2016-03-02

Software Engineering
Economics
(CS656) Overview

Jongmoon Baik

Class Information
• Instructor: Jongmoon Baik
– Office: N1, Rm. 502
– Phone: 042-350-3556
– Email: [email protected]
– Office Hour: MON & WED: 10:30AM-12:00PM
(or By Appointment)
• Class Info.
– Class Hours: MON & WED 13:00 PM– 14:15 PM
– Class Room: E3-1 Rm. 3445
– We’ll start on time with any questions and end on time

• Teaching Assistant: Duksan Ryu Email: [email protected]

– Office: N1, Rm. 525 , Office Hours: TBA
– Tel: 042-350-7756

1
 2016-03-02

Admin Notes

• Class Website:
http://spiral.kaist.ac.kr/wp/2016springcs656/
– Announcements: You must check periodically
– All assignments, lecture notes and supplemental
materials are available on Class Schedule

• E-Mail
– Be careful as it does not show other recipients
– Send e-mails with subject line starting with “[CS656]
XXXXXX”
3

Text Book & References

• Text Book
– Rick Selby, “Software Engineering: Barry W. Boehm’s Lifetime Contributions
to Software Development, Management, and Research”, Wiley, 2007, ISBN
978-0-470-14873-0

• References
– Boehm, Software Engineering Economics, Prentice Hall, 1981.
– Boehm et. al., Software Cost Estimation with COCOMO II, Prentice
Hall, 2000.
– Reifer, Don. Making the Software Business Case: Improvement by the
Numbers , Addison Wesley, 2001.
– Royce, Walker, Software Project Management, Addison Wesley, 1998.
– Reifer, Don, Software Management, 6th ed., IEEE-CS Press, 2002
– Stefan Biffl et. al, Value-Based Software Engineering, Springer, 2006

2
 2016-03-02

Class Schedule

Course Structure
• Lectures
– Check for slides DAY OF LECTURE please
– We try to update all the time
– Not all the supplemental materials will be on Class
Website due to copyright protections

– Proactive participation is critical

– Ask questions
– Prepare questions ahead of class
Challenge the instructors and each other!
6

3
 2016-03-02

Grading Policy
• Participation & Attendance (10%)
– Our perception! Not yours…. Ask if you don’t know
– I will call on people randomly at first, not so later
– Be proactive, but don’t just “run the mouth”

• Midterm Exam (20%)

– Cover all the materials you learned before the midterm

• Assignments (40%)
– Due at beginning of each week

• Final Term paper (30%) – Individual START NOW

Exam Policy
• Mid-Term Exam
– Midterm (4/20) :Tentative
– Open Book & Note
– Should attend (No make-up exam)

• No Final Exam !!!

– Term Paper Presentation

4
 2016-03-02

Assignments
• Each Assignment: Posted on the Web
• Due: At the beginning of the class on the due
date
• Submission
– hard copy to Instructor at the class
– Email soft copy to T.A. (CC it to Instructor)
• Late Penalty
– One day (30%), Two days (50%)
– Two days after due date: No Acceptance

Final Paper - I
• A Software Engineering subject of interest TO YOU.
– Describe Abstract (One page)
• Due Date: April 15(FRI)
• Graded on 10 point scale
– 2 points for Topic (on target, interesting, difficulty)
– 2 points for bibliography
• 10 min. “good” sources (URL’s aren’t “good”)
• 3 sources min. primary (original research)
– 6 points
• Readability (spelling, grammar, organization, etc.)
• Content: Needs to show analysis and synthesis, not just regurgitation
• Final Presentation
• Term Paper Format & Submission
– Papers should comply with IEEE style and be submitted in MSWord (.doc)
– Send the paper to T.A. and CC it to the instructor
– Due Date: June 10 (FRI)

10

5
 2016-03-02

Attendance & Participation

• More than 15% absent
– Fail the class (Grade: F)
• Being late more than 30 minutes
– Regarded as absent
• No disturbance in the class
– Cellular Phone: Turn off or Manner mode
– No walking in and out during class
• Pro-active participation
– Bring with your questions and ideas

11

Writing Suggestions
• First draft -don’t worry about length
• Don’t “format” it down, edit the paper to right
length
• Common errors
– Paper not proofed –Don’t rely on spell checker
– Doesn’t answer the question (s)
– Hard to read – format
“If you don’t like reading it, we won’t either”

12

6
 2016-03-02

General Writing Notes

• Must be readable
– Intended audience: Senior management
• PLEASE, 12 pitch font minimum
• Simple font
• 1.5 spacing is nice BUT not mandatory
• Use indentation, bold, etc. as needed
• Spelling and grammar count! (English)
• Must make sense to the reader

13

Course Readings
• Text
– May repeat for review
• Provided (Files or Links)
• Active reading – 3 times
– Skim first
– Read in full with questions in mind
• What happened, significance to the field, lessons
learned…
– Read Again

14

7
 2016-03-02

Citation of Your Source

• Typically one warning, with a reduced grade
• Then 0’s, or fail in class
• If using material verbatim
– Put in quotations with “according to”etc.
– I don’t need full source cite
• According to Jalote, “………..”
• Or at end of sentence, paragraph, “Jalote, pgs. 47-51”
• If in doubt, ask
– Paraphrasing, still state source, but quotations may
not be needed
15

Plagiarism !!!

