1. Final Exam Review June 2, 2011SE 477: Final Review1/19

2. Risk Management  Risk Management  Types of risk: schedule, cost, requirements  Risk Identification  Involve the team  Risk Analysis  Risk Exposure (RE = Prob. * Size)  Probability is 15%, size is 10 weeks .15 * 10w = 1.5w  Risk Prioritization  80-20 rule; large size or prob. 1st; grouping; ignoring June 2, 2011SE 477: Final Review2/19

3. Risk Management  Risk Control  Plan  Risk Resolution (5 Types)  Avoidance (ex: scrub)  Assumption (just monitor)  Control (contingency)  Knowledge Acquisition (learn/buy/prototype)  Transfer (off project, team, critical path)  Risk Monitoring  Top 10 Risk List  Mitigation vs. contingency planning June 2, 2011SE 477: Final Review3/19

4. Project Execution  Project Roles  Project Control  Planning  Measuring  Evaluating  Acting  Binary Reporting June 2, 2011SE 477: Final Review4/19

5. Feature Set Control  Minimal Specification  Requirements Scrubbing  Versioned Development  Effective Change Control  Feature Cuts June 2, 2011SE 477: Final Review5/19

6. Resource Leveling  Leveling techniques  Fast tracking – parallel tasks  Activity shifting  Activity splitting  Activity stretching  Resource substitution  Allocating overtime June 2, 2011SE 477: Final Review6/19

7. SE 477: Final Review Change Control  Average project has 25% requirements change  Sources of change  Change control is a process  Overly detailed specs. or prolonged requirements phase are not the answer  Change Control Board (CCB)CCB  Structure, process, triage June 2, 20117/19

8. SE 477: Final Review Configuration Control  Items: code, documents  Change & Version control  Configuration Management Plan  Maintenance June 2, 20118/19

9. Earned Value Analysis  BCWS  BCWP  Earned value  ACWP  Variances  CV: (Cost Variance) = BCWP – ACWP  SV: (Schedule Variance) = BCWP – BCWS  Ratios  SPI (BCWP / BCWS), CPI (BCWP / ACWP)  CR (SPI x CPI)  Benefits  Consistency, forecasting, early warning June 2, 2011SE 477: Final Review9/19

10. SE 477: Final Review QA & Testing  SQA Team  Defect Metrics  Why track metrics? June 2, 201110/19

11. SE 477: Final Review Defect Metrics  Open Bugs (outstanding defects)  Ranked by severity  Open Rates  How many new bugs over a period of time  Close Rates  How many closed over that same period  Ex: 10 bugs/day  Change Rate  Number of times the same issue updated  Fix Failed Counts  Fixes that didn’t really fix (still open)  One measure of “vibration” in project June 2, 201111/19

12. Migration  Migration Plan  Migration Strategies  Flash Cut  Immediate Replacement  Parallel Operation  Staged  One part at a time  Importance of 2-way communication  Find-out customer’s key dates  Minimize intrusiveness  Back-out Plan  Data Conversion SE 477: Final ReviewJune 2, 201112/19

13. SE 477: Final Review Other Final Steps  Roll-Out  Release Check-List  Training  More than just end-users  Users, systems ops, maintenance developers, sales  Documentation  Many types: End-user, sales & marketing, operations, design June 2, 201113/19

14. SE 477: Final Review Project Recovery  3 Approaches  1. Cut the size of the software  2. Increase process productivity  3. Slip the schedule, proceed with damage control  People Steps  Morale; focus; re-assign  Process Steps  Fix classic mistakes; mini-milestones  Product Steps  Stabilize; trim features; take out the garbage June 2, 201114/19

15. SE 477: Final Review Post Project Reviews  Focused on process not people  Steps  Prepare survey form  Email team with survey and schedule meeting  Gather data  Conduct meeting  Prepare PPR report June 2, 201115/19

16. SE 477: Final Review Resource Allocation  Responsibility Assignment Matrix  Who does What  Skills Matrix  Who has what skills  Hiring Guidelines  Hire for trait, train for skill  Smart, gets things done  Balance June 2, 201116/19

17. SE 477: Final ReviewTeams  Start with objective  Problem resolution, creativity, tactical execution  Decentralized vs. Centralized  Large teams  Decompose via hierarchy, into optimal sizes  Optimal size?  4-6 developers  Hiring  Hire for trait – train for skill June 2, 201117/19

18. SE 477: Final Review Team Models  Business team  Technical lead + team; most common  Can be strong or loose hierarchy  Democratic team organization  Self-organizing teams  Extreme programming groups  Chief-programmer team  Surgical team; star at top; ego issues  Skunk works team  Off-site; pro: buy-in; con: minimal visibility  Feature team  Interdisciplinary; balanced  SWAT team  Highly skilled/specialized; Ex: security team June 2, 201118/19

19. June 2, 2011SE 477: Final ReviewTeams  Understanding of human factors  Avoid making unrealistic demands on people  Staff selection factors  Software development groups  Group communications  The working environment 19/19