CIS 375 SOFTWARE ENGINEERING

UNIVERSITY OF MICHIGAN-DEARBORN

DR. BRUCE MAXIM, INSTRUCTOR

Date: 9/22/97

Week 3

DYNAMIC ESTIMATION MODELS:

1. Putnam Model (initially based on 200 military models).
   (Rayliegh curve -> skewed, median & mean offset from one another).

{ lines of code measure} {c represents the technology factor from survey}.

1. Parr Model (variation of above):
   Staff (t) = sech² [(a t + c)/2].

1. Jensen Model:
   (less sensitive to compression)
   S_s = C_te T K ^1/2
   { C_te represents technology constraint with environmental adjustment}.
2. COPMO (cooperative programming model):
   • Size of project.
   • Size of code.
   E = E₁ (S) + E₂ (m);
   {E represents total effort}.
   {E₁(S) represents the effort of one or people working independently

   E₁(S) = a + bs ,where a & b are empirically derived constants}.
   {E₂(m) represents the effort required to coordinate the development process

   E₂(m) = cm² where m is the average number of team members}.
   • It is not clear that all cost factors are incorporated in each model.
   • Dynamic models have the ability to make adjustments to accommodate the fact that all cost factors are not considered.

MAKE-BUY-DECISION:

1. Is a computer-based necessary (cost effective?):
   • If the company doesn't know enough about procedures without computer, you can't build a beneficial computer system for them.
   • If things are only done once, it's not worth designing a system.
   • It must be a repeatable task & cost effective.
2. Can an existing software package be purchased for the task?
   • For example: if MS Access can be purchased for $150, why write a new one? We cannot create a new one for this low cost.

DECISION STEPS:

1. Develop specifications.
2. Estimate internal cost & delivery.
3. Select 3 or 4 candidate packages.
4. Select reasonable components.
5. Build a cost-benefit comparison matrix, or Benchmarking (if from scratch you can't do this).
6. Evaluate each software package or component based on history with the product or vendor.
7. Contact other users.

RISKS ANALYSIS:

1. Risk identification.
2. Risk projection (impact of risks/likelihood of risk actually happening).
3. Risk assessment (the change if it happens).
4. Risk management.

KINDS OF RISK:

• Project risks:
  • Schedule, budget, people, resources, etc.
• Technical risks:
  • Design, implementation, interfacing, etc.
• Business risks:
  • Building a project that the public actually wants.

RISK PROJECTION (OR ESTIMATION):

• Likelihood that a risk is real.
• Consequences of problem, if it occurs.

RISK PLANNER:

1. Establish a scale indicating perceived likelihood.
2. Determine consequences.
3. Estimate impact on project (for each risk).
4. Note overall accuracy of risk projection (to avoid misunderstanding).

Date: 9/24/97

Week 3

SAMPLE IMPACT ASSESSMENT TABLE:

CATEGORY \ COMPONENTS		PERFORMANCE	SUPPORT	COST	SCHEDULE
	1	Failure to meet would result in mission failure		Failure results in increased costs and schedule delays with expected values in excess of $500K
CATASTROPHIC	2	Significant degradation to nonachievement of technical performance	Nonresponsive or unsupportable software	Significant, financial shortages, budget overrun likely	Unachievable delivery date
	1	Failure to meet the requirement would degrade system performance to a point where mission success is questionable		Failure results in operational delays and/or increased costs with expected value of $100K to $500k
CRITICAL	2	Some reduction in technical performance	Minor delays in software modifications	Some shortage of financial resources, possible overruns	Possible slippage in delivery date
	1	Failure to meet the requirement would result in degradation of secondary mission		Costs, impacts, and/or recoverable schedule slips with expected value of $1K to $100K
MARGINAL	2	Minimal to small reduction in technical performance	Responsive software support	Sufficient financial resources	Realistic, achievable schedule
	1	Failure to meet the requirement would create inconvenience or nonoperational impact		Error results in minor cost and/or schedule impact with expected value of less than $1K
NEGLIGIBLE	2	No reduction in technical performance	Easily supportable software	Possible budget underrun	Early achievable date

RISK PROJECTION:

1. List all risks listed in column one.
2. Categorizes risks by type:

{business, project, etc.}

1. Probability value (guess).
2. Impact.
3. Sort table based on impact and probability.

RISK ASSESSMENT:

1. Define referent levels.

{break even points - cost/schedule)

1. Develop relationships between each reference factor.
2. Determine region of termination, bounded by uncertainty curve.
3. Try to predict compound combinations of risks - affecting referent level.

RISK MANAGEMENT (MONITORING):

1. Determine causes of job turnover.
2. Eliminate causes before project starts.
3. After project starts, assume turnover is going to occur and ensure continuity.
4. Make sure teams are organized and distribute information widely.
5. Define documentation standards and be sure documents are produced in a timely manner.
6. Conduct peer review of all work.
7. Define backup staff.

CUSTOMER QUESTIONS

1. Do you understand my needs?
2. Can you design a system to meet my needs?
3. How long will it take?
4. How much will it cost?

PROJECT SCHEDULE

1. Deliverables:
   • Documents.
   • Demonstration of function.
   • Demonstration of subsystem.
   • Demonstration of accuracy.
   • Demonstration of
2. Milestones:
   Completion of an activity of delivery (must be measurable).
   Milestones must have definite a start and stop.
3. Work breakdown structure:
   Separate project development into phases, composed of steps (series of activities).

DIGRAPH = ACTIVITY GRAPH.

CRITICAL PATH METHOD:

• EST = earliest start time, EFT = earliest finish time.
• LST = latest start time, LFT = latest finish time.
• Slack = (LST - EST) or (LFT - EFT).

Activity	Precursor	Duration	EST	EFT	LST	LFT	Slack
Start	-	0	0	0	0	0	0
A	Start	2	0	2	0	2	0
B	Start	3	0	3	4	7	4
C	A	5	2	7	2	7	0
D	A,B	4	3	7	7	11	4
E	D	2	7	9	11	13	4
F	B,C	6	7	13	7	13	0
FINISH	E,F	0	13	13	13	13	0

• EST(1) always = 0, EFT(1) always = 0.
• LST(1) always = 0, LFT(1) always = 0.
• EFT(I) = EST(I) + DUR(I).
• EST(I) = max(EFT of all predecessors).
• LFT(I) = LFT(I) - DUR(I).
• LFT(I) = min(LST of all predecessors).
• LFT(FINISH) = LST(FINISH) = EST(FINISH) = EFT(FINISH).

PERT:

• * = Critical activity.
• - = Non critical activity.
• F = Float or slack.

1. Gant Charts:

1. Overlay People On Gant Chart:

1. Milestone Chart:

1. Line Graph:

PRESSMAN'S PROJECT PLAN:

1. Project scope.
2. Project schedule.
3. Project team organization.
4. Technical description of proposed system.
5. Project standards, procedures, proposed methodologies.
6. Quality assurance plan.
7. Special tools & techniques.
8. Configuration management plan.
9. Documentation plan.
10. Data management plan.
11. Resource management plan.
12. Test plan.
13. Training plan.
14. Security plan.
15. Maintenance plan.