IBM Prescriptive Maintenance on Cloud IBM

User Guide
IBM Prescriptive Maintenance on Cloud IBM

User Guide
ii User Guide
Note
Before using this information and the product it supports, read the information in
“Notices” on page 43.

© Copyright IBM Corp. 2017 iii

iv User Guide
 First edition (March 2018)

© Copyright IBM Corp. 2017 v

vi User Guide
Contents
Chapter 1. Product overview . . . . . . 1 Resampling and data handling rules . . . . . 14
Welcome to Prescriptive Maintenance on Cloud . . 1 Configuring model settings . . . . . . . . 15
How it works . . . . . . . . . . . . . . 2 Analysis results . . . . . . . . . . . . . 16
Required data . . . . . . . . . . . . . . 2 Notification . . . . . . . . . . . . . . 18
Optional data . . . . . . . . . . . . . . 4 Resetting the product . . . . . . . . . . . 18
Outputs . . . . . . . . . . . . . . . . 4
Training models by using a subset of data . . . . 5 Chapter 3. Integrating with other
Accuracy of predictions for long-range planning . . 5 products . . . . . . . . . . . . . . 19
How is data backed up and restored? . . . . . . 5 Integration with Watson IoT Platform . . . . . 19
What's new . . . . . . . . . . . . . . . 6 Integration with Maximo Asset Management . . . 20
Accessibility features . . . . . . . . . . . 6
Chapter 4. Application programming
Chapter 2. Using the product . . . . . 9 interface . . . . . . . . . . . . . . 23
Logging on . . . . . . . . . . . . . . . 9
Post entity data from Maximo Asset Management 23
Downloading the tutorial and sample data . . . . 9
Post event data from Maximo Asset Management . 24
Getting started . . . . . . . . . . . . . 10
Post entity data in JSON format . . . . . . . 26
Using data files . . . . . . . . . . . . . 10
Post event data in JSON format . . . . . . . 27
Data from multiple sources . . . . . . . . 10
Upload a file . . . . . . . . . . . . . . 28
Entity property files . . . . . . . . . . 10
Download data . . . . . . . . . . . . . 29
Event files . . . . . . . . . . . . . . 11
Retrieve analysis results . . . . . . . . . . 31
Translation file . . . . . . . . . . . . 12
Delete user data . . . . . . . . . . . . . 33
Uploading data files and previewing the data . . 13
View analysis status . . . . . . . . . . . 34
Downloading data files . . . . . . . . . 13
Deleting data files . . . . . . . . . . . 13
Configuring training settings . . . . . . . . 14 Chapter 5. Troubleshooting . . . . . . 37

© Copyright IBM Corp. 2017 vii

viii User Guide
Chapter 1. Product overview
IBM® Prescriptive Maintenance on Cloud focuses on the needs of the reliability
engineer to identify and manage asset reliability risks that could adversely affect
plant or business operations. It applies machine learning and analytics to
operational data generated by critical assets to gain a better understanding of asset
performance and then operationalize these analyses for continuous process
improvement regarding maintenance practices and resources.

Welcome to Prescriptive Maintenance on Cloud

IBM Prescriptive Maintenance on Cloud looks for patterns in how an equipment
asset is used and the environment in which it is operating. It then correlates this
information with any known failures in the equipment. These correlations are used
to evaluate new data about the equipment asset, resulting in predictive scores that
indicate the relative health of the equipment and the likelihood of future failures.

The ability to determine when maintenance should be performed on equipment

assets leads to the following business benefits:
v You can estimate and extend the life of your assets.
v You can increase the return on your assets.
v You can optimize your maintenance, inventory, and resource schedules.

Prescriptive Maintenance on Cloud goes beyond both preventive and regularly

scheduled maintenance to ensure asset performance, thereby enabling maximized
value at every step of a process. Using Prescriptive Maintenance on Cloud, you can
perform the following tasks:
v Predict the failure of an instrumented asset so that you can prevent costly
unexpected downtime.
v Make adjustments to maintenance schedules and tasks to reduce repair costs and
minimize downtime.
v Determine the most effective maintenance cycles.
v Identify the root cause of asset failure faster so that you can take corrective
actions.

Instrumented assets generate data such as device ID, timestamp, temperature, and
status code.

Examples of instrumented assets are manufacturing equipment, mining equipment,

drilling equipment, farming equipment, security equipment, cars, trucks, trains,
helicopters, engines, cranes, oil platforms, and wind turbines.

Data from instrumented assets and data from other sources such as maintenance
records, maintenance logs, inspection reports, repair invoices and warranty claims
can be collected and used in models that predict when an asset is likely to fail.

Prescriptive Maintenance on Cloud helps an organization optimize its maintenance

program by developing a set of recommendations to carry out when specific
changes in asset health are identified. These recommendations can be based upon
analysis of historical maintenance records, best practices and procedures provided
by subject matter experts, original equipment manufacturer recommendations, as

© Copyright IBM Corp. 2017 1

 well as analysis of correlations that indicate potential problems or pending asset
failure. The goal of developing these optimized decisions or recommendations is to
further improve maintenance practices. When an anomaly in asset performance is
identified specific recommendations can be made to maintenance personnel in
order to affect the most efficient remedy of the problem. In the case of a
manufacturing organization which provides service or warranty support for
products that are used by hundreds or thousands of customers, the ability to
monitor asset performance in the field and proactively initiate a service call, in
comparison to waiting for the client to initiate a repair request or warranty claim,
can significantly transform the way the manufacturer provides service. With a
better understanding of asset usage and performance in the field manufacturer
may wish to modify its warranty program based upon asset usage or warranty
costs. Additionally greater insight into asset usage and performance can help the
manufacturer to optimize the parts inventory and locations so as to reduce the
volume of inventory and identify locations to facilitate proactive customer service.

For example, an automobile assembly plant is a system that combines thousands of

pieces of equipment with interlocking pieces. It is critical that such a system is able
to work efficiently and produce safe, high quality products. Prescriptive
Maintenance on Cloud looks for patterns in the usage and environmental
information for equipment that correlate with failures that take place. These
patterns are used to create predictive models to score incoming new data in order
to predict the likelihood of failure. Scores that are generated from this information
give an indication of the health of the piece of equipment. In addition, key
performance indicators (KPIs) are collected, which are used for reporting. KPIs
help to identify assets that do not conform to normal patterns of behavior. The
plant employees can use dashboards and reports to monitor and track the lifecycle
of each piece of equipment.

How it works
The product discovers leading and lagging indicators of equipment failure and
builds models that predict future failure.

The product accepts generic input data in the form of metrics, categorical values,
and events. Metrics are time series that typically represent operating data obtained
from sensors or production systems. Categorical values may be time-variant, like
the operating state of a machine, or static, like the manufacturer of a machine.
Events are generally intermittent and record the fact that something occurred.
Examples of events include alarms, inspections, and maintenance that is
performed.

Required data
The minimum data that is required to create failure predictions by using
Prescriptive Maintenance on Cloud includes failure data, operating data, the
frequency of the collection of the operating data versus the analysis period, and
master data.

Failure data

The most critical piece of data that is required by Prescriptive Maintenance on

Cloud is the equipment failure history. Equipment failure is represented by a
boolean flag that is either true or false. A true flag means the equipment failed,
and a false flag means the equipment did not fail. This flag must be coded
accurately. Any errors in coding the failure flag translate directly into prediction

2 User Guide
inaccuracy. For example, if your coding is only 50% accurate, the ceiling on model
accuracy is 50%.

Operating data

Operating data describes anything that is known about the tasks that the
equipment performs or the signals that the equipment emits, for example,
temperatures, pressures, noise levels, and vibration levels. Good operating data
enables the product to determine how the historical usage or loading of each piece
of equipment corresponds with the signals that it produces and how load and
signals are correlated with failure.

Frequency of the collection of operating data versus the analysis

period
Prescriptive Maintenance on Cloud builds a history of operating data. When it
analyzes this history, it aligns various pieces of operating data to a fixed interval so
that it can examine the correlation between multiple variables that are collected at
different times. The default time period for analysis is daily.

When performing daily analysis, Prescriptive Maintenance on Cloud resamples and

aggregates all operating data to the daily level. Resampling and aggregation
speeds up analysis and smooths the data to make patterns more clearly evident.
Resampling to a fixed interval is particularly effective when loading on equipment
is reasonably consistent throughout the day.

In some cases, loading of equipment is cyclical in nature, that is, the value of
variables changes considerably depending on which stage of a manufacturing cycle
the equipment is performing. In these cases, it is usually better not to resample.
Instead, supply summary data that represents the values of operating variables for
each cycle. When you supply data that is pre-summarized to the end of a cycle,
supply all variables that describe a single cycle with a common time stamp that
represents the end of the cycle.

Master data

Each asset that is monitored by IBM Prescriptive Maintenance requires an asset_id

value and an asset_type value.

Next scheduled maintenance date

The next scheduled maintenance date is used to evaluate the current maintenance
strategy. If a piece of equipment is predicted to fail before its next scheduled
maintenance, Prescriptive Maintenance on Cloud flags it as under-maintained and
recommends that the maintenance schedule be advanced.

