use to complete their work?
򐂰 What other applications do they make use of throughout the course of the
day?
򐂰 Do they switch between applications frequently? If so, why? Do they re-key
information from one system to another? If so, what is the information and
why?

Chapter 9. Planning and designing a BPM solution 309

 The end result of this process is to eventually compute a return on investment
(ROI). The more information you gathered on both as-is and to-be states of the
business processes, the easier it is to create a solid ROI statement. Whatever
you ask, always determine the areas of current inefficiencies and the associated
costs so the new or improved process can be better.

9.1.5 XYZ Corporation as-is auto claim approval process

By using the foregoing information gathering techniques, we have been able to
determine the following facts about our case study of the XYZ Corporation auto
claim process.

XYZ Corporation offers a range of financial and insurance products and services
through a large network of field offices.

The key corporate business drivers include:

򐂰 Streamline™ business process efficiencies.
򐂰 Prove and maintain adherence to government regulations.
򐂰 Improve customer satisfaction.
򐂰 Increase competitive advantage.

Strategic objectives for the corporation include:

򐂰 Improve business processing.
򐂰 Grow business with less staff and leverage existing systems.
򐂰 Reduce cycle time for claims processing.
򐂰 Improve customer service.
򐂰 Launch self-service customer projects.
򐂰 Improve accounting.
򐂰 Better risk management.
򐂰 Control and manage e-mails.

From Information Technology (IT) perspectives, objectives include:

򐂰 Establishing new standards.
򐂰 Reducing IT expenditure through standardization and consolidation.

After gathering the existing process requirements and business initiatives of XYZ
Corporation, we create a flowchart describing the current as-is auto claim
approval process as shown in Figure 9-2.

310 Introducing IBM FileNet Business Process Manager

 START
Field Agent reviews
the claim, checks for
errors and completion Adjustor reviews the
Call receives from
policy holder claim package for
financial controls

Field Agent queries the Central filing

Customer Services policy underwriting receives and stores
Rep (CSR) receives system for claimant the claim package
the call and takes notes policy information Decison for seven years

CSR creates a word Field Agent prints out

the relevant document Accounting Clerk
document using call If the claim amount is If the claim amount is
that are needed to sends the claim
notes from policy holder greater than $3000, $3000 or less,
process the claim package along with
Adjustor sends it to Adjustor sends it for
a copy of the check
Supervisor payment
to central filing

CSR faxes the Field Agent determines

word document to the Adjustor who Supervisor receives Accounting Clerk Accounting Clerk
the head office would handle the claim the package and gets the approved sends a check to
based on the amount reviews it claim package the Claimant
and type of the claim,
Mailroom Clerk receives such as residential
the claim via fax policy, $3000 or under

Supervisor runs risk

Mailroom Clerk manually Field Agent faxes the assessment using
date/time stamps the document to Adjustor background history
claim package and notifies Adjustor
about the claim

Mailroom Clerk
determines which Field Decision
Agent should work on Adjustor reviews the END
the claim depending on claim and waits for
the location of accident police report and car
photos
Supervisor rejects Supervisor
the claim approves the claim
Mailroom Clerk manually
classifies the claim Field Agent receives
and sends the police
report and photos to the Supervisor writes
head office Mailroom Clerk gets the
a rejection letter
Mailroom Clerk sends letter and dispatches it
and sends it to the
the claim to Field Agent to claimant
mailroom
via internal office mail
Mailroom Clerk receives
the package, stamps it,
classifies and forwards
to Adjustor

Figure 9-2 Flowchart describing XYZ Corporation as-is auto claim approval process

Chapter 9. Planning and designing a BPM solution 311

 After reviewing the process, we discover the following pain points associated
with the current as-is auto claim process:
򐂰 It is manual and time intensive.
򐂰 There is a lot of rework due to errors.
򐂰 Many hours of overtime are required to correct errors.
򐂰 Documents are often not signed as required.
򐂰 Claimants have to wait a long time to receive checks.
򐂰 The file system is manual; not only is it time consuming to use, but also
difficult to find the information in it later.

9.2 Functional design

In this section, we discuss analyzing the requirements and transforming the
requirements into an IBM FileNet Business Process Manager functional design.
We show that using IBM FileNet Business Process Manager improves the
overall process from both tangible and intangible perspectives.

9.2.1 Transforming requirements into a solution

After you have gathered detailed requirements, identify and analyze the key
areas where IBM FileNet Business Process Manager can be used to provide
greater benefits.

Identifying the key requirements

First, make sure that the business process is important for business, and that the
improvement of this process is critical to long-term success of the company or
organization. Next, ensure that there is an easily identifiable role for IBM FileNet
Business Process Manager as part of the process improvement.

Analyzing the requirements

Analyzing the requirements determine how a manual process can be redesigned
to improve processing time. As a result, some part of the process can be
automated and some part might be completely eliminated.

In our case study of auto claim process, we can eliminate the mail-room
operation. This alone saves labor hours thus improving processing time and
reducing the claim processing errors and overall costs.

