Cost Estimation

1
Cost Estimation

• I've got Bad News and Bad News!

2
Cost Estimation

• I've got Bad News and Bad News!
• Accurate estimation of the cost to develop
  software projects is NEARLY IMPOSSIBLE.

3
Cost Estimation

• I've got Bad News and Bad News!
• Accurate estimation of the cost to develop
  software projects is NEARLY IMPOSSIBLE.
• NEARLY EVERY project manager expects you to come
  up with an accurate estimate of cost.

4
Some Observations

• The only time you can be 100 accurate in the
  cost of the project is after the project is
  completely finished.

5
Some Observations

• The only time you can be 100 accurate in the
  cost of the project is after the project is
  completely finished.
• Most software projects are never completely
  finished.

6
Some Observations

• The only time you can be 100 accurate in the
  cost of the project is after the project is
  completely finished.
• Most software projects are never completely
  finished.
• Experience helps.

7
Cont.

• If computer efficiency has increased multiple
  orders of magnitude, and size of memory has
  increased multiple orders of magnitude and disk
  space has increased multiple orders of magnitude
  and the object oriented paradigm has permitted
  software reuse,
• WHY HASN'T THE COST OF DEVELOPING SOFTWARE
  DRAMATICALLY BEEN REDUCDE??

8
Some Answers

• Software complexity has increased by many orders
  of magnitude.
• Salaries continue to increase and most of the
  cost to develop software is people oriented.

9
Observations

• A majority of the cost to develop software is the
  human effort involved.
• So, cost estimation is really effort estimation.
• It appears intuitive that the effort is
  proportional to the complexity of the project.

10
Cont.

• Complexity is relative. It's not always related
  just to size.
• A small project that is done by new software
  engineers unfamiliar with the company or the
  project is complex.
• A large project that is done by experienced
  software engineers familiar with previous (almost
  identical) projects is not as complex.

11
So what to do???

• Begin day one to establish a history of your
  development effort at your company.
• Keep records of the unexpected cost factors that
  come up.
• Customer changes their mind about a functional
  requirement.
• Environment changes (Windows NT -gt Windows XP)
• Additional functionality is requested.
• Plan to become an expert in estimating your
  effort in software development.

12
Causes of Inaccurate Estimates

• Frequent requests for changes by users.
• Overlooked tasks
• User's lack of understanding of their own
  requirements.
• Insufficient analysis when developing an estimate
  (pressure to get it done NOW!).
• Lack of coordination of systems development,
  technical services, operations, data
  administration, and other functions during
  development.
• Lack of an adequate method or guidelines for
  estimating.

13
Aspects of the projectthat are key estimate
influences

• Complexity of the proposed application system.
• Required integration with existing system.
• Complexity of the programs in the system.
• Size of the system expressed as number of
  functions or programs.
• Capabilities of the project team members.
• Project team's experience with the application.

14
Cont.

• Anticipated frequency or extent of potential
  changes in user requirements.
• Project team's experience with the programming
  language or hardware.
• Database management system.
• Number of project team members.
• Extent of programming or documentation standards.
• Availability or tools such as application
  generators. (Familiarity of the team)

15
Accuracy of Estimates

• (Show overhead slide from Boehm.)

16
Estimation Techniques

• Expert Judgement.
• Algorithmic Methods.
• Machine-Learning Methods.

17
Expert Judgement

• The accuracy of the prediction is based on the
  competence, experience, objectivity, and
  perception of the estimator.
• An educated guess
• Based on a top-down or bottom-up analysis of
  what is needed.
• Analogies are used to estimate effort.

18
Cont.

• The analogy process can be formalized by asking
  several experts to make three predictions
• A pessimistic one (x).
• A most likely guess (y).
• An optimistic one (z).
• Then these are combined as the mean of the beta
  probability distribution.
• ( x 4y z ) / 6

19
Delphi

• Experts are asked to make individual predictions
  secretly, based on their expertise and using
  whatever process they choose.
• Then the average estimate is calculated and
  presented to the group.
• Each expert has the opportunity to revise his or
  her estimate, if desired.
• The process is repeated until no expert wants to
  revise.

20
Delphi cont.

• Sometimes a discussion is allowed before the
  revised estimates are calculated.
• Another variation is to allow the extreme
  estimators to make a case for their decisions
  anonymously.

21
Wolverton Model Cost Matrix
22
Other conflicts

• Adding people to a late project simply makes it
  later.
• Two people cannot produce code twice as fast as
  one person.
• A model that may work for one organization may
  not apply to another.
• Most expert judgement techniques are simplistic,
  neglecting a large number of factors
  So.......

23
Algorithmic Methods

• Empirical models to express the relationship
  between effort and the factors that influence it.
• Models are usually described using equations
• E (a bS ) m(X)
• S estimated size of the system
• a,b,c constants
• X is a vector of cost factors, m is an adjustment
  multiplier based on X

Ac
24
One of the firstby Walston and Felix

• 0.91
• E 5.25 S
• For 60 projects ranging in size from 4000 to
  467,000 lines of code, written in 28 different
  high-level languages on 66 computers and
  representing from 12 to 11,758 person-months of
  effort.