Theory of Value-Based Systems and Software Engineering

1
Theory of Value-BasedSystems and Software
Engineering

• Apurva Jain, apurva.jain_at_usc.edu
• USC Center for Systems Software Engineering
• http//csse.usc.edu

2
Context and Definitions Value-Based SSE

• Definition
• the explicit concern with value (financial and
  non-financial) in the application of science and
  mathematics by which the properties of computer
  systems and software are made useful to people
• Practicing VBSSE
• integrating stakeholder value considerations
  into the full range of systems and software
  development principles and practices

3
Context and Definitions Value

• Origin
• from Latin valere to be worth
• Definition (Webster)
• relative worth, utility or importance
• Financial or non-financial (Maslow, Kaplan and
  Norton)
• Key non-financial corporate value drivers
  (Forbes.com with Wharton and EY)
• Innovation, ability to attract talented
  employees, alliances, quality of major processes,
  products, or services, environmental performance

4
Key Observations from Literature

1. Organizations are social units people-centric
2. Assume bounded rationality (Simon)
3. No silver-bullets, not one-size-fits-all (Brooks)
4. Stakeholder values are financial and
   non-financial (Maslow, Forbes-EY)
5. Timeless theories of physics will not apply (from
   1-4)
6. Organizational systems affect the bottom line
   (Burton and Obel)
7. Engineering theories must take the organization
   in context (from 4 and 6)

5
Successful Project? Multi-Contingency
Organizational Context (Burton and Obel)

• Key Observations from Literature (contd.)
• Management theories usually take at least a
  decade for conclusive evidence
• Problem and solution space is huge, balance on
  breadth and depth (T-shaped)
• Therefore Avoid reinventing the wheel,
  capitalize on existing research

6
What is a Theory?

• 1960s System of general laws
• Spatially and temporally unrestricted
  nonaccidental
• Does not work for systems and software
• 1994 System for explaining a set of phenomena
• Specifies key concepts, laws relating concepts
• Not spatially and temporally unrestricted
• Better for people-intensive activities

7
Theory W Enterprise Success Theorem

• Your enterprise will succeed
• if and only if
• it makes winners of your success-critical
  stakeholders
• Proof of if
• Everyone that counts is a winner(i)
• Nobody significant is left to complain(ii)
• Proof of only if
• Nobody wants to lose(iii)
• Prospective losers will refuse to participate, or
  will counterattack(iv)
• The usual result is lose-lose(v)

8
Theory W WinWin Achievement Theorem

• Making winners of your success-critical
  stakeholders requires
• Identifying all of the success-critical
  stakeholders (and the contingencies they
  bring-in) (SCSs)(i)
• Understanding how the SCSs want to win (ii)
• Having the SCSs negotiate a win-win set of
  product and process plans(iii)
• Controlling progress toward SCS win-win
  realization, including adaptation to change(iv)

9
VBSSE Theory 41 Model
10
Supporting Theories Contingency

• Provides insights into various organizational and
  project contingencies
• What the best way to do x? It depends.
• Spans socio-political, environment, cultural,
  technical dimensions
• Component theories include
• Benefits Chain, Model Clashes, Network Analysis
• Primary contributions include
• Helps identify contingent success-critical
  variables
• Applies to whole (socio-technical) system
• Appeals to intuition that systems fail because of
  mismatches.

11
Environment Framework (Porter, Burton and Obel)
Uncertainty Equivocality Complexity Hostility
Buyers Bargaining Power
Suppliers Bargaining Power
Threat of Substitutes
Threat of New Entrants
Inter-firm Rivalry

• Systems Software Project Implications
• Process
• System Architecture
• System Capabilities

12
Environment Propositions

• Propositions for organization structure
• If the environment has low equivocality, low
  complexity and low uncertainty then formalization
  should be high, organization complexity should be
  medium and centralization should be low (i)
• If the environment has low equivocality, high
  complexity and low uncertainty then formalization
  should be high, organization complexity should be
  medium and centralization should be medium (ii)
• If hostility is extreme, then formalization
  should be low, and centralization should be very
  high (iii)