We also look to automate certain areas for improvements. This includes

identifying and assigning field agents and adjustors based upon accident
location, type of a claim, and the claim amount involved.

312 Introducing IBM FileNet Business Process Manager

 Identifying a key business process
Identify a key business process that can be improved through IBM FileNet
Business Process Manager.

Map the current as-is and future (or want) to-be business processes. Analyze the
future business processes and find out more information about these business
processes. This includes identifying the following information:
򐂰 The functions and roles within the process
򐂰 The source of content and its method of acquisition within the process
򐂰 Steps that can be automated
򐂰 Integration point within the process

Overlay the IBM FileNet Business Process Manager to check for the right fit, and
identify the potential gaps in functionality between the as-is and to-be business
process.

The graphical view of the current requirements is presented in a flowchart in

Figure 9-3 on page 314. We also create a business context diagram, Figure 9-4
on page 316, to show user roles and interaction within the system.

9.2.2 XYZ Corporation future (to-be) auto claim approval process

In 9.1.5, “XYZ Corporation as-is auto claim approval process” on page 310, we
discuss XYZ Corporation’s as-is process with a flowchart. In this section, based
on our analysis of the process and XYZ Corporation requirements, we transform
the as-is process into a future to-be process shown in Figure 9-3.

The purpose of having a future to-be process is to ensure that we understand the
process in its entirety, including the roles and responsibilities of each player in
the process. This way, we can get an agreement on the process steps and on
the more low level requirements.

Chapter 9. Planning and designing a BPM solution 313

 Supporting
documents
START preparation by
Field Agent
(Human task)

CSR interfaces with

policy holder and files
the claim form NO
(Human task) Documents
complete?

YES
Claim case is
created
(System step) Final review by Payment
Adjustor processing
(Human task) (Human task)

Field Agent is
assigned
(System step)
Goes for second approval Approve if claim amount is
if claim amount is more less than or equal to
than $3000 $3000
Claim approves
Claim amount?
(System step)
Field Agent reviews the
case, completes
END
information, determines
the claim amount YES
(Human task)
Escalate for any
exception Is the claim
(System step) approved?
Adjustor is
assigned NO
(System step)
Supervisor reviews
(Human task)

Adjustor receives and Claim rejection

reviews the claim letter generated
(Human task) (System step)

Figure 9-3 XYZ Corporation to-be process auto claim approval process

For auto claims, the comparison of the as-is process and the to-be process
clearly shows a considerable reduction from 30 steps to 13 steps. This is done by
eliminating mail-room operation and automating manual processes such as field
agent and adjustor assignments.

314 Introducing IBM FileNet Business Process Manager

The auto claim process depicted in Figure 9-3 on page 314 has four roles:
򐂰 Customer Service Representative (CSR)
This is an individual, customer facing role. CSRs take customers’ calls and
fill out eForms with claim information such as customer name, policy number,
type of incident, and location of incident. After a CSR submits an eForm, a
process is automatically initiated in the IBM FileNet Business Process
Manager system.
򐂰 Field agent
After the system launches a claim process, it finds the appropriate field agent
and adjustor depending upon the location of accident, the type of claim, and
the value of claim. The field agent then completes any missing information,
including gathering supported documents such as police report and claim
estimates, and routes the claim to an adjustor.
򐂰 Adjustor
The adjustor receives the claim package from the inbox, reviews the claim,
checks and validates the policy document, and takes one of three actions:
approve the claim, reject the claim, or escalate it to supervisor due to certain
exceptions. If the adjustor approves the claim, the claim goes for payment
and an approval letter is sent to the claimant. If the adjustor rejects the claim,
a rejection letter is sent to the claimant. In the case of escalation, the claim
along with all supporting documents moves to the supervisor’s inbox for
further processing.
򐂰 Supervisor
Supervisor performs thoroughly reviews the claimant folder for supporting
documents, any previous claims. If required, the supervisor runs a fraud
investigation and risk assessment.

After identifying the business process and the responsible roles, we create a
business context diagram that explains users’ roles, connectivity, and
information flow for the auto claim process. See Figure 9-4.

Chapter 9. Planning and designing a BPM solution 315

 3. Mailroom Clerk receives the
claim via fax , timestamps the claim
package, searches for policy
document, gets the name of Field
2. CSR takes Agent based on location of loss,
information, and and sends the claim package to
sends to head Field Agent via US mail
office via fax

1.Customer calls
and submits a
claim
Front Office
CSRs

7. Accounting
Clerk gets the
approved claim
8. Filing Clerk
package and writes
gets the closed
a check and sends
file and files it .
it to the claimant.
The closed file is
then sent for filing.

Back Office
Mailroom Clerks
Adjustors
Supervisors
6. Supervisor reviewes 4. Field Agent gets the
the claim and either Accounting Clerks
claim package, reviews
approves or rejects it. Filing Clerks it , associates police
Approved claims go to report and other related
Accounting Department documents, then sends it
for payment to Adjustor

Field Agents
5. Adjustor receives the claim
package, reviews it, and either
approves or rejects the claim or
escalates it to Supervisor .
Approved claims go to
Accounting Department for
payment

Figure 9-4 Business context diagram with roles and connectivity for auto claim process

Using both the to-be process and the business context diagram, you can clarify
your understanding of the process and gain your customer’s commitment on the
requirements.