Well-maintained days

This parameter indicates the wanted number of days between expected failure and
planned maintenance. Tracking well-maintained days serves as a factor of safety
for maintenance scheduling. It is not advisable to schedule maintenance on the
expected failure date as allowances must be made for model inaccuracy and
schedule delays.

Chapter 1. Product overview 3

Optional data
Some data, such as the next scheduled maintenance date and date of manufacture,
is optional but useful in prescriptive maintenance.

Static equipment attributes

Supply information about each piece of equipment such as the manufacturer and
tested load rating. The product uses this information along with operating data to
understand why different assets fail at different intervals.

Alarms and other events

Prescriptive Maintenance on Cloud learns which alarms and other events, such as
the need for operator intervention, typically precede equipment failure. You can
supply any available event data as a single column of data per event. Like the
failure flag, event columns feature a true value at any points of time when the
event occurred and a false value for periods where no failure ocurred.

Manufactured date

If you provide this value, the product calculates the age of the equipment and
learns to what extent failure is correlated with age.

Outputs
After the input data is analyzed, Prescriptive Maintenance on Cloud provides the
maintenance margin, top drivers, and risk factors.

Maintenance margin

Maintenance margin is a calculation that represents the number of days between

the next scheduled maintenance and the expected failure date. Use this information
to adjust maintenance schedules. A negative number implies that equipment will
fail before the next scheduled maintenance, so the schedule should be advanced to
avoid failure. A positive number implies that equipment will fail after the next
scheduled maintenance so scheduled maintenance can be postponed.

Top drivers

Top drivers is a sensitivity analysis that shows the strength of correlation between
operating variables and failure for each equipment instance. Prescriptive
Maintenance on Cloud ranks the drivers, indicates the current value of each driver,
and provides an estimate of the remaining value before failure in the driver unit.
For example, the top failure for a pump might be the cubic feet of water pumped
since last repair. At the time of analysis, pump 2672 has pumped 40000 cubic feet
of water. Prescriptive Maintenance on Cloud estimates that it pump another 10000
cubic feet before it fails. This value is intended as an estimate. When this estimate
is produced, Prescriptive Maintenance on Cloud assumes that the values of the
other drivers remain constant. The actual number of operating hours is influenced
by the changes in other variables.

Risk factors

Risk factors are characteristics of equipment that make them more or less
susceptible to failure. Unlike drivers, which are continuous numeric variables, risk

4 User Guide
 factors are categorical characteristics of the equipment. They are useful in
explaining why different equipment instances that are used in a similar way have
different failure rates. For example, pump 2672 has pumped 40000 cubic feet of
water since the last repair, and can pump 10000 more cubic feet before failure.
However, pump 8251 has also pumped 40000 cubic feet of water, but it can pump
only 10 more cubic feet. Pump 2672 is used in a clean environment. Pump 8251 is
used in a corrosive environment. The environment is shown as a risk factor with a
clean environment shown as a positive contributor for pump 2672 and a corrosive
environment is shown as a negative contributor for pump 8251.

Training models by using a subset of data

Prescriptive Maintenance on Cloud trains models by using 100% of historical data.
However, you can train by using a subset of data.

It is common to use a subset of the available historical data to train a model. This
way you can compare multiple iterations during the model build process and pick
the iteration with the lowest error. However, Prescriptive Maintenance on Cloud
uses 100% of historical data. If some of the training data is eliminated, overall
model accuracy decreases.

If you prefer to use a subset of data for training, you can upload a subset of the
historical data for training, then upload the rest of the data for scoring without
retraining. You can measure accuracy by using the predictions that are obtained
from the data that is used for scoring. After you measure accuracy in this way,
retrain by using all data. Doing this improves model accuracy over the number
you measured previously.

Failure prediction models are sensitive to the number of failure records that they
are trained on because failures are rare events. If you eliminate a random number
of rare events from the training data, the accuracy of the model and prediction
stability of subsequent training jobs is compromised.

Accuracy of predictions for long-range planning

While model accuracy might decrease for long-range planning, this decrease is
generally not a problem. The purpose of long-range planning is to ensure that
sufficient resources are available to perform future maintenance, not to make
decisions about individual assets.

When you look at aggregated maintenance needs for a population of equipment

that is based on long-range predictions, errors are averaged out. With a large
enough population and a wide long-range planning window, the average errors
approach zero. You can confirm this fact by extracting historical predictions and
aggregating errors for an equipment population.

How is data backed up and restored?

IBM Prescriptive Maintenance on Cloud is one of many solutions hosted on a
cloud cluster. IBM has a backup cluster for the live Prescriptive Maintenance on
Cloud cluster. The operations team uses cron jobs to back up all Prescriptive
Maintenance on Cloud data, including user data, asset entity and event data,
trained models, and analysis results, from the live cluster to the backup cluster.

The recovery timing is as follows:

Chapter 1. Product overview 5

 v Recovery time objective (RTO): 10 hours
v Maximum recovery point objective (RPO): 4 hours

What's new
The What's new section contains a list of new, changed, and deprecated
configuration features for this release. It also contains a cumulative list of similar
information for previous releases. You can use these lists to plan your application
deployment strategies and the training requirements for your users.

New in this release

March 2018
v The view analysis status API call enables you to get the analysis status
of a specific asset type.

New in previous releases

December 2017
v When data from different sources has the same date, the data is merged
before training and scoring.
v When an event file is uploaded that contains more or fewer columns
than the previously loaded and trained event file, Prescriptive
Maintenance on Cloud detects this issue and displays a warning
message.
v In this release, separate API keys are required for each user. In previous
releases, one API key was used for all users.
v A new API call enables you to delete all data for a specified user.
September 2017
v The Factory Reset option enables you to reset the product. Selecting this
option deletes all data and configuration settings, returning the product
to its original state.
March 2017
v The manufacture date field MF_DATE from entity property files is now
used for training. Two new fields, age_entity and age_entity_cusum, are
automatically generated and populated with data. The age_entity field
contains the age of the asset in days based on its manufacturing date.
The age_entity_cusom field contains the CUSUM (cumulative sum) of
asset age. If you do not want to use the manufacture date in training,
you can disable it by using the metaInput JSON object in the translation
file.
v The root-mean-square error (RMSE) value now shows in your
notifications.
v You now have more control of your data to customize the training and
scoring processes.
v When you hover over an asset card, you can view more information
about that asset card.

Accessibility features
Accessibility features help users who have a physical disability, such as restricted
mobility or limited vision, to use information technology products.

6 User Guide
For information about the commitment that IBM has to accessibility, see the IBM
Accessibility Center (www.ibm.com/able).

HTML documentation has accessibility features. PDF documents are supplemental

and, as such, include no added accessibility features.

Chapter 1. Product overview 7

8 User Guide
Chapter 2. Using the product
You can use Prescriptive Maintenance on Cloud to upload data. With the data, you
can view and analyze the maintenance requirements of equipment assets. You can
also organize and track your equipment assets.

Logging on
Log on to access Prescriptive Maintenance on Cloud.

About this task

Provide your IBM ID to your local administrator. Your administrator is responsible

for ensuring that you have the security access level that is appropriate to your role
in your organization. Your administrator will also supply you with the web
address URL for accessing the solution portal.

Procedure
1. Enter the URL into the address field of the browser.

Note: The fully qualified domain name is required in the URL, for example,
https://web_hostname/ibm/pm/ where web_hostname is the host name of the
web server. If you use the IP address instead of the registered fully qualified
domain name, some windows do not open correctly.
2. At the IBM ID log in page, enter your IBM ID and password. The Prescriptive
Maintenance on Cloud home page appears.

What to do next

On first use, you see a Welcome screen in which you can do the following things:
v Get a tour of the main elements of the software.
v View a tutorial video.
v View the documentation.
v Download sample data.
v Load sample data.

After first use, you can click the User icon to access these functions.

Each time you close your browser window, you must enter your IBM ID and
password to log in again.

Downloading the tutorial and sample data

Prescriptive Maintenance on Cloud includes a tutorial and sample data.

About this task

The topic shows how to download the tutorial and sample data.

© Copyright IBM Corp. 2017 9

 Procedure
1. Click the User icon and click Tutorial download to download the mp4 file for
the steps for a process engineer.
2. Click Sample data download to download the sample data that is used for the
tutorial.

Getting started
When you first use the product, you need to upload entity property files and event
files to trigger the creation and training of models.

When you first use the product, complete the following tasks in the order shown.
1. Optional: Configure model settings.
2. Upload the first entity property file.
3. Upload the first event file. After the first event file is successfully uploaded, the
product analyzes the data, creates a new model for each asset type, trains the
models, and then uses the models to do prediction. The last step is called
scoring.
4. After the model training and analysis are complete, you can view the analysis
results.
5. Optional: Upload more files. The new uploaded data is analyzed by existing
models to make new predictions.
6. Optional: Retrain existing models.
7. Optional: Reset user data.

Using data files

On the home page, you can upload entity property files and event files. You can
preview, download, and delete uploaded files.

Data from multiple sources

When data from different sources has the same timestamp, the data is merged
before training and scoring.

