Unit 12.4A · Term 4 (Advanced Revision)

Software Engineering & SDLC

Deep dive into the System Development Life Cycle (SDLC). This revision covers everything from initial feasibility studies to final system maintenance, aligned with Cambridge 9618 standards.

Learning Objectives (9618 Syllabus)

  • 11.2.1 Show understanding of the stages of the SDLC
  • 11.2.2 Discuss feasibility study (TELOS)
  • 11.2.5 Compare Prototyping types (Evolutionary vs Throwaway)
  • 11.2.8 Analyze Software Implementation methods

1. The SDLC Workflow

The SDLC is a formal stages-based approach to building a system. Each stage must produce specific documentation (deliverables).

2. Feasibility Study (TELOS)

Before committing resources, we use TELOS to evaluate if the project should proceed.

Technical
Is the hardware/software available or buildable?
Economic
Cost-benefit analysis. Is it within budget?
Legal
Does it violate Copyright or Data Protection laws?
Operational
Will the end-users actually be able to use it?
Scheduling
Can it be completed within the required timeframe?

3. SDLC Models Comparison

Model Workflow Best For Risk Level
Waterfall Linear phases. No going back easily. Simple, rigid projects. High (late testing)
Iterative Developing in small chunks/versions. Large systems with evolving specs. Medium
Spiral Risk-driven cycles with analysis. High-budget, high-risk projects. Low (due to analysis)
Agile Customer focus, sprints, rapid updates. Startups, web apps. Variable

4. Prototyping Strategy

Throwaway Prototyping: Built only to demonstrate a concept, then discarded. Fast and cheap.


Evolutionary Prototyping: The prototype is refined and eventually becomes the final product. High quality but takes longer.

5. Implementation (Cut-over)

Choosing how to switch from the old system to the new one is a strategic decision.

Method Strategy Pros Cons
Direct Instant switch. Fast, least expensive. High risk of data loss.
Parallel Both run together. Safest; backup exists. High cost; double work.
Pilot One branch/office first. Errors isolated to one site. Slow full roll-out.
Phased Module by module. Easier training. System inconsistency.

[Image comparing Direct, Parallel, Pilot, and Phased implementation timelines]

6. Types of Maintenance

  • Perfective: Improving performance or adding new features (making it better).
  • Adaptive: Modifying the system to work in a new environment (e.g., new OS).
  • Corrective: Fixing bugs and errors discovered after release.

Exam-Style Revision Tasks

Analyze

Q1: A nuclear power plant is updating its safety control software. Justify why the Spiral Model is more appropriate than the Waterfall Model. [4 marks]

Apply

Q2: Explain two reasons why Parallel Running is often used by banks despite the high cost. [4 marks]

Understand

Q3: Describe the difference between Corrective and Adaptive maintenance. [4 marks]

Quick Self-Check

Q1: Which TELOS component checks if the system complies with Data Protection laws?

Legal Feasibility.

Q2: Which implementation method is the riskiest?

Direct Changeover.