Software Project Management

Durga Prasad Mohapatra

Professor
CSE Deptt.
NIT Rourkela
Project Estimation Techniques
cont…
• A taxonomy of estimating methods
• Size Estimation
 A taxonomy of estimating methods
 Top-down or Bottom-up - activity based, analytical

 Parametric or algorithmic models e.g. function points

 Expert Judgement - just guessing?

 Analogy - case-based, comparative

 Price to win
 Pricing to win
 ‘Price to win’ is setting a target that is likely to win business when
tendering for work
 The project costs whatever the customer can spend on it.
 Advantages:You get the contract
 Disadvantages: Costs do not accurately reflect the work required.
Either:
◦ (1) the customer does not get the desired system or
◦ (2) the customer overpays.
 Pricing to win
 This approach may seem unethical and unbusiness-like….
◦ However, when detailed information is lacking it may be the only
appropriate strategy…
 Which is the most ethical approach?
◦ The project cost is agreed on the basis of an outline proposal and
the development is constrained by that cost
◦ A detailed specification may be negotiated or an evolutionary
approach used for system development
 Project Parameters to be Estimated
 For project planning, we need:
◦ Effort (cost)
◦ Duration

 Hard to estimate effort (or cost) or duration directly from a

problem description.
 Effort and Duration can be measured in terms of project
size (indirect metric)
 Project Parameters to be Estimated
cont …
 Size is a fundamental measure of work
 Based on the estimated size, two parameters are estimated:
◦ Effort Effort
◦ Duration Size
Duration
 Effort is measured in person-months:
◦ One person-month is the effort an individual can typically put in a
month.
 What is size? A Measure of Work…
 Project size is a measure of the problem complexity in terms of
the effort and time required to develop the product.
 Two metrics are popularly used to measure project size:
◦ Source Lines of Code (SLOC)
◦ Function point (FP)
 SLOC is conceptually simple
◦ But, FP is now-a-days favoured over SLOC
◦ Because of the many shortcomings of SLOC.
 Lines of code
 What's a line of code?
◦ Originally proposed when programs were typed on cards
with one line per card;
◦ What happens when statements in Java span several lines
or where there can be several statements on one line?
 What programs should be counted as part of the system?
GUI? Built-in Class?
 Initially software development consisted of only writing
code…
 Lines of Code – Some Terminologies
 LOC ≡ Line of Code
 KLOC ≡
◦ Thousands of LOC
 KSLOC ≡
◦ Thousands of Source LOC
 NCKSLOC ≡
◦ New or Changed KSLOC
 LOC: Few things counter intuitive…
 The lower the level of the language, the more
productive the programmer is:
◦ The same functionality takes more code to implement in a lower-
level language than in a high-level language.

 The more verbose the programmer, the higher

the productivity:
◦ Measures of productivity based on lines of code suggest that
programmers who write verbose code are more productive than
programmers who write compact code.
 Major Shortcomings of SLOC
 Size can vary with coding style.
 Focuses on coding activity alone.
 Correlates poorly with quality and efficiency of code.
 Penalizes higher level programming languages, code reuse,
etc.
 Major Shortcomings of SLOC
 Difficult to estimate at start of a project from problem
description
Only way to estimate is to make a guess…
So not useful for project planning
 Only a code measure
 Programmer-dependent
 Measures lexical/textual complexity only,
 Does not consider structural or logical complexity.
 Further Difficulties with SLOC
 SLOC can become ambiguous due to rapid changes in
programming methodologies, languages, and tools:
- Language advancements
- Automatic source code generation
- Custom software and reuse
- Object-orientation
 Expert Judgment-based Techniques
1. Basic expert judgment

2. Weighted average estimating

3. Consensus estimating

4. Delphi
 Basic Expert Judgment
 One or more experts predict software costs.
◦ Process iterates until some consensus is reached.

 Advantages: Relatively simple estimation method. Can be

accurate if experts have direct experience of similar
systems
 Disadvantages:Very inaccurate if there are no experts
available!
 Basic Expert Judgment Method: Steps
 Coordinator presents each expert with a specification and an estimation
form.
 Coordinator calls a group meeting in which the experts discuss
estimation issues with the coordinator and each other.
 Experts fill out forms anonymously
 Coordinator prepares and distributes a summary of the estimation on an
iteration form.
 Coordinator calls a group meeting, specially focusing on having the
experts discuss points where their estimates varied widely.
 Experts fill out forms, again anonymously, and Steps 4 and 6 are iterated
for as many rounds as appropriate.
Basic Expert Judgement Method cont. …
 An expert is familiar with and knowledgeable about the
application area and the technologies

 Particularly appropriate where existing code is to be

modified.

