Software Project Management

W5HH of Project Management

Boehm suggests an approach (W5HH) that addresses project
objectives, milestones and schedules, responsibilities, management
and technical approaches, and required resources
Why is the system being developed?
Enables all parties to assess the validity of business reasons for the software
work. In another words - does the business purpose justify the expenditure of
people, time, and money?

What will be done?

The answers to these questions help the team to establish a project schedule by
identifying key project tasks and the milestones that are required by the
customer

When will it be accomplished?

Project schedule to achieve milestone
Unit 3 – Managing Software Projects 1
W5HH of Project Management Cont.
Who is responsible?
Role and responsibility of each member

Where are they organizationally located?

Customer, end user and other stakeholders also have responsibility

How will the job be done technically and managerially?

Management and technical strategy must be defined

How much of each resource is needed?

Develop estimation

W5HH
It is applicable regardless of size or complexity of
software project
Unit 3 – Managing Software Projects 2
Terminologies
 Measure  Indirect Metrics
 It provides a quantitative indication of the  Aspects that are not immediately quantifiable
extent (range), amount, dimension, capacity or  Ex., Functionality, Quality, Reliability
size of some attributes of a product or process
 Ex., the number of uncovered errors  Indicators
 It is a metric or combination of metrics that
 Metrics provides insight into the software process, project
 It is a quantitative measure of the degree or the product itself
(limit) to which a system, component or  It enables the project manager or software
process possesses (obtain) a given attribute engineers to adjust the process, the project or the
 It relates individual measures in some way product to make things better
 Ex., number of errors found per review  Ex., Product Size (analysis and specification
metrics) is an indicator of increased coding,
 Direct Metrics integration and testing effort
 Immediately measurable attributes
 Faults
 Ex., Line of Code (LOC), Execution Speed,
Defects Reported  Errors - Faults found by the practitioners during
software development
 Defects - Faults found by the customers after
release
Unit 3 – Managing Software Projects 3
Why to Measure Software?
1 To determine (to define) quality of a product or process.
2 To predict qualities of a product or process.
3 To improve quality of a product or process.
Metric Classification Base
 Process
 Specifies activities related to production of software.
 Specifies the abstract set of activities that should be performed to go from user needs to final product.
 Project
 Software development work in which a software process is used
 The actual act of executing the activities for some specific user needs
 Product
 The outcomes of a software project
 All the outputs that are produced while the activities are being executed
Unit 3 – Managing Software Projects 4
Process Metrics
 Process Metrics are an invaluable tool for We measure the effectiveness of a process by
companies to monitor, evaluate and improve deriving a set of metrics based on outcomes
their operational performance across the of the process such as,
enterprise
Errors uncovered before release of the software
 They are used for making strategic decisions
Defects delivered to and reported by the end users
 Process Metrics are collected across all
projects and over long periods of time Work products delivered
 Their intent is to provide a set of process Human effort expanded
indicators that lead to long-term software
process improvement Calendar time expanded

Conformance to the schedule

Ex., Defect Removal Efficiency (DRE) metric
Relationship between errors (E) and defects (D) Time and effort to complete each generic activity
The ideal is a DRE of 1
DRE = E / ( E + D )
Unit 3 – Managing Software Projects 5
Project Metrics
 Project metrics enable a software project manager to,
 Assess the status of an ongoing project
 Track potential risks
 Uncover problem areas before their status becomes critical
 Adjust workflow or tasks
 Evaluate the project team’s ability to control quality of software work products
 Many of the same metrics are used in both the process and project domain
 Project metrics are used for making tactical (smart) decisions
 They are used to adapt project workflow and technical activities
 Project metrics are used to
 Minimize the development schedule by making the adjustments necessary to avoid delays and
mitigate (to reduce) potential (probable) problems and risks
 Assess (evaluates) product quality on an ongoing basis and guides to modify the technical
approach to improve quality
Unit 3 – Managing Software Projects 6
Product Metrics
 Product metrics help software engineers to gain insight into the design and construction of
the software they build
 By focusing on specific, measurable attributes of software engineering work products
 Product metrics provide a basis from which analysis, design, coding and testing can be
conducted more objectively and assessed more quantitatively
 Ex., Code Complexity Metric