13
Management and Leadership Style Frameworks
(Burton and Obel)
Leader Producer Entrepreneur Manager
Preference for Delegation HIGH HIGH LOW LOW
Level of Detail in Decision-Making LOW HIGH HIGH HIGH
Reactive/Proactive Decision-Making PROACTIVE REACTIVE PROACTIVE REACTIVE
Decision-Making Time Horizon LONG SHORT LONG SHORT
Risk Preference HIGH LOW HIGH LOW
Motivation and Control INSPIRATION CONTROL INSPIRATION CONTROL

• Systems Software Project Implications
• Staffing
• Process

14
Management and Leadership Style Propositions

• Propositions for project structure
• If an individual is a leader, then
• Centralization should be low (i)
• Formalization should be low (ii)
• Complexity should be medium (iii)
• Incentives should be results based (iv)
• Coordination and control should be loose (v)
• If an individual is a manager, then
• Centralization should be high (vi)
• Formalization should be high (vii)
• Complexity should be high (viii)
• Incentives should be procedure based (ix)
• Coordination and control should be tight (x)
• If an individual is a producer, entrepreneur

15
Technology Frameworks (Perrow)
CRAFT NONROUTINE
ROUTINE ENGINEERING
ILL-DEFINED WELL-DEFINED
PROBLEM ANALYZABILITY
FEW EXCEPTIONS
MANY EXCEPTIONS
TASK VARIABILITY

• Systems Software Project Implications
• Staffing
• Process
• System Architecture

16
Technology Propositions

• vs. Strategy
• Nonroutine technology is a misfit with a
  defender strategy (i).
• vs. Management Style
• Nonroutine technology is a misfit with a manager
  leadership style, except in small organizations
  (ii)
• vs. Organizational Climate
• Nonroutine technology is a misfit with an
  internal process climate (iii)
• vs. Organizational Environment
• Nonroutine technology is a misfit with a high
  equivocality environment (iv)

17
Technology Frameworks (Al-Said, Boehm)

• Systems Software Project Implications
• Staffing
• Process
• System Architecture

18
Technology Propositions

• Maintainer vs. Developer
• Ease of transition is a misfit with freedom of
  COTS (i)
• User vs. Acquirer
• High levels of service is a misfit with freedom
  of COTS (ii)
• User vs. Acquirer
• Application compatibility is a misfit with
  freedom of COTS (iii)

19
Supporting Theories Utility

• Provides a rich theoretical method to infer
  subjective stakeholder value over a set of
  choices
• Component theories include
• Maslow, Simon, Multiple attribute utility theory
• Primary contributions include
• Helps determine Pareto optimality
• Works well with subjective preferences
• Provides rich fodder (stakeholder utility
  functions) for other theories

20
Supporting Theories Decision

• Provides a plethora of techniques and models to
  enable decision making
• Component theories include
• Game theory, options theory, statistical decision
  theory
• Primary contributions include
• Helps determine risks and opportunities
• Works well with uncertainty
• Not wedded to a particular decision theory, such
  as bounded rationality, economic man, etc.
• Provides rich fodder (competing investment
  options) for other theories

21
Supporting Theories Control

• Provides theory augmented models for state
  measurement
• Component theories include
• BSCs, BTOPP, Risk management
• Primary contributions include
• Helps determine necessary conditions for enabling
  control
• Works well in situations requiring stability AND
  adaptability
• Provides rich fodder (risks and opportunities)
  for other supporting theories

22
VBSSE Theory 6-Step Process
23
The Incremental Commitment Model (ICM)
24
VBSSE Phase Configuration
25
Conclusion

• It provides a unifying theory for practicing
  VBSSE that is
• Entirely theory-based
• There is nothing as practical as a good theory
  Karl Lewin
• Built on existing research
• Empirically validated (TBD)
• Simple
• Derived from simple rules, provides step-by-step
  guidance