When values have conflicts, the following rules apply.

v Values from Watson™ IoT Platform have lower priority than values from
Maximo® Asset Management and the Prescriptive Maintenance on Cloud user
interface. Values from Maximo Asset Management and Prescriptive Maintenance
on Cloud overwrite values from Watson IoT Platform.
v If files from Maximo Asset Management and Prescriptive Maintenance on Cloud
have conflicts, the values from the most recently uploaded files overwrite the
conflicting values from previous files.

Entity property files

Entity property files describe asset detail information and associate an asset with
an asset type.

The following table describes important columns in the entity property file.

10 User Guide
 Table 1. Important columns in the entity property file
Column name Description
ASSET_ID A string that identifies each asset. The ASSET_ID column
cannot contain empty or null values. This column is
required.
ASSET_TYPE A string that describes the type of asset. Values in the
ASSET_TYPE column cannot contain spaces. This
column is required.
NEXT_PM The next planned maintenance date for an asset. The
margin value is the predicted failure date minus
NEXT_PM. This column is required.
MF_DATE The manufacture date of an asset. This column is
optional.
WELL_MAINTAIN A window in days. Prescriptive Maintenance on Cloud
uses this window to determine if an asset is
well-maintained, under-maintained or over-maintained.
v If the next planned maintenance date (NEXT_PM) is
after the predicted failure date (the margin is less than
0), the asset is categorized as under-maintained.
v If the next planned maintenance date is
WELL_MAINTAIN days before the predicted failure
date (the margin greater than WELL_MAINTAIN), the
asset is categorized as over-maintained.
v If the margin value is greater than 0 and less than
WELL_MAINTAIN, the asset is well-maintained.
Margin = Predicted_Next_Failure_Date - NEXT_PM

This column is required.

Important:
v The file must not contain duplicate column names.
v The MF_DATE column must use one of the following date formats consistently
within each file:
– yyyy-MM-dd hh:mm:ss
– dd/MM/yyyy HH:mm:ss
– yyyy/MM/dd H:mm:ss
– yyyy/M/d H:mm:ss
– d/M/yyyy H:mm:ss
– dd.MM.yyyy HH:mm:ss
– MM-dd-yyyy HH:mm:ss tt, where tt = AM or PM

Event files
Event files record failure events and asset operating data. Event data must be in a
continuous time series that is recorded at a second, minute, hourly, daily, weekly
or monthly level.

Important:
v The file must not contain duplicate column names.
v Column names cannot contain spaces.

Chapter 2. Using the product 11

 v Column names cannot contain the following special characters: ' " < > + # . % * ;
: \ -.
v The file must contain an asset_id column. The asset_id column cannot contain
empty or null values.
v The file must record failure events.
v The failure event, maintenance event, and repair event column values must be 0
or 1.
v The file must contain a timestamp column. The timestamp column cannot
contain empty or null values.
v The file must contain an is_fail column. is_fail is a keyword for a failure event
field in an event file. If you use is_fail as a column name for another field, the
column will be renamed as is_fail_renamed. In the analysis results, the failure
event field will be renamed as is_fail. The values in the is_fail column must be
0 or 1.
v The nextMaintenance and manufacturedDate columns must use one of the
following date formats consistently within each file:
– yyyy-MM-dd hh:mm:ss
– dd/MM/yyyy HH:mm:ss
– yyyy/MM/dd H:mm:ss
– yyyy/M/d H:mm:ss
– d/M/yyyy H:mm:ss
– dd.MM.yyyy HH:mm:ss
– MM-dd-yyyy HH:mm:ss tt, where tt = AM or PM
v When multiple event files are uploaded, when there are potential conflicts, the
column values from the newer file will overwrite the column values from the
older file.

Translation file
You can use the translation file to map columns in your data files.

Uploaded files are automatically parsed to determine whether they are entity
property files or event files. Columns in uploaded files are automatically read to
determine the type of data they contain. If you want to customize the way data in
uploaded files is handled, you can edit the translation file.

The translation file contains the following JSON objects:

entityPropertyFile
Used to map columns in entity property files.
dataFile
Used to map columns in event files.
metaInput
Used to enable or disable fields in entity property files.
exclusion
Used to exclude fields in event files.

In the entityPropertyFile JSON object, assetType is the asset type,

nextMaintenance is the next maintenance date, and manufactureDate is the
manufactured date.

12 User Guide
 In the dataFile JSON object, failureEvent is the failure event flag,
maintenanceEvent is the maintenance event flag, and repairReplaceEvent is the
repair or replace event flag.

In the following example of the metaInput JSON object, IBM is the asset type, and
disable indicates that the non-required data fields for the IBM asset type from the
entity property file are disabled.
"metaInput":
{
"IBM":"disable"
}

In the following example of the exclusion JSON object, IBM is the asset type, and
index and f4 are the fields to be excluded.
"exclusion":
{
"IBM":"index,f4"
}

Uploading data files and previewing the data

On the home page, you can upload entity property files and event files. You can
upload the files one at a time. Upload entity property files first, and then upload
event files.

About this task

After an event file is uploaded, Prescriptive Maintenance on Cloud parses the

uploaded data. If a model does not exist for the assets, Prescriptive Maintenance
on Cloud builds the prediction model, predicts the next failure date, and gets the
analysis results for each asset. If a model already exists, the product performs
scoring and determines the new next failure date. You can manually retrain the
model if necessary.

Procedure
1. Select My Data > Files > Browse for file, select the file, and select Open.
2. In the Progress pane, view the status of the upload.
3. In the Uploaded Files section, view the files.
4. To view the uploaded file variable type and dimension information, hover over
the tags in each file card.
5. To preview the data in the file, click Preview. The first 10 rows of the data are
shown.

Downloading data files

On the home page, you can download a data file that has been uploaded.

Procedure
1. Select My Data > Files.
2. In the Uploaded Files section, view the files.
3. From the list of files, click the menu button of the file and click Download.

Deleting data files

On the home page, you can delete a data file that has been uploaded.

Chapter 2. Using the product 13

 About this task

When you delete a file, the operation does not delete the associated events in the
database, but only deletes the file from the system.

Procedure
1. Select My Data > Files.
2. In the Uploaded Files section, view the files.
3. From the list of files, click the menu button of the file and click Delete.

Configuring training settings

You can configure the training settings by configuring the data handling rules and
model settings. Training settings affect how the data for each asset type is
processed and how the model is trained. To access these settings, select the settings
icon.

You can configure training settings before you upload the first event data, or
retrain existing models.

Resampling and data handling rules

You can configure the resampling rule and data handling rules that are used by the
product.

Resampling rule
Resampling rules affect the way data is aggregated. Two options are available for
resampling data:
None No resampling technique is performed on the original data set.
Downsampling
Data is downsampled. Downsampling rebalances the distribution between
failure events and non-failure events to increase the proportion of failure
events.

Data handling rules

Data handling rules specify the aggregation rule and missing value rule for each
column in the event file. You can define data handling rules for individual fields.
The options for aggregation rule and missing value rule depend on the type of
field.

The following table describes the types of aggregation rules.

Table 2. Aggregation rules
Aggregation rule Description
T_IF_ANY_T True if any true. If one or more true values
exist, then true is used.
F_IF_ANY_F False if any false. If one or more false values
exist, then false is used.
mode The value that occurs most frequently is
used.
first The value that occurs first is used.

14 User Guide
 Table 2. Aggregation rules (continued)
Aggregation rule Description
last The value that occurs last is used.
sum The sum of all values is used.
mean The average of all values is used.
max The maximum value is used.
min The minimum value is used.
count The number of events in the time period is
used.

The following table describes the types of missing value rules.

Table 3. Missing value rules
Missing value rule Description
false If the event does not have this value, false is
used.
ffill The last available value is used.
bfill The first available value is used.
mean The average is used.
min The minimum value is used.
max The maximum value is used.
interpolate A linear interpolation is used.

Configuring model settings

The product uses default model settings to preprocess data and train models.
However, you can configure your model settings to train your model. You can
choose to automatically or manually configure the model settings. If you
automatically configure the model settings, the product uses default values for
model settings and data preprocessing. If you manually configure the model
settings, you have more control of the values.

About this task

You can use the model settings function to set the configuration settings for
currently selected asset type. You can switch asset types using the asset type menu
on the asset panel.

When you upload your data, an initial training is automatically completed by

using the following parameters:
v Number of Trees: 3
v Maximum Depth of Tree: 3
v Learning Rate: 0.3

For the Gradient Boosting Decision Tree (GBDT) regression model type, the
following parameters are available:
v Number of Trees, range [2, 100], default 10
v Maximum Depth of Tree, range [2, 10], default 4
v Loss Function, default leastSquaresError, logLoss, and leastAbsoluteError

Chapter 2. Using the product 15

 v Learning Rate, range [0.01, 0.3], default 0.1

During training, the settings icon and Retrain button are disabled.

Procedure
1. Click the settings icon.
2. On the Model Settings tab, select the analysis interval to define how you want
the data set to be aggregated before training. The default option for the analysis
interval is daily or you can select weekly, monthly, yearly, or the raw option
where no aggregation is done during training.
3. In the Define Model Settings box, do one of the following options:
a. Add the parameter values.
b. Select Automatically define model parameters to have the values be
automatically defined. If you select Automatically define model
parameters, the training speed is slow.
4. Optional: Select Automatically Retrain on Apply for the system to
automatically start a training that is based on the new values.
5. Select Apply.