316 Introducing IBM FileNet Business Process Manager

9.2.3 Locking down the requirements
Locking down the requirements serves two purposes. The first purpose is to gain
agreement from customers, including business users, IT staff, and the
management. The second purpose is to have a base process to start creating a
prototype solution. The best practice is to have a formal review with all partners
to achieve a common understanding and consensus.

Address the following points when locking down the requirements:

򐂰 Base requirements
򐂰 Associated costs
򐂰 Process improvement requirements
򐂰 Discussion of the solution scenario
򐂰 Discussion of the stages of implementation
򐂰 High level overview of IBM FileNet Business Process Manager products that
address the requirements
򐂰 Benefit summary

9.3 Detailed design

There might be gaps between the default functions provided by the IBM FileNet
Business Process Manager system as-is and the customer requirements.
Identify the alternatives. If the requirements are nearly all met, and the gaps are
not significant, you might need minor customization. If the gaps are significant,
either a third-party solution or add-on products can be used to close the gaps.

The alternatives will become part of the detailed solution design, wherein you
describe the readily available functions as-is and any of the added functions from
customization.

For the purpose of our case study, we use two interfaces. The eForms interface
is used by CSRs who take customers’ calls and prepare claims. The eForms
helps in filling the required information interactively while the customer is on the
phone. After the process is initiated, a case is created. Other users, such as field
agent, adjustor, and supervisor, require a review of the case. Therefore they use
a customized user interface that provides more functions, such as reviewing the
claimant folder, and associating more documents with the case, such as police
reports, creating comments, and approving or rejecting the claims.

Chapter 9. Planning and designing a BPM solution 317

 Note: For the actual implementation of the case study in Chapter 6,
“Implementing business processes: Case study” on page 127, in order to
simplify the explanation, we do not show an eForms implementation. We use
Workplace (XT) and the generic step processor provided with IBM FileNet
Business Process Manager to complete the entire implementation.

9.3.1 Solution components

After you have identified the key requirements and the business process, the
next step is to list the requirements and map it with IBM FileNet Business
Process Manager features in order to identify any gaps.

In our case study, we map the key business requirements with the features
provided by IBM FileNet Business Process Manager and other IBM FileNet P8
products. The result is shown in Table 9-1.

Table 9-1 Mapping of key business requirements with readily available P8 features
Business requirements IBM FileNet P8 features

Replace manual steps of creating the Use eForms to fully automate this
claim, such as creating and faxing word process.
documents, time-stamping, classifying,
and reviewing for input errors for the
claim.

Replace the manual process of locating Through the BPF customization, a case is
the policy document from a manual filing created automatically after the claim is
system, printing, and associating the submitted. The claim and all associated
policy with the claim package and re-filing. documents are part of the case folder.

Replace manual interaction between field Interaction through the IBM FileNet
agent and adjustor, which has usually Business Process Manager system, with
delayed the claim processing. built-in Time Delay, Exception, and
Escalation triggers.

Replace manual handling of associated The Content Engine used within IBM
documents that were causing delay in the FileNet Business Process Manager stores
processing, for example, police reports, these documents electronically.
car photos, and so on.

Improve risk assessment, which currently Intelligence is built into the IBM FileNet P8
is completely manual, using the previous process. Based upon the risk assessment
claims history. rules, the adjustor can make informed
decisions.

318 Introducing IBM FileNet Business Process Manager

 Business requirements IBM FileNet P8 features

Reduce claim approval’s overall IBM FileNet Business Process Manager’s

processing time. process has built in logic for claim
approval, rejection, and escalation
decisions based upon claim amount and
corporate rules.

Improve payment process, which takes up The claim case would have every
time, due to not having complete document that is needed for a final
information. payment.

Replace or improve the writing and This is completely automated in the

dispatching of letters to the claimant, process step.
which currently are manual processes.

Provide status about the process, Comprehensive auditing in BPF

including who did what, which currently is Business Activity Monitor.
hard to track down.

Properly retain and destroy claim files and IBM FileNet P8 compliance framework
associated documents, such as e-mails, can be proposed to address this
according to corporate compliance requirement.
policies, a feature that currently the
manual process does not offer.

We identified a number of manual processes that can be fully automated with

readily available IBM FileNet Business Process Manager functions. We identified
at least three major requirements such as case management, push method of
delivering the case to users, and comprehensive audit.

As we learned earlier, the business and strategic objectives of XYZ Corporation

are to gain a competitive edge, so we need to have a solution that can be
deployed in a short period of time. To address this and the other key
requirements such as case management, comprehensive audit, and using a
push method to deliver a case to adjustor’s queue, our plan is to use IBM FileNet
Business Process Framework (BPF) in conjunction with IBM FileNet Business
Process Manager.