 Research shows that expert judgement in practice tends to

be based on analogy…
 Stages
 Identify significant features of the current project

 Look at previous project(s) with similar features

 Observe differences between the current and previous

projects

 Find out the possible reasons for error (risk)

 Take measures to reduce uncertainty

 Estimation by Analogy
 The cost of a project is computed by comparing the project
to a similar project in the same application domain:
 Advantages:
◦ May be accurate if project data available and people/tools the same
 Disadvantages:
◦ Impossible if no comparable project has been tackled.
◦ Needs systematically maintained cost database
 Basic Expert Judgement: Cons
 Hard to quantify

 It is hard to document the factors used by the experts or

expert-group.

 Expert may be biased, optimistic, or pessimistic, even though

they have been decreased by the group consensus.

 The expert judgment method always complements the

other cost estimating methods such as algorithmic method.
 Weighted average estimates
•Weighted average estimating is also known as sensitivity analysis estimating.
•Three estimates are obtained rather than one.
• Best case (O = optimistic), worst case (P = pessimistic) and most likely
(M = median).
•This provides a more accurate estimate than when only one estimate is
used.
•These are then used in the following formula:

Estimated effort = (O + 4M + P ) / 6
 Consensus estimating
Steps in conducting a consensus estimating session:
 A briefing is provided to the estimating team on the project.
 Each person is provided with a list of work components to estimate.
 Each person independently estimates O, M and P for each work
component.
 The estimates are written up on the whiteboard.
 Each person discusses the basis and assumptions for their estimates.
 A revised set of estimates is produced.
 Averages for the O, M and P values are calculated.
 These values are used in the formula.
 Delphi Estimation
 A variation of consensus estimation technique
 Team of Experts and a coordinator.
 Experts carry out estimation independently:
◦ mention the rationale behind their estimation.
◦ coordinator notes down any extraordinary rationale:
 circulates the estimation rationale among experts.
 Experts re-estimate.
 Experts never meet each other
◦ to discuss their viewpoints.
 Delphi
 Delphi is an expert survey in two or more "rounds".
 Starting from the second round, a feedback is given (about
the results of previous rounds).
 The same experts assess the same matters once more -
influenced by the opinions of the other experts
 important: anonymity
 Delphi Method: Steps
1. Coordinator presents each expert with a specification & an estimation
form.
2. Coordinator calls a group meeting in which the experts discuss
estimation issues with the coordinator.
3. Experts fill out forms anonymously.
4. Coordinator prepares and distributes a summary of the estimation on
an iteration form.
5. Coordinator calls a group meeting, specially mentioning the noted
rationale where the estimates varied widely.
6. Experts fill out forms, again anonymously, and Steps 4 and 6 are iterated
for as many rounds as appropriate.
 Types of Estimation Techniques
 Though there are many techniques of estimating, they can
broadly be classified into:
◦ Top-down
◦ Bottom-up
 What about:
◦ Algorithmic models?
◦ Expert opinion?
◦ Analogy ?
◦ Price to win?
 Bottom-up versus top-down
 Bottom-up
◦ identify all tasks that have to be done – so quite time-
consuming
◦ use when you have no data about similar past projects
 Top-down
◦ produce overall estimate based on project cost drivers
based on past project data
◦ divide overall estimate between jobs to be done
 Bottom-up estimating
1. Break the project activities into smaller and smaller
components
 Stop when you get to what one person can do in
one/two weeks
2. Estimate costs for the lowest level activities
3. At each higher level calculate estimate by adding
estimates for lower levels
 Top-down estimates
Estimate  Produce overall estimate
overall 100 days using effort driver(s)
project
 Distribute proportions of
overall estimate to
design code test components
30% 30% 40%
i.e. i.e. i.e. 40 days
30 days 30 days
Top-down Example
 Top-Down Estimating: Pros
 It accounts for system-level activities such as integration,
documentation, configuration management, etc.:

◦ Many of these may be ignored in bottom-up estimating

methods.

 It requires minimal project details:

◦ It is usually faster, easier to implement.