Analysis results
After you upload entity property files and event files, the product analyzes the
information, trains the model based on the data that is uploaded, and then
provides the analysis result. You can select an asset type to see the analyses for all
equipment assets of that type.

On the asset pane, click the asset type menu and select an asset type.

Equipment assets are grouped by the following categories:

All Assets
All equipment assets, or all equipment assets of the selected type.
Over Maintained
Scheduled maintenance is long before the predicted failure of the asset.
Well Maintained
Scheduled maintenance is close to the predicted failure of the asset.
Under Maintained
Scheduled maintenance is long after the predicted failure of the asset.
Needs Data
More data is required to analyze the asset.

Each equipment asset appears on a separate card that provides important

indicators of asset health:
Top Driver
Top driver is a sensitivity analysis that shows the strength of correlation
between operating variables and failure for each equipment instance.
Prescriptive Maintenance on Cloud ranks the drivers, indicates the current
value of each driver, and provides an estimate of the remaining value
before failure in the driver unit. For example, the top failure for a pump
might be the cubic feet of water pumped since last repair. At the time of
analysis, pump 2672 has pumped 40000 cubic feet of water. Prescriptive
Maintenance on Cloud estimates that it pump another 10000 cubic feet

16 User Guide
 before it fails. This value is intended as an estimate. When this estimate is
produced, Prescriptive Maintenance on Cloud assumes that the values of
the other drivers remain constant. The actual number of operating hours is
influenced by the changes in other variables.
Margin
The number of days' difference between the next scheduled maintenance
and the next predicted failure of the asset.

Attention: Negative margin means that the asset is predicted to fail

before the next scheduled maintenance.
Risk Factors
Risk factors are characteristics of equipment that make them more or less
susceptible to failure. Unlike drivers, which are continuous numeric
variables, risk factors are categorical characteristics of the equipment. They
are useful in explaining why different equipment instances that are used in
a similar way have different failure rates. For example, pump 2672 has
pumped 40000 cubic feet of water since the last repair, and can pump
10000 more cubic feet before failure. However, pump 8251 has also
pumped 40000 cubic feet of water, but it can pump only 10 more cubic
feet. Pump 2672 is used in a clean environment. Pump 8251 is used in a
corrosive environment. The environment is shown as a risk factor with a
clean environment shown as a positive contributor for pump 2672 and a
corrosive environment is shown as a negative contributor for pump 8251.
Graph The graph gives a concise visual reference of asset health:
v The gray horizontal bar indicates the total top driver count before the
next predicted failure of the asset.
v The dark portion of the horizontal bar indicates the current top driver
count.
Attention: If the entire horizontal bar is dark, then the asset is already
past its next predicted failure.
v The vertical bar indicates the current average top driver count for this
asset type.
v The arrow at the end of the horizontal bar indicates whether the current
top driver count is more than twice the average top driver count for this
asset type.

You can sort the cards based on Margin, or filter the cards based on Top Driver or
Asset Property Type.

Click a card to open a window that shows more analysis results for the asset,
including the following details:
v The number of days until the next scheduled maintenance date.
v All drivers (that is, variables that contribute to asset failure). A driver whose
name ends with _LTD is derived from an existing field by accumulating between
failure events.
v The maintenance history of the asset.
v A histogram of all drivers or a particular driver.

Tip: Specify a time scale to view a particular period.

Chapter 2. Using the product 17

Notification
You can receive system notifications in Prescriptive Maintenance on Cloud. The
notifications show system events such as file upload and model training.

The notification icon shows the number of notifications that you have received and
show in chronological order.

The notification types are:

File uploaded
A system notification indicates the user has completed uploading a file.
File deleted
A system notification indicates the user has deleted a file.
Analysis complete
A system notification indicates an analysis of an asset is complete.
Asset added
A system notification shows that assets were added to the system.
Errors found
Shows errors and the cause of each error. Errors may be caused by failures
during file upload, failures during analysis of an asset, or other reasons.

Resetting the product

You can use the Factory Reset feature to reset the product. Selecting this option
deletes all data, trained models, and configuration settings, returning the product
to its original state.

About this task

To reset the product, complete the following steps.

Procedure
1. Log in to the product.
2. Click the user icon.
3. Select the Factory Reset option.
4. Click Continue. Data deletion begins.

Results

When the factory reset operation is complete, the message "Successfully deleted
user data" is displayed.

18 User Guide
Chapter 3. Integrating with other products
You can integrate Prescriptive Maintenance on Cloud with other products to
import and analyze maintenance data from those products.

Integration with Watson IoT Platform

Prescriptive Maintenance on Cloud integrates with IBM Watson® IoT Platform to
make predictions about asset failure based on the current state of the asset.

Configure your organization in Watson IoT Platform

Before integrating Prescriptive Maintenance on Cloud with Watson IoT Platform,

you must configure your organization in the platform. This guide does not explain
how to create device types and device IDs.

You must add ASSET_ID in the Metadata field of a device ID. This metadata is
used to determine with which asset ID this device associated. For example,
{
"ASSET_ID": "JK84-JAS-144"
}

Next, send data to devices. Prescriptive Maintenance on Cloud supports receiving

simple JSON data only. See the following example. The timestamp and asset ID are
added automatically when generating files:
{
“is_fail”:”true”,
“temperature_flag”:”true”,
“event_count”:”3”
}

Add Integration

Before integrating with Watson IoT Platform, you need to create an integration in
Prescriptive Maintenance on Cloud. You must enter the organization ID, API key,
and authentication token to connect to Watson IoT Platform. After connecting to an
organization, you can select the devices from which to receive data.

In Prescriptive Maintenance on Cloud, in the IoT tab, click Add Integration. Enter
the organization ID, API key, and authentication token to connect to an
organization. You can create only one integration for each organization.

Once connected to the organization, select the devices from which you want to
receive data. Select Add all data from all devices, or click Select
Devices/Types/Events to select specific devices.

Set Analysis Interval to Daily. The analysis interval is how often Prescriptive
Maintenance on Cloud triggers IoT data analysis. Only the Daily interval is
supported.

© Copyright IBM Corp. 2017 19

 Analyze IoT data

After an integration with Watson IoT Platform is created, Prescriptive Maintenance

on Cloud receives data from that integration and starts analyzing daily, at
midnight.

You can disconnect from the integration by clicking Disconnect. Prescriptive

Maintenance on Cloud stops receiving data from that integration. You can
reconnect to integration by clicking Connect.

You can pause analysis data from an integration by clicking Pause Analysis.
Prescriptive Maintenance on Cloud stops analyzing data from that integration. You
can continue analysis by clicking Continue Analysis. During a pause, Prescriptive
Maintenance on Cloud can still receive data from that integration.

Integration with Maximo Asset Management

Prescriptive Maintenance on Cloud uses historical maintenance and metric data to
make predictions about asset failure based on the current state of the asset. IBM
Maximo Asset Management is an application that supports device maintenance. It
can be the source of some or all of the asset data that is needed by Prescriptive
Maintenance on Cloud.

Overview

Asset data from Maximo Asset Management may be provided to PM either as a

file (csv) or through a http interface. The csv format may be most convenient for a
large historical data set while the http interface is convenient for ongoing updates.

Maximo Asset Management provides the capability to define and filter the data for
export to PM. In general, the exported data fields correspond to the columns of a
csv file for upload to PM. Thus they include timestamp, asset id, failure,
maintenance, and metric data.

Maximo Asset Management integration capabilities are defined using several

components. Object Structures define the data fields. Publish Channels can
determine when the data is exported and can filter which data to include. End
Points determine the export method (ex. File or HTTP). And External Systems
combine a Publish Channel with an End Point.

Work Orders

Maximo Asset Management work orders can be an import source of maintenance

data for the PM application. This includes asset failure, asset repair or replace, and
routine maintenance. The following is a description of a set of Maximo Asset
Management external system artifacts that allow export of this work order data to
Prescriptive Maintenance on Cloud.

Object Structure – PMEX_WORKORDER

The PMEX_WORKORDER Object Structure includes the following fields of

WORKORDER. The extended fields (PMEX_) are added as non-persistent attributes
of WORKORDER.
v ASSETNUM
v SITEID

20 User Guide
v WORKTYPE
v FAILDATE
v FAILURECODE
v STATUS
v STATUSDATE
v PMEX_TIMESTAMP
v PMEX_ASSET_ID
v PMEX_IS_FAIL
v PMEX_IS_MAINTENANCE
v PMEX_IS_REP_REPLACE
v WELL_MAINTAIN

Prescriptive Maintenance on Cloud needs the values of timestamp, asset_id, is_fail,

is_maintenance, is_rep_replace. By default, Prescriptive Maintenance on Cloud will
infer these values. Asset_id is the combination of ASSETNUM+SITEID. If there is a
FAILURECODE then is_fail is TRUE and the timestamp is FAILDATE (otherwise
timestamp is STATUSDATE). If the WORKTYPE is CM or EM then is_rep_replace
is TRUE and otherwise if it is PM then is_maintenance is TRUE.

This default mapping can be overridden by using a publish channel to explicitly

set the value of any of the fields beginning with PMEX_.