While the BPF user interface is critical to claim processors, it is “overkill” for
CSR’s role. We decide to use a simple eForms-based user interface. A form can
be used as a document or can be customized as an intelligent user interface. For
our case study, we use the document type eForms.

Chapter 9. Planning and designing a BPM solution 319

 By automating the process steps in the IBM FileNet Business Process Manager
system, we address a number of XYZ Corporation business objectives along
with the strategic and the IT objectives such as improving customer satisfaction
by reducing overall process cycle time. Automating the process also reduces
labor time and errors, so more claims can now be handled correctly and speedily
by the same number of people, thus reducing the overall cost per claim.

Besides automating the claim process, XYZ Corporation also ask for monitoring
the claim process and using the results for further process improvements. This
can be addressed by implementing Process Analyzer, Process Simulator, and
Business Activity Monitor BAM.

As XYZ Corporation is looking to implement IBM FileNet Business Process

Manager as an enterprise solution, they prefer to ensure that the system can
scale, expand, and be able to handle future growth.

Compliance is another critical requirement that includes storing, discovering, and

retaining claim files, e-mails, and other related documents for a specific period of
time, and being able to disposed them according to corporate governance and
regulatory requirements.

IBM FileNet Business Process Manager can use various high availability options,
including server farms for scalability and expendability; while IBM FileNet
Records Manager, Email Manager, and Record Crawler can be used in
conjunction to address the compliance requirements.

9.3.2 Modeling the process

In this phase, we design a high level solution that shows process steps, roles, the
user’s interactions with respective user interfaces, and process logic. All the
design is based around IBM FileNet Business Process Manager and the
additional components, in this case, eForms and BPF.

Finally we model the process in Visio using business process modeling notation,
BPMN. The final BPMN design is shown in Figure 9-5. It can be imported as a
BPM process design diagram into Process Designer (see Figure 9-7 on
page 322).

320 Introducing IBM FileNet Business Process Manager

 Field Agent
completes the case
CSR information and attaches Adjustor Supervisor
Creates a case supporting documents Reviews the case Reviews the case
(eForms ) (BPF) (BPF) (BPF)

Application Auto Claim Assign Field Agent

Engine Process Assign Adjustor

File documents LDAP users’ authentication

LDAP

Content
Engine
operation

Figure 9-5 XYZ Corporation BPM system context diagram

9.3.3 Technical architecture overview

The system context diagram shown in Figure 9-5 provides a view of the proposed
solution. For IT infrastructure, we use an architectural overview diagram so that
XYZ Corporation IT staff can see both the logical and physical fit within their IT
infrastructure.

Technical architecture is designed based upon the analysis of the requirements

and identifying a process that shows a prototype solution. Some key reasons for
creating a technical architecture are:
򐂰 To show how the solution fits within the existing IT infrastructure
򐂰 To map the solution features with the required functions
򐂰 To show connectivity to the internal and external systems
򐂰 To evaluate the delta between required and proposed specifications and its
impact on the development approach
򐂰 To use it as a prototype to present solution’s functionality

Figure 9-6 shows the created technical architecture of our auto claim process.
The auto claim BPM process design diagram is shown in Figure 9-7.

Chapter 9. Planning and designing a BPM solution 321

 BPF customized
interface used by Field
Agent, Adjustor, and
eForms interface Supervisor
used by CSR Web Server / eForms and BPF Process Engine Content Engine
Application Engine Application Server

Ethernet

DB2 Database CM OnDemand

Email Server Content Manager
Server for statements
system

Figure 9-6 High level architecture for XYZ Corporation prototype auto claim process

Creates case, validates Reject Claim

Claim Setup Generate letter
customer policy , and
Sub-process Sub-process
assigns adjustor
Application

Start End

Claim Settle Claim

processed Sub-process
ClaimID End
First Notice of Loss
Agent

Field Agent
Review
If <=3000 .00
Approved

Police Report Witness Statement

Adjustor Approved
Adjustor

Review
Reject

Escalate
If > 3000.00 For exception
Second approval
Supervisor

Supervisor
Review

Figure 9-7 Auto claim BPM process design diagram

322 Introducing IBM FileNet Business Process Manager

9.3.4 Cost analysis metrics
As we discussed earlier, business process management relates to change
management that results in improvements in several business areas. Some of
these areas include:
򐂰 Elimination of manual labor, such as a process step that automatically sends
an email transaction confirmation to a customer, which in turn replaces a
manual process.
For the auto claim process, we identified the following manual processes:
– Manual filing and re-filing
– Mail room handling
– Calls/email
򐂰 Capturing of process information for understanding, such as the automated
gathering of data showing when each item in a process is entered and exited
at each processing step.
򐂰 Improvement in analysis of information and decision-making, for example,
based on current data, calculate the average wait time for an item in each
process step, In the current auto claim process, it would be several work
hours.
򐂰 Coordination of processes across distances, making processes independent
of geography, for example, a data entry process step executed by operators
in home-based or remote offices, regardless of where the process initiated. In
other words, CSRs and adjustors are location independent.
򐂰 Coordination between tasks and processes, such as a process step that
automatically looks up client information in a line of business system based
on a client account number in a CRM system that triggered the process.
CSRs’ call time are greatly reduced.
򐂰 Capturing, storing, searching, and distributing of unstructured content by
providing direct online access to content such as claim cases and associated
documents in IBM FileNet P8 content repository.