Unit 3 – Managing Software Projects 7

Types of Measures
Categories of Software Measurement Software Measurement
Direct measures of the Indirect measures of the Metrics for Software
Software process Software product Cost and Effort estimations
Ex., cost, effort, etc. Ex. functionality, quality, Size Oriented Metrics
complexity, efficiency,
Software product
reliability, etc. Function Oriented Metrics
Ex., lines of code produced,
execution speed, Object Oriented Metrics
defects reported, etc.
Use Case Oriented Metrics

Unit 3 – Managing Software Projects 8

Size-Oriented Metrics
 Derived by normalizing (standardizing) quality and/or productivity measures by considering
the size of the software produced
 Thousand lines of code (KLOC) are often chosen as the normalization value
A set of simple size-oriented metrics can  Size-oriented metrics are not universally accepted
be developed for each project as the best way to measure the software process
Errors per KLOC (thousand lines of code)
Defects per KLOC $ per KLOC Opponents argue that KLOC measurements
Pages of documentation per KLOC Are dependent on the programming language
In addition, other interesting metrics can Penalize well-designed but short programs
be computed, like Cannot easily accommodate nonprocedural languages
Errors per person-month Require a level of detail that may be difficult to achieve
KLOC per person-month
$ per page of documentation
Unit 3 – Managing Software Projects 9
Function Oriented Metrics
 Function-oriented metrics use a measure of the functionality delivered by the application as
a normalization value
 Most widely used metric of this type is the Function Point
 FP = Count Total * [0.65 + 0.01 * Sum (Value Adjustment Factors)]
 Function Point values on past projects can be used to compute,
 for example, the average number of lines of code per function point
 Advantages
 FP is programming language independent
 FP is based on data that are more likely to be known in the early stages of a project, making it more
attractive as an estimation approach
 Disadvantages
 FP requires some “sleight of hand” because the computation is based on subjective data
 Counts of the information domain can be difficult to collect
 FP has no direct physical meaning, it’s just a number

Unit 3 – Managing Software Projects 10

Object-Oriented Metrics Use Case Oriented Metrics
 Conventional software project metrics  Like FP, the use case is defined early in the
(LOC or FP) can be used to estimate software process, allowing it to be used for
object-oriented software projects estimation before significant (valuable)
 However, these metrics do not provide modeling and construction activities are
enough granularity (detailing) for the initiated
schedule and effort adjustments that are  Use cases describe (indirectly, at least) user-
required as you iterate through an visible functions and features that are basic
evolutionary or incremental process requirements for a system
 Lorenz and Kidd suggest the following set  The use case is independent of programming
of metrics for OO projects language, because use cases can be created at
 Number of scenario scripts vastly different levels of abstraction, there is no
 Number of key classes (the highly standard “size” for a use case
independent components)
 Without a standard measure of what a use case
 Number of support classes
is, its application as a normalization measure is
suspect (doubtful).
 Ex., effort expended / use case
Unit 3 – Managing Software Projects 11
Function Point Metrics
 The function point (FP) metric can User / Event
be used effectively as a means for
measuring the functionality Other
delivered by a system User / Event external Applications
 Using historical data, the FP metric inquiries (EQs)
can be used to
external
 Estimate the cost or effort required to external interface
inputs (EIs)
design, code, and test the software files (EIFs)
 Predict the number of errors that will
be encountered during testing
external
 Forecast the number of components outputs (EOs) Internal
and/or the number of projected
source lines in the implemented Logic Files
system

Function / Application

Unit 3 – Managing Software Projects 12

Function Point Components Cont.
Information domain values (components) are defined in the following manner
 Number of external inputs (EIs)
 input data originates from a user or is transmitted from another application
 Number of external outputs (EOs)
 external output is derived data within the application that provides information to the user
 output refers to reports, screens, error messages, etc.
 Number of external inquiries (EQs)
 external inquiry is defined as an online input that results in the generation of some immediate software
response in the form of an online output
 Number of internal logical files (ILFs)
 internal logical file is a logical grouping of data that resides within the application’s boundary and is
maintained via external inputs
 Number of external interface files (EIFs)
 external interface file is a logical grouping of data that resides external to the application but provides
information that may be of use to another application