Meters

Maximo Asset Management meters can provide metric data that the PM
application uses to train and score asset failure models.

Object Structure – PMEX_METER

The PMEX_METER Object Structure includes the following fields of ASSETMETER.

The extended fields (PMEX_) are added as non-persistent attributes of
ASSETMETER.
v SITEID
v METERNAME
v LASTREADING
v LASTREADINGDATE
v PMEX_TIMESTAMP
v PMEX_ASSET_ID
v PMEX_METRIC_NAME
v PMEX_METRIC_VALUE

Prescriptive Maintenance on Cloud uses Maximo Asset Management meters as

metrics. It needs the values of timestamp and asset_id. It also needs the metric
name and value. By default, Prescriptive Maintenance on Cloud will infer these
values. Asset_id is the combination of ASSETNUM+SITEID. Timestamp is
LASTREADINGDATE. The metric name is the METERNAME and the metric value
is LASTREADING.

As with the work order, this default mapping can be overridden by using a
publish channel to explicitly set the value of any of the fields beginning with
PMEX_.

Chapter 3. Integrating with other products 21

 Endpoints

The object structures for Prescriptive Maintenance on Cloud can be exported using
an HTTP or Flat File (csv) endpoint. A Flat File endpoint may be more appropriate
for a large amount of historical data while an HTTP endpoint is ideal for
automatically exporting new data.

The HTTP endpoint must be configured with the URL, username, and password
for Prescriptive Maintenance on Cloud. It must also include the header
“content-type application/xml” and the http method “POST”.

Summary

Maximo Asset Management is an application with powerful capabilities to support

asset maintenance. Historical data in Maximo Asset Management can be used to
train and score predictive models in Prescriptive Maintenance on Cloud. The
resulting "days to failure" prediction in Prescriptive Maintenance can provide a
valuable insight to help optimize asset maintenance and thus provide efficient and
productive asset use.

22 User Guide
Chapter 4. Application programming interface
The application programming interface enables you to perform actions such as
posting entity and event data, uploading a file, and retrieving a result.

Post entity data from Maximo Asset Management

Posts entity data from IBM Maximo Asset Management.
URL
/ibm/pm/service/apiWrapper?apiName=maximo
Method
The request type POST
URL parameters
Required
apiName="Maximo"
Username=IBM_ID
type=property
Optional
None
Headers for IBM ID authorization
Content-Type: text/html
Authorization: Basic authorization_string. Optional. Use IBMid credentials
when an API key is not available.
Headers for API key authorization
Content-Type: text/html
APIKEY: API_Key

Sample body
<?xml version="1.0" encoding="UTF-8"?>
<PublishMXPM_PM xmlns="http://www.ibm.com/maximo"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
creationDateTime="2017-03-14T13:37:20-04:00" transLanguage="EN"
baseLanguage="EN" messageID="3456160.1489513040315874425"
maximoVersion="7 6 $build$ V7608-46" event="1">
<MXPM_PMSet>
<ASSET action="Add">
<ASSETNUM>80901</ASSETNUM>
<PMEX_ASSET_ID> 80901vBEDFORD </PMEX_ASSET_ID>
<PMEX_ASSET_TYPE> AIRCRAFTBRAKES</PMEX_ASSET_TYPE>
<PMEX_MANUFACTURE_DATE>2016-03-01T00:00:00-05:00</PMEX_MANUFACTURE_DATE>
<PMEX_NEXT_MAINTENANCE>2017-03-01T00:00:00-05:00</PMEX_NEXT_MAINTENANCE>
<SITEID>BEDFORD</SITEID>
<WELL_MAINTAINED>14</WELL_MAINTAINED>
</ASSET>
</MXPM_PMSet>
</PublishMXPM_PM>
Success response

© Copyright IBM Corp. 2017 23

 Code
200 OK
Content
{}
Error response
Code
403
Content
The server encountered an internal error and was unable to complete your
request. Please contact system administrator.
Sample call
https://server_url/ibm/pm/service/apiWrapper?apiName=maximo
&Username=user_name&type=property
Notes This REST API call supports both user API key and IBM ID for
authentication. To generate the authorization_string for an IBM ID, base64
encode the user name and password separated by a colon, for example:
Authorization: Basic base64_encode(IBM_ID ':' IBM_ID_password)

Post event data from Maximo Asset Management

Posts event data from IBM Maximo Asset Management.
URL
/ibm/pm/service/apiWrapper?apiName=maximo
Method
The request type POST
URL parameters
Required
apiName=maximo
Username=IBM_ID
type=data
Optional
None
Headers for IBM ID authorization
Content-Type: text/html
Authorization: Basic authorization_string. Optional. Use IBMid credentials
when an API key is not available.
Headers for API key authorization
Content-Type: text/html
APIKEY: API_Key

24 User Guide
Sample Body
<?xml version="1.0" encoding="UTF-8"?>
<PublishPMEX_WO_ANDMETER
xmlns="http://www.ibm.com/maximo"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
creationDateTime="2016-10-18T12:11:29-04:00"
transLanguage="EN" baseLanguage="EN" messageID="3114181.1476807089991377382"
maximoVersion="7 6 20151202-1713 V7603-143" event="1">
<PMEX_WO_ANDMETERSet>
<WORKORDER action="Replace">
<ASSETNUM>811</ASSETNUM>
<FAILDATE xsi:nil="true" />
<PMEX_ASSET_ID>811vBEDFORD</PMEX_ASSET_ID>
<PMEX_IS_FAIL>TRUE</PMEX_IS_FAIL>
<PMEX_IS_MAINTENANCE>TRUE</PMEX_IS_MAINTENANCE>
<PMEX_IS_REP_REPLACE>FALSE</PMEX_IS_REP_REPLACE>
<PMEX_TIMESTAMP>2016-12-05T12:11:25-04:00</PMEX_TIMESTAMP>
<SITEID>BEDFORD</SITEID>
<STATUS changed="1" maxvalue="COMP">COMP</STATUS>
<STATUSDATE changed="1">2016-12-05T12:11:25-04:00</STATUSDATE>
<WONUM>6011</WONUM>
<WORKTYPE>PM</WORKTYPE>
<PMEX_METER_READINGS>
<meterdata>
<metername>TEMP</metername>
<meterreading>200</meterreading>
</meterdata>
<meterdata>
<metername>PRESSURE</metername>
<meterreading>2.0</meterreading>
</meterdata>
<meterdata>
<metername>VIBRATION</metername>
<meterreading></meterreading>
</meterdata>
</PMEX_METER_READINGS>
<PMEX_OPTIONAL_FIELDS>
<optionalfield>
<fieldname>OptionalField1</fieldname>
<fieldvalue>TestValue</fieldvalue>
</optionalfield>
<optionalfield>
<fieldname>OptionalField2</fieldname>
<fieldvalue>TestValue</fieldvalue>
</optionalfield>
</PMEX_OPTIONAL_FIELDS>
</WORKORDER>
</PMEX_WO_ANDMETERSet>
</PublishPMEX_WO_ANDMETER>
Success response
Code
200 OK
Content
{}
Error response
Code
403
Content

Chapter 4. Application programming interface 25

 The server encountered an internal error and was unable to complete your
request. Please contact system administrator.
Sample call
https://server_url/ibm/pm/service/apiWrapper?apiName=maximo
&Username=user_name&type=data
Notes This REST API call supports both user API key and IBM ID for
authentication. To generate the authorization_string for an IBM ID, base64
encode the user name and password separated by a colon, for example:
Authorization: Basic base64_encode(IBM_ID ':' IBM_ID_password)

Post entity data in JSON format

Posts entity data in JSON format.
URL
/ibm/pm/service/apiWrapper?apiName=json
Method
The request type POST
URL parameters
Required
apiName="json"
Username=IBM_ID
type=property
Optional
None
Headers for IBM ID authorization
Content-Type: application/json
Authorization: Basic authorization_string. Optional. Use IBMid credentials
when an API key is not available.
Headers for API key authorization
Content-Type: application/json
APIKEY: API_Key
Sample body
[
{
"ASSET_TYPE": " Aircraft ",
"ASSET_ID": "810vBEDFORD ",
"NEXT_PM": "2016-08-16 00:00:00",
"MF_DATE": "2015-05-01 00:00:00",
"LOCATION": "BEDFORD",
"MF": "IBM-US",
"WELL_MAINTAIN": 10,
"MACHINE_LEVEL": 1
}
]
Success response
Code
200 OK

26 User Guide
 Content
{}
Error response
Code
403
Content
The server encountered an internal error and was unable to complete your
request. Please contact system administrator.
Sample call
https://server_url/ibm/pm/service/apiWrapper?apiName=json
&Username=user_name&type=property
Notes This REST API call supports both user API key and IBM ID for
authentication. To generate the authorization_string for an IBM ID, base64
encode the user name and password separated by a colon, for example:
Authorization: Basic base64_encode(IBM_ID ':' IBM_ID_password)

