2.6 Empirical estimation models & The make-buy decision.ppt

INTRODUCTION
 Estimation models uses empirically derived formulas to
predict the estimates.
 Here we conduct a study on some completed projects.
 From those observation we form some statistical
formulas.
 We can use this formulas to estimate the cost of other
projects.
 The structure of empirical estimation models is a
formula, derived from data collected from past software
projects.

THE STRUCTURE OF ESTIMATION
MODELS
 The overall structure of such models takes the form –
 Where,
 A, B and C – empirically derived constants
 E – effort in person-months
 ev – estimation variable (LOC or FP)
C
v
e
B
A
E )
(
*



THE COCOMO II MODEL
 Stands for COnstructive COst MOdel
 Introduced by Barry Boehm in 1981
 Became one of the well-known and widely-used
estimation models in the industry
 It has evolved into a more comprehensive estimation
model called COCOMO II, with reuse property.

THE COCOMO II MODEL
 COCOMO II is actually a 3 level hierarchy of estimation
models that address the following areas:
 Application composition model - Used during the
early stages of software engineering process.
 Early design stage model - Used once requirements
have been stabilized and basic software architecture
has been established.
 Post-architecture-stage model - Used during the
construction of the software.

THE COCOMO II MODEL
 The COCOMO II models require sizing information
 Three different sizing options are available as part of
the model hierarchy:
 Object points (OP)
 Function points (FP)
 Lines of source code (LOC)
 The COCOMO II application model uses object points
(OP)

THE COCOMO II MODEL
 Object Point is an indirect software measure that is
computed using counts of no. of –
 Screens
 Reports
 Components likely to be required to build the
application
 Each of the above object instance is classified into one
of the three complexity levels – simple, medium or
difficult

THE COCOMO II MODEL
 The object point count is then determined by
multiplying the original no. of object instances by the
weighting factor and summing to obtain a total object
point count
 NOP = (object points) x [(1-%reuse)/100]
 Where, NOP – new object points
 PROD = NOP / OP person-month
 Where, PROD – productivity rate
Estimated project effort = NOP / PROD

Object Point Estimation Procedure

Example
The system includes:
 6 screens: 2 simple + 3 medium + 1 difficult
 3 reports: 2 medium + 1 difficult
 2 3GL components
 30 % of the objects could be supplied from
previously developed components
 Productivity is high.
Calculate estimated effort.

THE SOFTWARE EQUATION
 The software equation is a dynamic multivariable model
that assumes a specific distribution of effort over the life
of a software development project
 The model has been derived from productivity data
collected for over 4000 contemporary software projects.
 An estimation model form –
4
3
333
.
0
1
*
*
t
P
B
LOC
E 

THE SOFTWARE EQUATION
 Where,
 E = effort in person-months/years
 t = project duration in months/years
 B = special skills factor
 P = productivity parameter that reflects overall
process and management practices
 Typical values –
 P = 2000 (real-time embedded software)
 P = 10,000 (telecommunication & system software)
 P = 28,000 (business applications)

2.6 Empirical estimation models & The make-buy decision.ppt

INTRODUCTION
 It is often more cost effective to acquire rather than to
develop software
 Managers have many acquisition options-
 Software may be purchased (or licensed) off the
shelf
 “Full-experience” or “partial-experience” software
components may be acquired and integrated to meet
specific needs
 Software may be custom built by an outside
contractor to meet the purchaser’s specifications

INTRODUCTION
 The make/buy decision can be made based on the
following conditions
 Will the software product be available sooner than
internally developed software?
 Will the cost of acquisition plus the cost of
customization be less than the cost of developing the
software internally?
 Will the cost of outside support (e.g., a maintenance
contract) be less than the cost of internal support?

CREATING A DECISION TREE
 Consider a decision tree for a software based system X
 In this case, the software engineering organization can
 Build system X from scratch
 Reuse existing partial-experience components
construct the system
 Buy an available software product and modify it to
meet local needs
 Contract the software development to an outside
vendor

COMPUTING EXPECTED COST
 Expected cost =
∑(path probability)i x (estimated path cost)i
 Where, i – decision tree path
 For the build path,
Expected costbuild = 0.30 ($380K) + 0.70 ($450K)
= $429K
 Similarly,
 Expected cost reuse = $382K
 Expected costbuy = $267K
 Expected costcontract = $410K

OUTSOURCING
 Software engineering activities are contracted to a
third party who does the work at lower cost and,
hopefully, higher quality
 The decision to outsource can be either strategic or
tactical
 Strategic – business managers consider whether
significant portion of all software work can be
contracted to others
 Tactical – a project manager determines whether part
or all of a project can be best accomplished by
subcontracting the software work

SUMMARY
 Empirical Estimation Models
 The Structure of Estimation Models
 The COCOMO II Model
 The Software Equation
 The Make/Buy Decision
 Creating a Decision Tree
 Expected cost
 Build
 Reuse
 Buy
 Contract
 Outsourcing
 Strategic
 Tactical

2.6 Empirical estimation models & The make-buy decision.ppt

More Related Content

What's hot (20)

Similar to 2.6 Empirical estimation models & The make-buy decision.ppt (20)

Recently uploaded (20)

2.6 Empirical estimation models & The make-buy decision.ppt