Unit 3 – Managing Software Projects 13

Compute Function Points
Value Adjustment Factors
FP = Count Total * [ 0.65 + 0.01 * ∑(Fi) ]
• F1. Data Communication
Count Total is the sum of all FP entries • F2. Distributed Data Processing
• F3. Performance
Fi (i=1 to 14) are complexity value adjustment factors (VAF).
• F4. Heavily Used Configuration
Value adjustment factors are used to provide an indication of problem
• F5. Transaction Role
complexity • F6. Online Data Entry
• F7. End-User Efficiency
• F8. Online Update
• F9. Complex Processing
• F10. Reusability
• F11. Installation Ease
• F12. Operational Ease
• F13. Multiple Sites
• F14. Facilitate Change

Unit 3 – Managing Software Projects 14

Function Point Calculation Example
Used Adjustment
Factors and assumed
values are,
F09. Complex internal
processing = 3
F10. Code to be reusable =
2
F03. High performance = 4

FP = Count Total * [ 0.65 + 0.01 * ∑(Fi) ] F13. Multiple sites = 3

F02. Distributed
FP = [50]* [0.65 + 0.01 * 17]
processing = 5
FP = [50]* [0.65 + 0.17]
Project Adjustment
FP = [50]* [0.82] = 41 Factor (VAF) = 17
Unit 3 – Managing Software Projects 15
Function Point Calculation Example 2
Study of requirement specification
for a project has produced following
results 7 28
10 40
6 18
Need for 7 inputs, 10 outputs, 6
inquiries, 17 files and 4 external 17 119
interfaces 4 28
233
Input and external interface
function point attributes are of
Value adjustment factors (VAF) = 32 given
average complexity and all other
function points attributes are of low
complexity FP = Count Total * [ 0.65 + 0.01 * ∑(Fi) ]
= 233 * [ 0.65 + 0.01 * 32]
Determine adjusted function points
assuming complexity adjustment = 233 * 0.97 = 226.01
value is 32.
Unit 3 – Managing Software Projects 16
Software Project Estimation
To achieve reliable cost and effort estimates, several options arise:
 Delay estimation until late in the project (obviously, we can
achieve 100 percent accurate estimates after the project is
complete!)
 Base estimates on similar projects that have already been
completed
 Use relatively simple decomposition techniques to generate
project cost and effort estimates
It can be transformed
 Use one or more empirical models for software cost and effort
from a black art to a estimation.
series of systematic
steps that provide
estimates with
acceptable risk

Unit 3 – Managing Software Projects 17

Software Project Decomposing
 Software project estimation is a form of problem solving and in most cases, the problem to be
solved is too complex to be considered in one piece
 For this reason, decomposing the problem, re-characterizing it as a set of smaller problems is
required
 Before an estimate can be made, the project planner must understand the scope of the
software to be built and must generate an estimate of its “size”

Decomposition Techniques

1. Software Sizing 3. Process based Estimation

2. Problem based Estimation 4. Estimation with Use-cases
LOC (Lines of Code) based,
FP (Function Point) based

Unit 3 – Managing Software Projects 18

Software Sizing
Putnam and Myers suggest four different approaches to the sizing problem

 “Fuzzy logic” sizing

 This approach uses the approximate reasoning techniques that are the cornerstone of fuzzy logic.
 Function Point sizing
 The planner develops estimates of the information domain characteristics
 Standard Component sizing
 Estimate the number of occurrences of each standard component
 Use historical project data to determine the delivered LOC size per standard component.
 Change sizing
 Used when changes are being made to existing software
 Estimate the number and type of modifications that must be accomplished
 An effort ratio is then used to estimate each type of change and the size of the change

Unit 3 – Managing Software Projects 19

Problem Based Estimation
 Start with a bounded statement of scope
 Decompose the software into problem functions that can each be estimated individually
 Compute an LOC or FP value for each function
 Derive cost or effort estimates by applying the LOC or FP values to your baseline productivity
metrics
 Ex., LOC/person-month or FP/person-month
 Combine function estimates to produce an overall estimate for the entire project
 In general, the LOC/pm and FP/pm metrics should be computed by project domain
 Important factors are team size, application area and complexity