There are two areas that directly impact ROI metrics:

򐂰 Cost reduction metrics
This is typically supportable with hard data and is generally well-understood
by all levels.
򐂰 Competitive advantage metrics
This requires revenue forecasting but might have a supportable model. It
might be understood only at executive and management levels.

Chapter 9. Planning and designing a BPM solution 323

 The cost reduction metrics provide tangible benefits in the following areas:
򐂰 Reduced error rates
򐂰 Increased process efficiency
򐂰 Reduced/consistent end-to-end service time (cost of compliance failure)
򐂰 Reduced time to implement process changes
򐂰 Increased process visibility (cost savings through self service)
򐂰 Reduced effort to measure process
򐂰 For document-driven processes, reduced manual distribution

The competitive metrics provide intangible benefits in the following areas:

򐂰 Improved time to market due to process improvement and greater agility
򐂰 Reduced/consistent end-to-end service time
򐂰 Increased process visibility (for example, the quality of customer service)
򐂰 Increased capacity
򐂰 Technology as a differentiated against competition
򐂰 Increased efficiency from handling customer inquiries on-line

9.3.5 Final walkthrough

In the final walkthrough, our objective is to gain agreements on what has been
done so far and proceed to the next step of creating a prototype solution,
followed by a customized demonstration and a high level return on investment
(ROI) for both hard and soft dollar savings.

The walkthrough discusses:

򐂰 Detailed solution design
򐂰 Configuration and process design
򐂰 Customizing and integration with other systems
򐂰 Monitoring and reporting
򐂰 Process management
򐂰 Customized product demonstration; using simulation and analysis, an ROI
can be demonstrated
򐂰 Return on investment on a business case

Implementing IBM FileNet Business Process Manager means introducing

change management. It is important to talk about it and point to any additional
savings due to changes in the process.

324 Introducing IBM FileNet Business Process Manager

 A

Appendix A. Additional material

This book refers to additional material that can be downloaded from the Internet
as described below.

Locating the Web material

The Web material associated with this book is available in softcopy on the
Internet from the IBM Redbooks Web server. Point your Web browser at:
ftp://www.redbooks.ibm.com/redbooks/SG247509

Alternatively, you can go to the IBM Redbooks Web site at:

ibm.com/redbooks

Select the Additional materials and open the directory that corresponds with
the IBM Redbooks form number, SG24-7509.

© Copyright IBM Corp. 2008. All rights reserved. 325

Using the Web material
The additional Web material that accompanies this book includes the following
file:
File name Description
SG247509.zip Contains object store data model and workflow definition
used for the case study.

System requirements for downloading the Web material

The following system configuration is recommended:
Hard disk space: 40 MB minimum
Operating System: Windows XP
Processor: Pentinum IV
Memory: 512 MB

How to use the Web material

Create a subdirectory (folder) on your workstation, and unzip the contents of the
Web material zip file into this folder. Follow the instruction in the README.txt file.

326 Introducing IBM FileNet Business Process Manager

Related publications

The publications listed in this section are considered particularly suitable for a
more detailed discussion of the topics covered in this book.

IBM Redbooks
For information about ordering these publications, see “How to get Redbooks” on
page 327. Note that some of the documents referenced here might be available
in softcopy only.
򐂰 IBM FileNet Content Manager Implementation Best Practices and
Recommendations, SG24-7547

Online resources
These Web sites are also relevant as further information sources:
򐂰 Product documentation for the IBM FileNet P8 Platform
http://www.ibm.com/support/docview.wss?rs=3247&uid=swg27010422
This is where you can download the IBM FileNet P8 Platform ecm_help
documentation (ecm_help.zip), FileNet P8 System Overview (in PDF), and
many other valuable product documentation.

How to get Redbooks

You can search for, view, or download Redbooks, Redpapers, Technotes, draft
publications and Additional materials, as well as order hardcopy Redbooks, at
this Web site:
ibm.com/redbooks