Post event data in JSON format

Posts event data in JSON format.
URL
/ibm/pm/service/apiWrapper?apiName=json
Method
The request type POST
URL parameters
Required
apiName="json"
Username=IBM_ID
type=data
Optional
None
Headers for IBM ID authorization
Content-Type: application/json
Authorization: Basic authorization_string. Optional. Use IBMid credentials
when an API key is not available.
Headers for API key authorization
Content-Type: application/json
APIKEY: API_Key
Sample body
[
{
"ASSET_ID": "810vBEDFORD ",
"Timestamp": "2016-11-18 00:00:00",
"IS_FAIL":"TRUE",
"IS_MAINTENANCE": "FALSE",

Chapter 4. Application programming interface 27

 "IS_REP_REPLACE": "TRUE",
"Speed": 15.383333,
"Win_Angle": 180.665922,
"Wind_Grade": "Soft",
"Power_Grade": "Large"
}
]
Success response
Code
200 OK
Content
{}
Error response
Code
403
Content
The server encountered an internal error and was unable to complete your
request. Please contact system administrator.
Sample call
https://server_url/ibm/pm/service/apiWrapper?apiName=json
&Username=user_name&type=data
Notes This REST API call supports both user API key and IBM ID for
authentication. To generate the authorization_string for an IBM ID, base64
encode the user name and password separated by a colon, for example:
Authorization: Basic base64_encode(IBM_ID ':' IBM_ID_password)

Upload a file
Enables users to upload a file.
URL
/ibm/pm/service/apiWrapper?apiName=file_upload
Method
The request type POST
URL parameters
Required
apiName=file_upload
Username=IBM_ID
Optional
None
Headers for IBM ID authentication
Authorization: Basic authorization_string. Optional. Use IBMid credentials
when an API key is not available.
Headers for API key authentication
APIKEY: API_Key
Body Attach a CSV file.

28 User Guide
 Success response
Code
200 OK
Content
{"result":[{"fileId":"file_id","fileName":"file_name"}]}
Error response
Code
200 OK
Content
"result":["error_message": appropriate_error_message]
Sample HTTP call
https://server_url/ibm/pm/service/apiWrapper?apiName=file_upload
&Username=user_name
Sample CURL call with IBM ID authentication
curl –k -X post --user IBM_ID:IBM_ID_password --connect-timeout 600
-F file=@c:\inputdata_4_1M_part.csv "https://server_url/ibm/pm/
service/apiWrapper?apiName=file_upload&Username=user_name"
Sample CURL call with API key authentication
curl –k -X post -H "APIKEY:API_Key" --connect-timeout 600 -F
file=@c:\inputdata_4_1M_part.csv "server_url/ibm/pm/service/
apiWrapper?apiName=file_upload&Username=user_name"
Notes
CSV is the only file type that this REST API supports.
This REST API call supports both user API key and IBM ID for
authentication. To generate the authorization_string for an IBM ID, base64
encode the user name and password separated by a colon, for example:
Authorization: Basic base64_encode(IBM_ID ':' IBM_ID_password)

Download data
Downloads the analysis result data for one asset.
URL
/ibm/pm/service/apiWrapper ?apiName=file_download
Method
The request type POST
URL parameters
Required
apiName=file_download
Username=IBM_ID
Optional
Assetid=my_asset_ID
Assettype=my_asset_type
FutureOnly=true or false (default is false)

Chapter 4. Application programming interface 29

 Headers for IBM ID authentication
Authorization: Basic authorization_string. Optional. Use IBMid credentials
when an API key is not available.
Headers for API key authentication
APIKEY: API_Key
Success response
Code
200 OK
Content
Returns data in CSV format as part of the response.
Error response
Code
200 OK
Content
"result":["error_message": appropriate_error_message]
Sample HTTP calls
To download the file for a specific asset:
https://server_url/ibm/pm/service/apiWrapper?apiName=file_download
&Assetid=my_asset_ID&Username=user_name
To download the file for a specific asset type (all assets included):
https://server_url/ibm/pm/service/apiWrapper?apiName=file_download
&Assettype=my_asset_type&Username=user_name
To download the file with predictions for a specific AssetID:
https://server_url/ibm/pm/service/apiWrapper?apiName=file_download
&Assetid=my_asset_ID&Username=user_name&FutureOnly=true
To download the file with predictions for a specific AssetType:
https://server_url/ibm/pm/service/apiWrapper?apiName=file_download
&Assettype=AIRCRAFTBRAKES&Username=user_name&FutureOnly=true
Sample CURL calls with IBM ID authentication
To download the file for a specific asset:
curl -X post -u IBM_ID:IBM_ID_password -o filename.csv -v
"https://server_url/ibm/pm/service/apiWrapper?apiName=file_download
&Assetid=JK84-JAS-141&Username=user_name"
To download the file for a specific asset type (all assets included):
curl -X post -u IBM_ID:IBM_ID_password -o filename.csv -v
"https://server_url/ibm/pm/service/apiWrapper?apiName=file_download
&Assettype=my_asset_type&Username=user_name"
To download the file with predictions for a specific AssetID:
curl -X post -u IBM_ID:IBM_ID_password -o filename.csv -v
"https://server_url/ibm/pm/service/apiWrapper?apiName=file_download
&Assetid=my_assetID&Username=user_name&FutureOnly=true"
To download the file with predictions for a specific AssetType:

30 User Guide
 curl -X post -u IBM_ID:IBM_ID_password -o filename.csv -v
"https://server_url/ibm/pm/service/apiWrapper?apiName=file_download
&Assettype=AIRCRAFTBRAKES&Username=user_name&FutureOnly=true"
Sample CURL calls with API key authentication
To download the file for a specific asset:
curl -X post -H "APIKEY:API_Key" -o filename.csv -v
"https://server_url/ibm/pm/service/apiWrapper?apiName=file_download
&Assetid=JK84-JAS-141&Username=user_name"
To download the file for a specific asset type (all assets included):
curl -X post -H "APIKEY:API_Key" -o filename.csv -v
"https://server_url/ibm/pm/service/apiWrapper?apiName=file_download
&Assettype=my_asset_type&Username=user_name"
To download the file with predictions for a specific AssetID:
curl -X post -H "APIKEY:API_Key" -o filename.csv -v
"https://server_url/ibm/pm/service/apiWrapper?apiName=file_download
&Assetid=my_assetID&Username=user_name&FutureOnly=true"
To download the file with predictions for a specific AssetType:
curl -X post -H "APIKEY:API_Key" -o filename.csv -v
"https://server_url/ibm/pm/service/apiWrapper?apiName=file_download
&Assettype=AIRCRAFTBRAKES&Username=user_name&FutureOnly=true"
Notes
The output response for this REST API call produces data in CSV format.
This REST API call supports both user API key and IBM ID for
authentication. To generate the authorization_string for an IBM ID, base64
encode the user name and password separated by a colon, for example:
Authorization: Basic base64_encode(IBM_ID ':' IBM_ID_password)

Retrieve analysis results

Gets analysis result data in JSON format for a specific asset type or asset ID.
URL
/ibm/pm/service/apiWrapper?apiName=analysisResult
Method
The request type POST
URL parameters
Required
apiName=analysisResult
Username=IBM_ID
Optional
Assetid=my_asset_ID
Assettype=my_asset_type
Headers for IBM ID authentication
Authorization: Basic authorization_string. Optional. Use IBMid credentials
when an API key is not available.

Chapter 4. Application programming interface 31

 Headers for API key authentication
APIKEY: API_Key
Success response
Code
200 OK
Content
Returns JSON data that contains analyzed results.
Error response
Code
403
Content
The server encountered an internal error and was unable to complete your
request. Please contact system administrator.
Sample HTTP calls
To retrieve analysis results for a specific asset:
https://server_url/ibm/pm/service/apiWrapper?apiName=analysisResult
&Assetid=my_asset_ID&Username=user_name
To retrieve analysis results for a specific asset type, all assets included:
https://server_url/ibm/pm/service/apiWrapper?apiName=analysisResult
&Assettype=my_asset_Type&Username=user_name
Sample CURL calls with IBM ID authentication
To retrieve analysis results for a specific asset:
curl -X post -u IBM_ID:IBM_ID_password --connect-timeout 600
"https://server_url/ibm/pm/service/apiWrapper?apiName=analysisResult
&Assetid=my_asset_ID&Username=user_name" --insecure
To retrieve analysis results for a specific asset type, all assets included:
curl -X post -u IBM_ID:IBM_ID_password --connect-timeout 600
"https://server_url/ibm/pm/service/apiWrapper?apiName=analysisResult
&Assettype=my_asset_type&Username=user_name" --insecure
Sample CURL calls with API key authentication
To retrieve analysis results for a specific asset:
curl -X post -H "APIKEY:API_Key" --connect-timeout 600
"https://server_url/ibm/pm/service/apiWrapper?apiName=analysisResult
&Assetid=my_asset_ID&Username=user_name" --insecure
To retrieve analysis results for a specific asset type, all assets included:
curl -X post -H "APIKEY:API_Key" --connect-timeout 600
"https://server_url/ibm/pm/service/apiWrapper?apiName=analysisResult
&Assettype=my_asset_type&Username=user_name" --insecure
Sample response
{
"rmse": "0.630609733315",
"asset_type": "AIR_COOLER",
"data": [
{