Unit 3 – Managing Software Projects 20

Problem Based Estimation Cont.
 LOC and FP estimation differ in the level of detail required for decomposition with each value
 For LOC, decomposition of functions is essential and should go into considerable detail (the more detail, the
more accurate the estimate)
 For FP, decomposition occurs for the five information domain characteristics and the 14 adjustment factors
▪ External Inputs, External Outputs, External Inquiries, Internal Logical Files, External Interface Files
 For both approaches, the planner uses lessons learned to estimate,
 An optimistic (Sopt), most likely (Sm), and pessimistic (Spess) estimates Size (S) value for each function or
count
 Then the expected Size value S is computed as
▪ S = (Sopt + 4 Sm + Spess)/6

Historical LOC or FP data is then compared to S in order to cross-check it.

Unit 3 – Managing Software Projects 21

Process Based Estimation
 This is one of the most commonly used technique
Process-based estimation is
obtained from “process  Identify the set of functions that the software needs to perform as
obtained from the project scope
framework”
 Identify the series of framework activities that need to be performed
for each function
Framework Activities
Frame
 Estimate the effort (in person months) that will be required to
accomplish each software process activity for each function
Effort required to  Apply average labor rates (i.e., cost/unit effort) to the effort
Application
Functions

accomplish estimated for each process activity

each framework  Compute the total cost and effort for each function and each
activity for each framework activity.
application  Compare the resulting values to those obtained by way of the LOC
function and FP estimates
 If both sets of estimates agree, then your numbers are highly reliable
 Otherwise, conduct further investigation and analysis concerning the
function and activity breakdown
Unit 3 – Managing Software Projects 22
Estimation with Use Cases
Developing an estimation approach with use  Before use cases can be used for estimation,
cases is problematic for the following reasons:  the level within the structural hierarchy is
established,
 Use cases are described using many different
 the average length (in pages) of each use
formats and styles—there is no standard form. case is determined,
 Use cases represent an external view (the  the type of software (e.g., real-time,
user’s view) of the software and can therefore business, engineering/scientific, WebApp,
be written at many different levels of embedded) is defined, and
abstraction  a rough architecture for the system is
considered
 Use cases do not address the complexity of
the functions and features that are described  Once these characteristics are established,
 empirical data may be used to establish the
 Use cases can describe complex behavior (Ex., estimated number of LOC or FP per use
interactions) that involve many functions and case (for each level of the hierarchy).
features  Historical data are then used to compute the
 Although a number of investigators have effort required to develop the system.
considered use cases as an estimation input.
Unit 3 – Managing Software Projects 23
Empirical Estimation Models
Source Lines of Code (SLOC)

 The project size helps to determine the resources, effort,

and duration of the project.
 SLOC is defined as the Source Lines of Code that are
delivered as part of the product
 The effort spent on creating the SLOC is expressed in
relation to thousand lines of code (KLOC)
Source Lines of Code (SLOC)  This technique includes the calculation of Lines of Code,
Documentation of Pages, Inputs, Outputs, and Components
Function Point (FP) of a software program
Constructive Cost Model  The SLOC technique is language-dependent
(COCOMO)  The effort required to calculate SLOC may not be the same
for all languages

Unit 3 – Managing Software Projects 24

Software Development Project Based on the development complexity

Software Development Project Classification

Organic Semidetached Embedded

Application programs Utility programs System programs

e.g. data processing programs e.g Compilers, linkers e.g OS real-time systems
A development project can be A development project can be A development project is
considered of organic type, if considered of semidetached type, considered to be of embedded
the project deals with if the development consists of a type, if the software being
developing a well understood mixture of experienced & developed is strongly coupled
application program, the size inexperienced staff. Team to complex hardware, or if the
of the development team is members may have limited strict regulations on the
reasonably small, and the team experience on related systems but operational procedures exist
members are experienced in may be unfamiliar with some
developing similar types of aspects of the system being
projects developed.
Unit 3 – Managing Software Projects 25
 Software Development Project Cont.