© Copyright IBM Corp. 2008. All rights reserved. 327

Help from IBM
IBM Support and downloads
ibm.com/support

IBM Global Services

ibm.com/services

328 Introducing IBM FileNet Business Process Manager

Index
active content 4
A definition 2
abdication 61
primary focus 2
active content 4
transactional and content centric processes 3
Adjustor Review 148, 151
business requirement 136
APIs
business rule 28, 34, 267, 269, 306
Process Engine 10, 12, 43
business rules engines 13
application
IBM FileNet Enterprise Manager 129
Process Administrator 43, 205–206, 210–211 C
Process Analyzer 6, 13–16, 44, 320 case 11
Process Configuration Console 43, 268–269, case management model 11
274 case study 127, 129, 132, 135–136, 310, 312
Process Designer 5–6, 9, 39, 43, 47, 128, 132, claim approval process 24, 26–28
158, 267, 269, 274, 276, 280–281, 304, 320 CE_Operations 258–259
Process Simulator 5–6, 13–14, 42, 44, 320 claim approval process 24, 26–28, 56, 310, 313
Process Task Manager 43 claim folder 169, 172–173, 176
Step Processor 43 claim process 212, 227, 311
Task Manager 283 Claim Setup submap
Tracker 43, 150, 205–206 step 177, 182
tracker 5 collector step 59, 99, 104
Application Engine (AE) 13, 36, 39 component adaptor 272
attachment 57, 139 Component Integrator 5, 8, 12, 39, 43, 65, 128,
attachments 258–259, 271–272
workflow definition 82 runtime interaction 272
automation use case examples 259
process 2 component integrator 12
Component Manager 272, 274, 277
component queue 274
B importing operations 279
BAM Dashboard 46
required library entry 284
Base Workflow 78–79
component queues 272, 276
BPF_Operations 260
Component step 8, 72–73, 108, 160, 169, 274,
BPM Palette 72, 90, 146–147
280–281
Business Activity Monitor 42
component step
Business Activity Monitor (BAM) 13–14, 16, 45, 320
incoming and outgoing routing properties 286
business process 3, 13, 20, 24, 26, 35–37, 53, 58,
conditional route 114, 128, 176
61, 127–128, 257, 259, 286, 304, 306–309
conditional routing 59
definition 2
content
escalation mechanism 60
active content 4
first step 110
definition 3
key elements 55
federated 19
Web services 49
content centric process 3
Business Process Framework (BPF) 11, 46
Content Engine (CE) 13, 21, 36, 44, 129, 258
business process management

© Copyright IBM Corp. 2008. All rights reserved. 329

 database 42 G
federates content 19 General step 8, 72–73, 91, 160
operations 259
workflow process 143
workflow process definition 142 I
IBM FileNet Business Process Framework (BPF)
Content Engine (CE) Kernel 41
10
Content Federation Services for Image Services
IBM FileNet Content Manager 19
(CFS-IS) 19
IBM FileNet Email Manager 21
Content Manager 19
IBM FileNet Enterprise Manager 129
IBM FileNet P8 Platform 36
D IBM FileNet Records Crawler 21
data field 59, 81, 136–137, 267, 286–287 IBM FileNet Records Manager 20
data tier 42 IBM FileNet System Manager 21
data type 11, 81, 136, 281, 285 IBM FileNet System Monitor 21
database lookups 28 IBM FileNet Workplace 10, 12
deadline 60 import
definition 9 XML file 275
delegation 61 incoming message 40, 288
Directory Services 42 inheritance
workflow 117
Instruction Sheet Interpreter (ISI) 268
E
eForms 10–11, 27 integration options
electronic form 27, 31, 38 process 5
Email Manager 21 interview 307–308
email notification Invoke 264
disable 81 Invoke step 293
engine invoke Web services 287
definition 36 isolated region 49–50, 79, 275, 282, 286, 289
enterprise content management (ECM) 10, 18 limited UDDI registry list 286
Enterprise Information System (EIS) 258 UDDI Registry List 296
escalation UDDI registry list 289, 302
business process 60
Event Dispatcher 44 J
event log 44, 49, 51, 80, 117, 136 Java Authentication and Authorization Service
Process Engine 15 (JAAS) 273
external system 29–30 Java class 272
Java Message Service (JMS) 39
Java Messaging System (JMS) 12
F
F_Trackers 57 Java Naming and Directory Interface (JNDI) 278
farm JMS 258
servers 50 job shadowing 308
federated content 19
field agent 2, 29, 33, 172–173, 314–315, 317–318 K
FileNet System Monitor (FSM) 21 Key Performance Indicators (KPIs) 16, 45
focus group 308
L
Launch step 8, 110, 161, 265, 288, 299