16

8
 2016-03-02

“The Problem” - I
• From www.academicintegrity.org, in U.S.
– 70% of students admit to some cheating
– 25% admit to cheating on major tests
– 50% on written assignments in past year
– 40% to plagiarizing from the internet
– 77% don’t see this as a “serious issue.”
– 49% admit to unpermitted student collaboration
– Faculty reluctance to be “bad guys.”
• Cheating, copying other work, plagiarism is on the rise in US universities.
• Many students feel that they need to “cheat” in order to be competitive
• Some students have stated that “cheating” is acceptable in some cultures
• Some have stated that plagiarism is a form of flattery

17

“The Problem” - II

“In any presentation, creative, artistic, or

research, it is the ethical responsibility of each
student to identify the conceptual sources of
the work submitted. Failure to do so is
dishonest and is the basis for a charge of
cheating or plagiarism, which is subject to
disciplinary action. “
Plagiarism is considered cheating!

18

9
 2016-03-02

What is plagiarism?
• According to the Merriam-Webster Online
Dictionary, to "plagiarize" means
– to steal and pass off (the ideas or words of another)
as one's own
– to use (another's production) without crediting the
source
– to commit literary theft
– to present as new and original an idea or product
derived from an existing source.
• An act of fraud (stealing someone else's work)
19

The Solution
• Professional integrity
• Unlike study, faculty here have no problem
dealing with Plagiarism/Cheating
– No greater offense
– Allowing yourself to be copied…
• Reputation will follow you
• KAIST has a site license for Turnitin
– Web and DB search for similarity

20

10
 2016-03-02

Software Economics Overview

21

What do you have to deal with if you

were a CEO of a software company
or a manager of a software
development team?

22

11
 2016-03-02

What is Software Economics?-I

- Software Economics is the field that seeks to

enable significant improvements in software
design and engineering through economic
reasoning about product, process, program,
and portfolio and policy issues -

Source: Barry Boehm & Kevin Sullivan, “Software Economics: A Roadmap”

23

What is Software Economics?-II

The study of how scarce project resources are

allocated for software projects
The economics of the Software Industry

Psychology

Social Psychology Organizational Behavior

Software
Economics
Economics Statistics

Software Development

http://www.softwaremetrics.com/s.htm

24

12
 2016-03-02

Objectives of Software Economics

• To provide an empirical view of where the

effort and money goes when we build large
software systems

• To suggest ways of reducing and controlling

software development costs

25

Macro vs. Micro Economics

• Macro economics
– SE supports a commercial S/W sector that earns
$200B to $240B in US
– SE drove $1T of economic growth in US
• Micro economics
– About ¼ of software projects are delivered
successfully
– Commercial developers typically write 1.5K lines
of code per year
– Government developers typically write 1.2K lines
of code per year
26

13
 2016-03-02

Software Economics: A Roadmap

• The Goal of Software Economics

– Develop fundamental knowledge to enable
significant, measurable increases in the value
created over time
– There are subtleties
– Value itself can be a complex & subtle quantity
• Value can be assessed from monetary profits to the
solution of social problems
– Identify a network of important intermediate
outcomes

27

Problems in Past & Current Works

• Past
– Focuses largely on costs, not on benefits
– Not on value-added

• Current
– Technical software design
• Not linked to value creation

Software Economics can lead to fundamental

improvements in software design and engineering, in
theory and practice.

28

14
 2016-03-02

Software Failures & Risks

• Failed project cost for U.S in 1998
– Estimated at $80 billion
• Project, program, & business failures
– Inevitable (about 70%: chaos report, 2003)
– Desirable in a dynamic market place
• Software development and use
– Unpredictable Value Destroy & Risk Exposure
• Software costs
– Jumps in ways inconsistent with expectable risk
• Schedule Delays
– Lost value, quality shortfall, and missed market opportunities

29

Need for SE Research (I)

“Software Engineering Decision-Making Today”

• Principals of separation of concerns

– Economic-independent “Flatland”
• Focus on representation structure and logical semantics
• No links b/w technical issues & value creation

• Design decisions are intimately coupled with

fundamental business decisions
– Decisions: linked to value creation in a business context
• Profit vs. not-profit organizations

• Software development involves the investment

of valuable resources
30

15
 2016-03-02

Need for SE Research (II)

“Software Engineering as a Value-Creation

Activity”

• No Connection b/w Decision and Value-Creation Criteria

• Need to Understand & Reason about the connections b/w

technical decisions & enterprise-level value maximization

31

Value & Risk based Project Types

Business Value
Low High

High
Project Risk

Low Who cares?