Not Tight Dead Line

Little Innovation
Project Development
Model Size Nature of Project Environment
Organic Typically Small Size Project, Experienced developers in
Familiar &
2-50 the familiar environment, e.g. Payroll,
In-house
KLOC Inventory projects etc.
Semi Typically Medium Size Project, Medium Size Team,

Medium

Medium
Detached 50-300 Average Previous Experience, e.g. Utility Medium
KLOC Systems like Compilers, Database Systems,
editors etc.
Complex hardware &
Embedded Typically Large Project, Real Time Systems, Complex

Significant

Tight
Required
Over 300 interfaces, very little previous Experience. E.g. customer Interfaces
KLOC ATMs, Air Traffic Controls

Unit 3 – Managing Software Projects 26

COCOMO Model Basic COCOMO Model
The basic COCOMO model gives an approximate estimate of the project parameters
COCOMO (Constructive
Cost Estimation Model) The basic COCOMO estimation model is given by the following expressions
was proposed by 𝑬𝒇𝒇𝒐𝒓𝒕 = 𝑎1 + 𝐾𝐿𝑂𝐶 𝑎2
𝑃𝑀 𝑻𝒅𝒆𝒗 = 𝑏1 × 𝐸𝑓𝑓𝑜𝑟𝑡 𝑏2
𝑀𝑜𝑛𝑡ℎ𝑠
Boehm
• KLOC is the estimated size of the software product expressed in Kilo Lines of Code
• a1, a2, b1, b2 are constants for each category of software products,
According to Boehm, • Tdev is the estimated time to develop the software, expressed in months,
software cost estimation • Effort is the total effort required to develop the software product, expressed in
person months (PMs).
should be done through
three stages: Project a1 a2 b1 b2
Basic COCOMO Organic 2.4 1.05 2.5 0.38
Semidetached 3.0 1.12 2.5 0.35
Intermediate COCOMO
Embedded 3.6 1.20 2.5 0.32
Complete COCOMO
Unit 3 – Managing Software Projects 27
Basic COCOMO Model Cont.
 The effort estimation is expressed in units of person-months (PM)
 It is the area under the person-month plot (as shown in fig.)
 An effort of 100 PM
 does not imply that 100 persons should work for 1 month
 does not imply that 1 person should be employed for 100 months
 it denotes the area under the person-month curve (fig.)
 Every line of source text should be calculated as one LOC irrespective of the actual number of
instructions on that line
 If a single instruction spans several lines (say n lines), it is considered to be nLOC
 The values of a1, a2, b1, b2 for different categories of products (i.e. organic, semidetached, and
embedded) as given by Boehm
 He derived the expressions by examining historical data collected from a large number of
actual projects

Unit 3 – Managing Software Projects 28

Basic COCOMO Model Cont.
 Insight into the basic COCOMO model can be obtained by plotting the
estimated characteristics for different software sizes
 Fig.1 shows a plot of estimated effort versus product size
 From fig. we can observe that the effort is somewhat superlinear in the
size of the software product
 The effort required to develop a product increases very rapidly with
project size Fig. 1
 The development time versus the product size in KLOC is plotted in fig. 2
 From fig., it can be observed that the development time is a sublinear
function of the size of the product
 i.e. when the size of the product increases by two times, the time to
develop the product does not double but rises moderately
Fig. 2
 From fig., it can be observed that the development time is roughly the
same for all the three categories of products
Unit 3 – Managing Software Projects 29
 Basic COCOMO Model Cont.
 Effort and the duration estimations obtained using the COCOMO model are called as nominal
effort estimate and nominal duration estimate
 The term nominal implies that
 if anyone tries to complete the project in a time shorter than the estimated duration, then the cost will
increase drastically
 But, if anyone completes the project over a longer period than the estimated, then there is almost no
decrease in the estimated cost value