330 Introducing IBM FileNet Business Process Manager

Lotus Forms 11 active content 4
business process definition 2
case study 24, 26–28
M content centric 3
main workflow map 128, 184
deadline definition 9
manual process 2, 24, 26, 306, 312, 319
disable email notification 81
map
engine definition 36
definition 9
integration options 5
malfunction map 121
integration to external systems 3
workflow map 70–71, 91, 128, 142, 184, 242,
isolated region 79
268, 285, 293
map definition 9
message 40
map, malfunction map 121
message type 261, 295
map, workflow map 70–71, 91, 128, 142, 184,
Microsoft Visio 5–6
242, 268, 285, 293
middleware tier 41
modeling 304, 320
modeling
palette 72, 146–147
process 4
process voting 61
modeling process 304, 320
queue, work queue 164
monitoring 5
route definition 8
process 6
routing 58
More_Operations 169, 172, 260, 275–276
routing, conditional route 114, 128, 176
multiple routes 150, 152
routing, conditional routing 59
routing, multiple routes 150, 152
O routing, simple majority 62
object store 19, 42, 82, 87, 138–139, 259, 291 routing, single routing 58
attachment item 139 step definition 7, 58
attachment template 139 step, collector step 59, 99, 104
definition 36 step, Component step 8, 72–73, 108, 160, 169,
On-Line Analytical Processing (OLAP) 15 274, 280–281
operation parameter 170, 231 step, first step 110
operations step, General step 8, 72–73, 91, 160
import, component queue 279 step, Invoke step 293
optimization 5 step, Launch step 8, 110, 161, 265, 288, 299
process 3 step, Reply step 288, 296–297
step, reply step 49, 231, 245
step, Submap step 72, 84, 147, 160, 172, 185
P
palette 72 step, System step 72–73, 105, 181, 266, 293,
BPM Palette 72, 90, 146–147 297
parameter 161 subprocess 9
participant 55 subscription 192, 202
Partner link 87, 122, 233, 290, 292, 294, 296, 299, system function 73, 80, 105
301 Reply 87
Web services 291 task 55
police report 30, 56, 130, 173, 315, 317 timer definition 9
policy holder 24, 30, 55–56 transactional 3
portlet 10, 12 use case 56, 310, 313
presentation tier 39 Web services 255–256
process work queue 51
workflow definition 68, 79–80

Index 331
 workflow process 143 properties pane 71, 148, 152, 159
workflow process definition 142 select condition 235
workflow properties 71 Publisher 44
XML process definition language (XPDL) 5–6
Process Administrator 43, 205–206, 210–211
Process Analyzer 6, 13–16, 41, 44, 320
Q
questionnaire 309
database 42
queue
OLAP 42
component queue 274
process automation 2
component queue, import operations 279
process configuration 39
definition 51
Process Configuration Console 43, 268–269, 274
Instruction Sheet Interpreter 268
process definition 6, 9, 51, 305–306
work queue 164
definition 7
queues
Process Designer 5–6, 9, 39, 43, 47, 128, 132, 158,
component queues 272, 276
267, 269, 274, 276, 280–281, 304, 320
component queues 276
malfunction map 121 R
step palette 72 Receive system function 264–265
workflow properties 71 receive Web services 288
process designing 4 Records Crawler 21
Process Engine (PE) 4, 6, 13, 15, 21, 36, 43, 47, Records Manager 20
49–51, 129, 186, 258, 261, 272–273 Redbooks Web site 327
APIs 10, 12, 43 Contact us xiv
component 36 region
components 47 isolated region 49–50, 79, 275, 282, 286, 289
database 42, 44, 46, 50 Reply function 87
event log 15 Reply step 288, 296–297
logical view 49 reply step 49, 231, 245
server 50–51, 272, 275 Reply system function 264
system 41, 50 reply Web services 288
Process Engine (PE) Kernel 41 requirements gathering 305
process flow 305 Return on investment (ROI) 310, 323
process improvement cycle 14 role 55
process instance roster 80
definition 51 definition 51
process modeling 4 route 7
process monitoring 5–6 conditional route 114, 128, 176
process optimization 3, 5 definition 8
Process Orchestration 40 parallel 9
process orchestration 263–264 routes
process server 50 multiple routes 150, 152
Process Simulation Console 44 routing 28, 58
Process Simulation Designer 44 conditional routing 59
Process Simulator 5–6, 13–14, 42, 44, 320 simple majority 62
process subscription 191 single routing 58
Process Task Manager 43, 283 Rules connectivity framework (RCF) 267
Process Tracker 150, 205–206 rules engine 41, 267–268
process voting 61 rules engine framework 5
properties 71 Rules Framework 41

332 Introducing IBM FileNet Business Process Manager

Rules Listener 268 template
search template 260
timer
S definition 9
sample solution 55
tool
search template 138–139, 260
IBM FileNet Enterprise Manager 129
service level agreement (SLA) 16
Process Administrator 43, 205–206, 210–211
Service Orientated Architecture (SOA) 47
Process Analyzer 6, 13–16, 44, 320
Service Oriented Architecture 48
Process Configuration Console 43, 268–269,
simple majority 62
274
single routing 58
Process Designer 5–6, 9, 39, 43, 47, 128, 132,
single-sign-on (SSO) 42
158, 267, 269, 274, 276, 280–281, 304, 320
step 7
Process Simulator 5–6, 13–14, 42, 44, 320
collector step 59, 99, 104
Process Task Manager 43
Component step 8, 72–73, 108, 160, 169, 274,
Step Processor 43
280–281
Task Manager 283
definition 7, 58
Tracker 43, 150, 205–206
first step 110
tracker 5
General step 8, 72–73, 91, 160
Tracker 43, 150, 205–206
Invoke step 293
tracker 5
Launch step 8, 110, 161, 265, 288, 299
transactional process 3
Reply step 288, 296–297
reply step 49, 231, 245
Submap step 8, 72, 84, 147, 160, 172, 185 U
System step 8, 72–73, 105, 181, 266, 293, 297 UDDI registry 262, 264
Step Processor 43 UDDI registry entry 290
Submap step 8, 72, 84, 147, 160, 172, 185 UDDI registry list 286, 289
subprocess 9 Universal Description, Discovery and Integration
subscription 191–192, 202 (UDDI) 261–262
system function 73, 80, 181, 287–288, 294 use case 55
Reply function 87 claim approval process 56, 310, 313
system functions user interface 164, 305, 319
Invoke 264 user queue 51
Receive 264–265
Reply 264
System Manager 21
V
voting
System Monitor 21
process voting 61
system property 199, 201
symbolic name 202
system queue 51 W
System step 8, 72–73, 105, 181, 266, 293, 297 WaitForCondition 80
walkthrough 307, 324
Web Application Toolkit (WAT) 10, 12
T Web services 5, 8, 12–13, 28, 34, 40, 47–49, 65,
target object
72, 75, 128, 230–231, 240, 244–245, 251, 256,
class 188–189, 198
261–262, 271, 283, 286–287, 294, 299
class type 188
inbound requests 287
information 216
invoke 287
task 55
outbound requests 287
Task Manager 43, 283