32

16
 2016-03-02

Need for SE Research (III)

“Sources of Mismatch b/w Software Decisions &
Value Creation”

• Lack of adequate frameworks for modeling, measuring,

analyzing the connections b/w technical properties and
decisions and value-creation
• No Reason about value creation as an objective or about
technical parameters can be manipulated for value creation
purposes
• Inadequate design space where software designers operate

33

More Emphasis on Software Economics

• Situated at the intersection b/w information economics and

software engineering
• From government projects to commercial sector (different
measure of value and dynamics) after the cold war
– e.g.:Time-to-market  critical factor
• Much further impacts of software-enabled change across and
into organizations
– e.g.: order fulfillment of electronic retailing
– Demand of holistic approach
• Increasing understanding of value creation
– Strategic considerations dedicates
• Holistic approach
• One that treats uncertainty, incomplete knowledge and competition in a sophisticated manner

34

17
 2016-03-02

Software Economics
Roadmap - I Strategic Concerns
(reasoning about real
options and synergies)

Tactical Concerns
(cost estimation)

35

Software Economics Roadmap - II

• Better Decisions for Value Creation

– Need richer design spaces
• Less able to build systems from specialized, efficiently produced, volume-
priced thirty-party components
• Less able to use markets to manage risk through warranties, liability
insurance etc. – major hurdle to efficient production
– Need to understand links b/w technical design mechanisms, context,
and value
– Need to educate decision-makers in how to employ technical means
more effectively to create value
– Need dynamic monitoring and controlling mechanisms to better guide
decision makers
• Dynamically responsive to new information and changing conditions

36

18
 2016-03-02

Modeling Cost, Benefit, and

Value - I
• Modeling Software Development Cost,
Schedule, and Quality
– COCOMO II, PRICE S, SEER-SEM, etc.
– Still about +/-15% range of variation b/w projects and organization due
to counting rules
– Another source of variation – proliferation of new processes and
technologies

• Elusive Nature of Software Estimation

Accuracy
– Previous models get obsolete by better method and technology

37

Modeling Cost, Benefit, and Value - II

• Modeling Benefits and Value

– Productivity – Difficult, Controversial
• Sizing (SLOC, FP), Language level
– Highly domain-specific
• General Benefit-Modeling Techniques
– Frequently take the form of causal chain linking the organization goals
or objective to the development or acquisition of software
– e.g.: QFD, GQM, DMR-BRA, etc.
• Modeling Value: Relating Benefits to Costs
– Example: SPL & Reusable components
• Primary value: not in cost-avoidance but rather in reduced time-to-
market

38

19
 2016-03-02

Tracking and Managing for Value

• SEI SW-CMM & CMMI

– Good indicator of the current status and trends in models for software
project tracking and managing
– A sound implementation of project management process based on
KPAs (or PAs)
– CMMI
• Quality needs are exemplified
– Functionality, reliability, maintainability, usability, cycle time, predictability,
timeliness, and accuracy
• Emphasis on traceability not only to requirements but also to business
objectives, customer discussions, and market surveys
– Focus on tracking and managing the execution of project, rather
than on the value it will presumably deliver
• e.g.: EVMS  does not track business value associated with the
product’s operation implementation.

39

SE Important Issue
“ Better, Cheaper, & Faster”
• Software costs are big and growing

• Many useful software products are not getting developed

• Better Software Not Many Software

40

20
 2016-03-02

Software Cost Distribution Insight

• Development & Rework Costs

• Code and Documentation Costs

• Labor and Capital Cost

• Software Cost by Phase and Activity

41

Dev. and Rework Costs

• Large fraction of software costs
– Devoted to rework
• Fixing & Reworking costs
– Much smaller (by factor of 50-200) in the earlier
phases of software life cycle than in the later
phases
– Left Shift via Early Detection and Removal of high
risk problems
• Rework Instances
– Tend to follow a Pareto distribution
42

21
 2016-03-02

Code and Documentation Costs

• A larger proportion of costs

– Attributable to produce intermediate documents
(Specifications, Manuals, Plans, Studies, Report)
than code

– Exact proportion varies by applications and

process models
• (Commercial Inventory System, Space Shuttle Control
System etc), (Agile, Spiral etc)

43

Labor and Capital Costs

• Software development and evolution
– Very labor Intensive

• Productivity Leverage
– Can be obtained by making software production
more capital intensive

– e.g: Outsourcing (offshoring), Global S/W Dev…

44

22
 2016-03-02

Productivity Improvement Opportunities

• Productivity Improvement by
– Make people more effective
– Make steps more efficient
– Eliminate steps
– Eliminate rework
– Build simpler products
– Reuse components
– etc.

45

1st Assignment
• Fill out the questionnaire and submit it before the
class on Mar. 7

• Find out a article that deals with a project failure

– From Computer World, IEEE Computer, IEEE Software
etc.
– Summarize main causes of the failure
– Suggest possible solutions with rationale
– Due: Mar. 14
– Submit a hardcopy at the class and send a softcopy to T.A.
before the class

46

23
 2016-03-02

Q&A

47

24