 Top-Down Estimating: Cons
 It often does not take into consideration the difficult low-
level problems:

◦ Tends to underestimate and overlook complexities of low-

level components.

 It provides no detailed basis for justifying decisions or

estimates.
 Bottom-up Estimating: Pro
 It permits the software group to estimate in an almost
traditional fashion:
◦ Each group estimates components for which it has
adequate experience.
 It is more stable because the estimation errors in the various
components more or less balance out.
 Bottom-up Estimating: Con
 It may overlook many of the system-level costs:
◦ Integration,
◦ configuration management,
◦ quality assurance, etc.
 It may be inaccurate because the necessary information may
not be available in the early phase.
 It tends to be more time-consuming.
 Criteria for a Good Algorithmic Model
 Defined—clear procedure
 Accurate
 Objective—avoids subjective factors
 Results understandable
 Stable— Results valid for a wide range of parameter values
 Easy to Use
 Causal—future data not required
 Algorithmic Models
 For project planning, we need:
◦ Effort (cost)
◦ Duration
 Hard to estimate effort (or cost) or duration directly from
a problem description.
 Effort and Duration can be measured in terms of certain
project characteristics that correlate with it:
◦ Called size
 Software Size
 What exactly is the size of a software project?
◦ How do you measure it?
 Any characteristic of software that is easily measured and
correlates with effort.
◦ SLOC
◦ Function point
 Algorithmic/Parametric models
 COCOMO (lines of code) and function points examples of
these
 A Problem with LOC based models (COCOMO etc):

guess algorithm estimate

but what is desired is

system algorithm estimate

characteristic
 Algorithmic Methods: Pro
 It is able to generate repeatable estimations.

 It is easy to modify input data, refine and customise formulas.

 It is efficient and able to support a family of estimations or a

sensitivity analysis.

 It can be meaningfully calibrated to previous experience.

 Algorithmic Methods: Cons
 It lacks means to deal with exceptional conditions, such as:

◦ exceptional teamwork, exceptional match between skill-

levels and tasks, etc.
 Poor sizing inputs and inaccurate cost driver rating will
result in inaccurate estimation.

 Some factors such as experience cannot be easily quantified.

 Model Calibration
 Many models are developed for specific situations and are, by
definition, calibrated to that situation.
 Such models usually are not useful outside of their particular
environment.
◦ Calibration is needed to increase the accuracy of one of
these general models.
◦ Calibration is in a sense customising a generic model.
 Items that can be calibrated in a model include:
◦ product types, operating environments, labour rates and
factors, various relationships between functional cost
items, etc.
 Some recommendations
 Do not depend on a single cost or schedule estimate.
 Use several estimating techniques or cost models:
◦ Compare the results and determine the reasons for any
large variations.
 Document the assumptions made when making the
estimates.
 Monitor the project to detect when assumptions that turn
out to be wrong jeopardize the accuracy of the estimate.
 Improve software process:
◦ Maintain a historical database
 Simplistic Model
 estimated effort = (system size) / productivity
 As example:
system size = lines of code What is wrong with the
simplistic model?
productivity = lines of code per day
 productivity = (system size) / effort
◦ based on past projects
 Example 1
Consider a transaction project of 38,000 lines of code, what is
the shortest time it will take to develop? Productivity is about
400 SLOC/staff month
Effort = (productivity)-1 (size)
= (1/.400 KSLOC/SM) (38 KSLOC)
= 2.5 (38) ≈ 100 SM
Min time = .75 T= (.75)(2.5)(SM)1/3
≈ 1.875(100)1/3
≈ 1.875 x 4.63 ≈ 9 months
 Summary
 Discussed the Price to win estimation method.
 Presented Size estimation.
 Also, discussed Expert Judgement – based estimation
techniques such as
 Basic judgement
 Weighted average estimating
 Consensus estimating
 Delphi Technique
 Explained the Analogy based estimation.
 Discussed top-down and bottom-up estimation techniques
 Discussed the algorithmic/parametric models for project
estimation.
References :
1. B. Hughes, M. Cotterell, R. Mall, Software Project Management, Sixth Edition,
McGraw Hill Education (India) Pvt. Ltd., 2018.
2. R. Mall, Fundamentals of Software Engineering, Fifth Edition, PHI Learning Pvt.
Ltd., 2018.
Thank you