Index 333
 Partner link 291, 294, 299, 301 multiple Reply steps 265
publishing 301 publishing as Web services 301
reply, receive 288 validate and check 231, 245
standards 261 Web services 142, 230, 244
use case examples 262 Web services step 249
workflow properties 291, 296 work item 137
Web services (WS) 255–256 workflow process definition 142
Web Services Description Language (WSDL) 261, workflow properties 57, 71, 77
287, 289 workflow property 128, 134, 288, 290
Web Site 325 data fields 231
WebSphere Business Modeler 5–6 Web services 294, 299
work item 51, 74, 80, 137, 152, 258, 264, 274, Web services information 288, 291
286–288 workflow group 206
data field 286 workflow subscription 128, 186
data field values 74 special expressions 195
work packet 28, 30, 61 Workplace 10, 12, 68
routing 28, 30 Workplace XT 10
work queue 51, 164 WS-Adapter 265, 287
workflow 2, 128–129, 142, 186, 258–259, 272, 274 WS-Listener 265, 287
Base Workflow 78–79 WSPending table 287
definition 2 WSRequest 265
process 189 WSRequest queue 287
process definition 6–7, 9
Process Designer 5–6, 9, 39, 43, 47, 128, 132,
158, 267, 269, 274, 276, 280–281, 304, 320
X
XML data field 291
Process Simulator 5–6, 13–14, 42, 44, 320
XML file 275, 282
step, Submap step 8
import 275
step, System step 8
XML message 295, 300
version 192
XML process definition language (XPDL) 5–6
work queue 51
XML schema 87, 117, 261, 291
workflow definition 68, 79–80, 143, 189, 207, 285,
XML Web services 261
291
XYZ Corporation
disable email notification 81
sample solution 55
Reply functions 87
system functions 105
workflow group 57, 78, 83, 140–141
definition 117
name 140
setup 206
workflow inheritance 117
workflow map 70–71, 91, 128, 142, 242, 268, 285,
293
main 128, 184
workflow name 79, 135, 232, 297
workflow process 121, 128, 132, 140, 212,
231–232, 241, 244, 259, 261, 272, 288, 303, 306
definition 142, 221
document title 144
map 128, 146, 157

334 Introducing IBM FileNet Business Process Manager

Introducing IBM FileNet Business Process Manager
(0.5” spine)
0.475”<->0.875”
250 <-> 459 pages
 Back cover ®

Introducing IBM FileNet

Business Process Manager
®

Business process This IBM Redbooks publication provides a basic introduction

design to IBM FileNet Business Process Manager (BPM) V4.0. BPM INTERNATIONAL
enables organizations to create, modify, and manage content TECHNICAL
Integration with centric business processes. One key advantage of BPM is its SUPPORT
external systems ability to work with active content, which refers to the ability ORGANIZATION
of content to trigger or affect business processes.
and services
In this book, we cover the key elements that make up a
Step-by-step business process, including tasks, participants, roles, steps, BUILDING TECHNICAL
instructions routing, and deadlines. We describe how to use Process INFORMATION BASED ON
Designer (a BPM application) to design your business PRACTICAL EXPERIENCE
processes. In addition, we provide step-by-step instructions
on how to implement a use case business process scenario.
IBM Redbooks are developed by
BPM also supports integration with external systems and the IBM International Technical
services through Component Integrator, Web services, and Support Organization. Experts
Rules Connectivity Framework. We discuss these integration from IBM, Customers and
options and provide instructions on how to implement the Partners from around the world
create timely technical
use case scenario by demonstrating some of the options. information based on realistic
This book is useful for system architects, process analysts, scenarios. Specific
recommendations are provided
and process designers who require an understanding of IBM to help you implement IT
FileNet Business Process Manager. It also serves as a solutions more effectively in
practical guide for those who want detailed instructions in your environment.
order to implement a BPM system.

For more information:

ibm.com/redbooks

SG24-7509-00 ISBN 0738431052