ACQUISITION STRATEGY

1
ACQUISITION STRATEGY PLANNING

• MN3309 Session 5
• Prof. Mark Nissen

2
Agenda

• Project Reflection Acquisition Plan
• System Life Cycle Methodologies
• S/W Production Modeling
• Summary

3
Project Reflection Acquisition Strategy Plan

• Helpful to reflect on C-SAWS project
• Real S/W acquisition
• Instantiate course principles
• C-SAWS acquisition strategy
• Read acquisition strategy plan?
• Summarize overall approach?
• Key aspects?
• Strengths weaknesses?

4
Requirements Architecture Exercise

• Note-taking requirements architecture
• Small groups (3)
• Setup 1 blank page to compare/contrast
  development methodologies (in ink)
• Pass CW. Notes during lecture (in ink)
• Pass CW. Evaluate after lecture
• Answer questions
• Compare with requirements
• Pass to one person for integration
• Discussion - understanding rqmts

5
Weapon System Life Cycle

• Needs analysis (-1) - MNS
• Concept exploration (0)
• ORD, IMP/S, TEMP, acquisition strategy
• Demonstration/Validation (1)
• SDP, risk reduction, development B/L
• Development (2) - EMD for W/S
• Requirements CM, DTE, Production B/L
• Production Deployment (3) - FOTE
• Operations support - maintenance

6
Software System Life Cycle

• Life cycle methodology
• Phases
• Milestones
• Content of deliverables
• Evaluation criteria for deliverables
• Phases
• Analysis design (rqmts logical design)
• Code unit test (executable modules)
• System acceptance test (tested system)
• Implementation maintenance (fielded)

7
Software Development Notes

• S/W subsystem integrates with W/S
• S/W LC integrates with W/S LC
• S/W on critical path of W/S
• S/W H/W intimately linked (interface)
• 70 S/W LCC committed _at_ MS I
• Programmers make design decisions
• S/W changes user environment

8
S/W Functionality Requirements cycle
User environ
Software requirements
Software functionality
9
S/W Life Cycle Methodologies

• Code fix (just do it)
• Waterfall (aka Grand Design)
• Incremental (P3I)
• Evolutionary (P2I)
• Spiral (phased waterfall risk demo)
• RAD? Prototyping?

10
Single vs. Incremental Builds
Single Build
C
C2
T
D
Incremental Build
A
I
P P-Mo, Mo SB 140, 5 IB 140, 7
11
SDLC Comparison2 Modules A B
Time
Grand Design
Anal Design Code Test Implem Support
AB
Incremental
Anal Design
AB
Code Test Implem Support
A
Code Test
Implem Support
B
Evolutionary
Anal Design Code Test Implem Support
A
Anal Design Code
B
12
Selecting the Appropriate Methodology

• Program-dependent described in SDP
• Waterfall - low risk, certain funding
• Almost never crystalline requirements
• Incremental - low to med risk (funding)
• Extensible architecture known rqmts
• Evolutionary - med to high risk (funding)
• Extensible architecture evolve rqmts
• Spiral - high risk (uncertain funding)
• Extensible architecture determine rqmts

13
COTS Implications

• COTS
• Make vs. buy?
• What parts of SDLC involved?
• COTS problems/limitations?
• Packaged applications
• What are they? Examples?
• Effect on enterprise processes?
• SDLC approaches?

14
MIL-STD-498 ISO/IEC-12207

• Establish uniform rqmts for S/W development
  documentation
• Merges 2167A 7935A
• Applies to all SDLC phases
• Applies to in-house contracted S/W
• Independent of specific SDLC method
• Interprets MIL-Q-9858A ISO9001
• Invokes no other stds self-contained
• DIDs usage require tailoring

15
Contents

• Scope definitions
• General requirements
• S/W development methods stds
• S/W reuse
• Critical requirements
• H/W resource utilization
• Recording rationale access for review
• Detailed requirements
• Project planning oversight

16
Contents (Cont)

• Detailed requirements (cont)
• Establishing a SEE
• System rqmts analysis design
• S/W rqmts analysis design
• S/W implementation
• Testing (S/W unit - H/W system)
• Use and transition preparation
• S/W CM, product eval, QA, corrective action,
  joint reviews, other activities

17
Guidance

• S/W development presumed to occur independently
  of CDRLs
• On-site, online CDRL delivery
• Tailor-out unnecessary items
• App B interpretation for reusable S/W
• App C problem reporting
• App D S/W product evaluation
• App F candidate indicators/metrics
• App G strategy, tailoring, build plans

18
S/W Production Modeling (K3)

• Model is abstraction - partially right
• Validation calibration critical
• Kemerer
• Validation vs. calibration?
• Uncalibrated models 600 MREs!
• Calibrate to reflect local conditions
• S/W scale economies/diseconomies?
• Banker Kemerer
• Most productive S/W size (MPSS)?
• Implications for S/W PM?

19
Most Productive Software Size (MPSS)
Productivity
MPSS is organization-unique
Max
Average
Marginal
Software Size
MPSS
Increasing returns
Decreasing returns
Which region of PM Exercise (book)? Effect of
incremental/evolutionary development?
20
Summary

• C-SAWS reflection is helpful
• S/W changes user environment
• Select appropriate life cycle method
• COTS advantages disadvantages
• Understand stretch-out effects
• MPSS and returns to scale