Example: Assume that the size of an organic type software product has been estimated to be 32,000 lines of
source code. Assume that the average salary of software engineers be Rs. 15,000/- per month. Determine the
effort required to develop the software product and the nominal development time
𝑬𝒇𝒇𝒐𝒓𝒕 = 𝑎1 + 𝐾𝐿𝑂𝐶 𝑎2 𝑃𝑀 𝑻𝒅𝒆𝒗 = 𝑏1 × 𝐸𝑓𝑓𝑜𝑟𝑡 𝑏2 𝑀𝑜𝑛𝑡ℎ𝑠
= 2.4 + 32 1.05 𝑃𝑀 = 2.5 × 91 0.38 𝑀𝑜𝑛𝑡ℎ𝑠
= 91 𝑃𝑀 = 14 𝑀𝑜𝑛𝑡ℎ𝑠
Cost required to develop the product = 14 x 15000 = Rs. 2,10,000/-
Unit 3 – Managing Software Projects 30
Intermediate COCOMO model
 The basic COCOMO model assumes that effort and development time are functions of the
product size alone
 However, a host of other project parameters besides the product size affect the effort required
to develop the product as well as the development time
 Therefore, in order to obtain an accurate estimation of the effort and project duration, the
effect of all relevant parameters must be considered
 The intermediate COCOMO model recognizes this fact and refines the initial estimate obtained
using the basic COCOMO expressions by using a set of 15 cost drivers (multipliers) based on
various attributes of software development
 For example, if modern programming practices are used, the initial estimates are scaled downward by
multiplication with a cost driver having a value less than 1
 It requires the project manager to rate these 15 different parameters for a particular project on
a scale of one to three.
 Then, depending on these ratings, appropriate cost driver values which should be multiplied
with the initial estimate obtained using the basic COCOMO.
Unit 3 – Managing Software Projects 31
Intermediate COCOMO model Cont.
The cost drivers can be classified as being attributes of the following items
 Product: The characteristics of the product that are considered include the inherent complexity
of the product, reliability requirements of the product, etc.
 Computer: Characteristics of the computer that are considered include the execution speed
required, storage space required etc.
 Personnel: The attributes of development personnel that are considered include the experience
level of personnel, programming capability, analysis capability, etc.
 Development Environment: Development environment attributes capture the development
facilities available to the developers. An important parameter that is considered is the
sophistication of the automation (CASE) tools used for software development

Unit 3 – Managing Software Projects 32

Intermediate COCOMO model Cont.
Cost Drivers Very Low Low Nominal High Very High
Project A is to be a 32,000 DSI semi-
Product
detached software. It is in a mission
Required Software Reliability 0.75 0.88 1.00 1.15 1.40
critical area, so the reliability is high.
Size of Application Database 0.94 1.00 1.08 1.16
Complexity of The Product 0.70 0.85 1.00 1.15 1.30
Then we can estimate:
Computer
Runtime Performance Constraints 1.00 1.11 1.30 Effort = 1.15*3.0*(32)1.12
Memory Constraints 1.00 1.06 1.21 = 167 man-months
Volatility of the virtual machine environment 0.87 1.00 1.15 1.30
Required turnabout time 0.94 1.00 1.07 1.15 Schedule = 2.5*(167)0.35
Personnel = 15 months
Analyst capability 1.46 1.19 1.00 0.86 0.71
Applications experience 1.29 1.13 1.00 0.91 0.82
Productivity
Software engineer capability 1.42 1.17 1.00 0.86 0.70
Virtual machine experience 1.21 1.10 1.00 0.90
= 32,000 DSI/167 MM
Programming language experience 1.14 1.07 1.00 0.95 = 192 DSI/MM
Development Environment
Application of software engineering methods 1.24 1.10 1.00 0.91 0.82 Average Staffing
Use of software tools 1.24 1.10 1.00 0.91 0.83 = 167 MM/15 months
Required development schedule 1.23 1.08 1.00 1.04 1.10 = 11 FSP
Unit 3 – Managing Software Projects 33
Complete COCOMO model
 A major shortcoming of both the basic and intermediate COCOMO models is that they
consider a software product as a single homogeneous entity
 Most large systems are made up several smaller sub-systems
 These sub-systems may have widely different characteristics
 E.g., some sub-systems may be considered as organic type, some semidetached, and some embedded
 Also, for some subsystems the reliability requirements may be high, for some the development team might
have no previous experience of similar development etc.
 The complete COCOMO model considers these differences in characteristics of the
subsystems and estimates the effort and development time as the sum of the estimates for
the individual subsystems
 The cost of each subsystem is estimated separately
 This approach reduces the margin of error in the final estimate

Unit 3 – Managing Software Projects 34