32 User Guide
 "drivers": {
"driver": "normalcount,
2.872262773722628,1.9,2.868065693430657,1.2234554290771484",
},
"asset_id": "80901AIR",
"top_driver": "empty_col_cusum",
"missing_data": "0",
"risk_factor": {
"negative": "",
"positive": ""
},
"failure_date": "2019-04-04",
"top_driver_avg": "2493.2096715328466",
"top_driver_current": "235.43868613138685",
"margin": "436",
"maintainence_status": "3",
"top_driver_prediction": "591.5"
}
]
}
Notes This REST API call supports both user API key and IBM ID for
authentication. To generate the authorization_string for an IBM ID, base64
encode the user name and password separated by a colon, for example:
Authorization: Basic base64_encode(IBM_ID ':' IBM_ID_password)

Delete user data

Deletes all the data from the specified user's account. The account is not deleted.
URL
/ibm/pm/service/apiWrapper?apiName=deleteUserData
Method
The request type POST
URL parameters
Required
apiName=deleteUserData
Username=IBM_ID
Optional
None
Headers for IBM ID authorization
Content-Type: text/html
Authorization: Basic authorization_string. Optional. Use IBMid credentials
when an API key is not available.
Headers for API key authorization
Content-Type: text/html
APIKEY: API_Key
Success response
Code
200 OK
Content

Chapter 4. Application programming interface 33

 {}
Error response
Code
403
Content
The server encountered an internal error and was unable to complete your
request. Please contact system administrator.
Sample HTTP call
To delete all data for a specific user:
https://server_url/ibm/pm/service/apiWrapper?apiName=deleteUserData
&Username=user_name
Sample CURL call with API key authentication
To delete all data for a specific user:
curl -X post -H "APIKEY:API_Key" --connect-timeout 600
"https://server_url/ibm/pm/service/apiWrapper?apiName=
deleteUserData&Username=user_name" --insecure
Notes This REST API call supports both user API key and IBM ID for
authentication. To generate the authorization_string for an IBM ID, base64
encode the user name and password separated by a colon, for example:
Authorization: Basic base64_encode(IBM_ID ':' IBM_ID_password)

View analysis status

Gets analysis status of a specific asset type.
URL
/ibm/pm/service/apiWrapper?apiName=analysisStatus
Method
The request type POST
URL parameters
Required
apiName=analysisStatus
Assettype=my_asset_type
Username=IBM_ID
Optional
None
Headers for IBM ID authorization
Authorization: Basic authorization_string. Optional. Use IBMid credentials
when an API key is not available.
Headers for API key authorization
APIKEY: API_Key
Success response
Code

34 User Guide
 200 OK
Content
Returns JSON data that contains analysis status.
Sample HTTP call
To retrieve analysis results for a specific asset type, all assets included:
https://server_url/ibm/pm/service/apiWrapper?apiName=analysisStatus
&Assettype=my_asset_Type&Username=user_name
Sample CURL call with API key authentication
To retrieve analysis results for a specific asset type, all assets included:
curl -X post -H "APIKEY:API_Key" --connect-timeout 600
"https://server_url/ibm/pm/service/apiWrapper?apiName=analysisStatus
&Assettype=my_asset_type&Username=user_name" insecure
Sample response
{
"message": "NOT_TRAINED",
"status": "0"
}
Response items
NOT TRAINED: The asset type is not yet trained.
ANALYZING: The product is analyzing the asset type.
UNREAD: The asset type has been trained, but the user has not viewed
the result.
VIEWED: The asset type has been trained and the user has viewed the
result.
Notes This REST API call supports both user API key and IBM ID for
authentication. To generate the authorization_string for an IBM ID, base64
encode the user name and password separated by a colon, for example:
Authorization: Basic base64_encode(IBM_ID ':' IBM_ID_password)

Chapter 4. Application programming interface 35

36 User Guide
Chapter 5. Troubleshooting
This section contains troubleshooting information for Prescriptive Maintenance on
Cloud.

The following table shows possible error messages and corrective actions:
Table 4. Prescriptive Maintenance on Cloud error codes, messages, explanations, and
corrective actions
Error code Error message Explanation Action
PM-W-0003 The file X uploaded File uploaded
successfully successfully to the
data lake and finished
parser.
PM-W-0004 The file X was deleted The file deleted
successfully from the
data lake and Hbase.
PM-W-0005 The File X upload Exception happened Check that the file
failed during file upload to contains the correct
the data lake. timestamp column
and other required
fields. Event files
require failure event,
timestamp, and asset
id fields. Entity
property files require
asset type, asset id.
PM-W-0021 There is no asset_id ASSET_ID is a Add ASSET_ID
column in the file X required column in column in the file
Prescriptive which report this error
Maintenance data. But and upload it again.
file parser can not
find it from uploaded
file.
PM-W-0022 There is no timestamp Timestamp is a Add timestamp
column in the file X required column in column in x file and
Prescriptive upload it again.
Maintenance data. But
file parser can not
find it from x file.
PM-W-0023 There are repeated File parser found Remove or rename
column names in the duplicated column duplicated column
file X names in x file. name in x file, and
upload it again.
PM-W-0024 The type of File parser use Edit translation file
failurEvent, translation file to map correct flag value
maintenancEvent or parse x file and find fields to these
repairReplaceEvent failureEvent, specified columns. Or
column in the file X is maintenanceEvent or edit x file to make
not correct repairReplaceEvent these columns contain
didn't contain flag flag value
value (True/False). (True/False).

© Copyright IBM Corp. 2017 37

 Table 4. Prescriptive Maintenance on Cloud error codes, messages, explanations, and
corrective actions (continued)
Error code Error message Explanation Action
PM-W-0025 The type of File parser use Edit translation file
nextMaintenance or translation file to map correct
manufacturedDate parser the data find timestamp fields to
column in the file X is nextMaintenance or these specified
not correct manufacturedDate columns. Or edit data
didn't contain to make these
timestamp value. columns contain
correct timestamp
data(True/False)
PM-W-0026 There is no asset_type File parser find user Add one more column
column in the file. upload file is a entity ASSET_TYPE in this
Choose a column as property file. But this file or use translation
the asset_type file didn't contain file to map one
ASSET_TYPE column. column to asset_type.
PM-W-0027 None of the assets in User didn’t upload Upload entity
the file X have an asset entity property file property file contain
type. Upload an entity related with asset_id asset_type information
property file and in event data. for asset_id in event
manually trigger the data.
analysis
PM-W-0030 Successfully Disconnected from
disconnected from the IoT platform.
IBM Watson IoT
Platform
PM-W-0031 You haven't connected User disconnect from Refresh UI and
to this organization a unconnected IoT disconnect option
organization. disappear.
PM-W-0032 You already connected User try to connect to Refresh UI and
to this organization IoT organization connect option
which has already disappear.
been connected to.
PM-W-0033 Successfully connected User connected to IoT
to the IBM Watson IoT platform.
Platform
PM-W-0034 You must create a User try to connect to Create a IoT
connection to this an IoT platform which connection in IoT UI
organization before connection definition and then connect to it.
attempting to connect haven’t been created
yet.
PM-W-0035 Successfully deleted User delete one IoT
connection connection.
PM-W-0036 An error occurred An exception happen Disconnect to IoT
when deleting an IoT when delete IoT platform first then
connection connection. delete this connection
again.
PM-W-0039 An error occurred A timeout error Rerun analyzing by
while analyzing file X occurred when click retrain button on
analyzing file x UI.

38 User Guide
Table 4. Prescriptive Maintenance on Cloud error codes, messages, explanations, and
corrective actions (continued)
Error code Error message Explanation Action
PM-W-0040 File X upload failed Error happen when Check file format
with file parser system cannot parse matches with
File x translation file and
upload it again.
PM-W-0041 You already added User try to add a new User can edit old one
integration to integration connection to make the change.
organization X which already existed. IoT platform on
support one
connection from a
client.
PM-W-0043 An error occurred User uses invalid Check translation file
when saving the JSON file as is in correct JSON
translation file translation file. format.
PM-W-0044 The API Key or Failed to authenticate Check API key and
Authentication Token with IoT platform Authentication token
is wrong with input API key input correctly.
and authentication
token.
PM-W-0045 You must disconnect User try to save Disconnect from IoT
from IoT platform integration while it platform and then edit
before saving this connected. the connection.
integration
PM-W-0046 There is no dataFile User may delete dataFile object is
object in the translation dataFile object in required in translation
file translation file file. User need to add
it back or fix incorrect
spelling.
PM-W-0047 There is no User may delete entityPropertyFile
entityPropertyFile entityPropertyFile object is required in
object in the translation object in translation translation file. User
file file need to add it back or
fix incorrect spelling.
PM-W-0048 The dataFile object is After user edit User need to fix the
not a valid json object translation file format of dataFile
dataFile object is not a object in translation
valid JSON object. file before save it or
replace it.
PM-W-0049 The entityPropertyFile After user edit User need to fix the
object is not a valid translation file format of
json object entityPropertyFile entityPropertyFile
object is not a valid object in translation
JSON object. file before save it or
replace it.
PM-W-0050 You must provide After user edit User need to add
failureEvent, translation file failureEvent,
maintenanceEvent and failureEvent, MaintenanceEvent
repairReplaceEvent in MaintenanceEvent or and
the dataFile object repairReplaceEvent repairReplaceEvent in
can not be found in dataFile object.
dataFile object.

Chapter 5. Troubleshooting 39
 Table 4. Prescriptive Maintenance on Cloud error codes, messages, explanations, and
corrective actions (continued)
Error code Error message Explanation Action
PM-W-0051 You must provide After user edit User need to add
assetType, translation file assetType,
nextMaintainance, assetType, nextMaintainance,
manufacturedDate and nextMaintainance, manufacturedDate
wellMaintainThreshold manufacturedDate or and
in the wellMaintainThreshold wellMaintainThreshold
entityPropertyFile can not be found in in dataFile object.
object entityPropertyFile
object.
PM-W-0052 We found "key":"value" User use a column Change either column
pairs in the translation name in file x as Key name in file x or Key
file that duplicate in translation file. in translation file.
column titles in the file
X. "key" must use a
different name than
column titles
PM-W-0053 X rows contain an Asset id is null in File Provide a value for
empty asset id in the Y Asset id in File Y, and
uploaded file Y upload it again.
PM-W-0055 No asset type was System didn’t found Upload an entity
found for X assets in asset type related to x property file contain
event file Y number of assets in relation between asset
file Y. and its asset type.
PM-W-0056 The schema in file X Auto scoring process Edit file x to make
does not match the found uploaded file x columns consistent
previous model, so we columns didn’t match with previous model,
cannot score the previous trained or click retrain button
model. You must model. to retrain a new
manually train a new model.
model
PM-W-0057 There is no failure Prescriptive Add failure event in
event in file X maintenance analysis file x then upload it
require failure event again.
in event file, but file x
doesn’t have it. File x
will automatically
delete from system.
PM-W-0058 The event data for To build analytic Upload more data for
asset type X is not model it required at asset type x and click
sufficient to build a least 10 record. But retrain button to
model data for asset type x analysis it.
doesn’t have enough
data.
PM-W-0059 The model is not Analytic model is Upload more
accurate for asset type built, but the model historical event data
X accuracy is very low. and click retrain to
analysis.
PM-W-0060 Event data for asset Asset type x can’t Upload more
type X cannot fit a build a model with historical event data
model current event data and click retrain to
analysis.

40 User Guide
Table 4. Prescriptive Maintenance on Cloud error codes, messages, explanations, and
corrective actions (continued)
Error code Error message Explanation Action
PM-W-0061 The model built Entity property for Upload new entity
successfully, but the this model doesn’t property file
entity property file did contained next contained next
not contain the next planned maintenance planned maintenance
planned maintenance date. Model score date. And click retain
date. Margin cannot be cannot complete to analysis.
calculated. without that date.
PM-W-0062 The asset type X was Asset type x has
analyzed successfully finished analysis
successful.
PM-W-0063 The asset type X was Asset type x has
scored successfully finished score
successful.
PM-W-0064 The asset type X Asset type x fail to Click retrain to
analysis failed analysis. analyze. If problem
still exist contact
support, and provide
event data and entity
property data for
further
troubleshooting.
PM-W-0065 The asset type X Asset type x fail to Make sure data
scoring failed score. schema is consistent
with previous build
model. And click
retrain button. If
problem still exist
contact support, and
provide event data
and entity property
data for further
troubleshooting.
PM-W-0066 The uploaded file X is User uploaded file x Check file x and fix
not a valid json file is doesn’t contained incorrect format.
correct JSON format
PM-W-0067 The asset id X scoring Asset id x score failed Make sure the asset
failed contains event data
and that the data
schema is consistent
with the previous
build model. Click the
retrain button. If
problem still exists,
contact support and
provide event data
and entity property
data for further
troubleshooting.
PM-W-0068 The asset id X was Asset id x score
scored successfully successfully

Chapter 5. Troubleshooting 41
42 User Guide
Notices
This information was developed for products and services offered in the US. This
material might be available from IBM in other languages. However, you may be
required to own a copy of the product or product version in that language in order
to access it.

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 grant you
any license to these patents. You can send license inquiries, in writing, to:

IBM Director of Licensing

IBM Corporation
North Castle Drive, MD-NC119
Armonk, NY 10504-1785
US

For license inquiries regarding double-byte character set (DBCS) information,

contact the IBM Intellectual Property Department in your country or send
inquiries, in writing, to:

Intellectual Property Licensing

Legal and Intellectual Property Law
IBM Japan Ltd.
19-21, Nihonbashi-Hakozakicho, Chuo-ku
Tokyo 103-8510, Japan

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 jurisdictions 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 websites are provided for

convenience only and do not in any manner serve as an endorsement of those

© Copyright IBM Corp. 2017 43

 websites. The materials at those websites are not part of the materials for this IBM
product and use of those websites is at your own risk.

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

Licensees of this program who wish to have information about it for the purpose
of enabling: (i) the exchange of information between independently created
programs and other programs (including this one) and (ii) the mutual use of the
information which has been exchanged, should contact:

IBM Director of Licensing

IBM Corporation
North Castle Drive, MD-NC119
Armonk, NY 10504-1785
US

Such information may be available, subject to appropriate terms and conditions,

including in some cases, payment of a fee.

The licensed program described in this document and all licensed material
available for it are provided by IBM under terms of the IBM Customer Agreement,
IBM International Program License Agreement or any equivalent agreement
between us.

The performance data and client examples cited are presented for illustrative
purposes only. Actual performance results may vary depending on specific
configurations and operating conditions.

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 is for planning purposes only. The information herein is subject to
change before the products described become available.

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 actual people or business enterprises 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. The sample

44 User Guide
 programs are provided "AS IS", without warranty of any kind. IBM shall not be
liable for any damages arising out of your use of the sample programs.

Trademarks
IBM, the IBM logo, and ibm.com are trademarks or registered trademarks of
International Business Machines Corp., registered in many jurisdictions worldwide.
Other product and service names might be trademarks of IBM or other companies.
A current list of IBM trademarks is available on the web at "Copyright and
trademark information" at www.ibm.com/legal/copytrade.shtml.

Java and all Java-based trademarks and logos are trademarks or registered
trademarks of Oracle and/or its affiliates.

Linux is a trademark of Linus Torvalds in the United States, other countries, or

both.

Microsoft, Windows, Windows NT, and the Windows logo are trademarks of
Microsoft Corporation in the United States, other countries, or both.

UNIX is a registered trademark of The Open Group in the United States and other
countries.

Terms and conditions for product documentation

Permissions for the use of these publications are granted subject to the following
terms and conditions.

Applicability

These terms and conditions are in addition to any terms of use for the IBM
website.

Personal use

You may reproduce these publications for your personal, noncommercial use
provided that all proprietary notices are preserved. You may not distribute, display
or make derivative work of these publications, or any portion thereof, without the
express consent of IBM.

Commercial use

You may reproduce, distribute and display these publications solely within your
enterprise provided that all proprietary notices are preserved. You may not make
derivative works of these publications, or reproduce, distribute or display these
publications or any portion thereof outside your enterprise, without the express
consent of IBM.

Rights
Except as expressly granted in this permission, no other permissions, licenses or
rights are granted, either express or implied, to the publications or any
information, data, software or other intellectual property contained therein.

Notices 45
 IBM reserves the right to withdraw the permissions granted herein whenever, in its
discretion, the use of the publications is detrimental to its interest or, as
determined by IBM, the above instructions are not being properly followed.

You may not download, export or re-export this information except in full
compliance with all applicable laws and regulations, including all United States
export laws and regulations.

IBM MAKES NO GUARANTEE ABOUT THE CONTENT OF THESE

PUBLICATIONS. THE PUBLICATIONS ARE PROVIDED "AS-IS" AND WITHOUT
WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING
BUT NOT LIMITED TO IMPLIED WARRANTIES OF MERCHANTABILITY,
NON-INFRINGEMENT, AND FITNESS FOR A PARTICULAR PURPOSE.

IBM Online Privacy Statement

IBM Software products, including software as service solutions, (“Software
Offerings”) may use cookies or other technologies to collect product usage
information, to help improve the end user experience, to tailor interactions with
the end user or for other purposes. In many cases no personally identifiable
information is collected by the Software Offerings. Some of our Software Offerings
can help enable you to collect personally identifiable information. If this Software
Offering uses cookies to collect personally identifiable information, specific
information about this offering’s use of cookies is set forth below.

Depending upon the configurations deployed, this Software Offering may use
session and persistent cookies that collect each user’s name, user name, password,
or other personally identifiable information for purposes of session management,
authentication, single sign-on configuration or other usage tracking or functional
purposes. These cookies can be disabled, but disabling them will also likely
eliminate the functionality they enable.

If the configurations deployed for this Software Offering provide you as customer
the ability to collect personally identifiable information from end users via cookies
and other technologies, you should seek your own legal advice about any laws
applicable to such data collection, including any requirements for notice and
consent.

For more information about the use of various technologies, including cookies, for
these purposes, see IBM’s Privacy Policy at http://www.ibm.com/privacy and
IBM's Online Privacy Statement at http://www.ibm.com/privacy/details in the
section entitled “Cookies, Web Beacons and Other Technologies” and the "IBM
Software Products and Software-as-a-Service Privacy Statement" at
http://www.ibm.com/software/info/product-privacy.

46 User Guide
Notices 47
IBM®

Printed in USA