BIT402 Software Project Management

Software Project ManagementUnit 711 min read

Project Monitoring, Control & Quality Assurance: Techniques, Tools & Metrics

Unit 7 of Software Project Management covers systematic approaches to track project health, correct deviations, and ensure deliverables meet quality standards—using techniques like Earned Value Analysis, quality gates, and control charts, with real-world applications in Nepalese tech firms.

TAKEAWAYS:

  • Monitoring uses metrics (e.g., SPI, CPI) to compare planned vs. actual progress in real-time, while control applies corrective actions via tools like Gantt charts and Pareto analysis.
  • Quality assurance relies on preventive techniques (reviews, inspections) and corrective techniques (testing, metrics) to reduce defects before they reach users.
  • Critical path analysis identifies the longest sequence of tasks (e.g., in a Daraz delivery system) that dictates project duration—delaying any task here delays the entire project.
  • Earned Value Management (EVM) integrates scope, schedule, and cost into a single metric (e.g., SV = -5 means behind schedule by 5 days).
  • Quality models (e.g., McCall’s, ISO 9001) break quality into factors like product operation, revision, and transition, helping teams prioritize improvements.
  • Agile monitoring uses burndown charts and velocity tracking to adjust sprints dynamically, while traditional projects use milestone tracking and risk registers.

1. Definitions: Monitoring vs. Control vs. Quality Assurance

Monitoring and controlling are interdependent processes in project management. While monitoring involves collecting data (e.g., progress reports, defect logs), controlling uses that data to take corrective actions (e.g., reallocating resources, adjusting timelines). Quality Assurance (QA) is a broader discipline that prevents defects through processes like reviews and testing, whereas Quality Control (QC) detects and fixes defects after they occur.

Analyze DeviationsReplan/ReallocateImprove ProcessesFeedback LoopMonitoringControlQuality AssuranceCorrective ActionsProcess Improvement
Flow of Monitoring → Control → Quality Assurance (QA) in project management

2. Techniques for Project Monitoring and Control

A. Earned Value Management (EVM)

EVM integrates scope, schedule, and cost into three key metrics:

  • Planned Value (PV): Budgeted cost of work scheduled (BCWS).
  • Earned Value (EV): Budgeted cost of work performed (BCWP).
  • Actual Cost (AC): Real cost incurred.
025005000750010000Planned Value (PV)10000Earned Value (EV)8000Actual Cost (AC)9500
EVM Example: EV < PV → Project behind schedule (Cost Variance = -1500)

Key Formulas:

Metric Formula Interpretation
Schedule Variance (SV) EV - PV Positive = Ahead of schedule; Negative = Behind
Cost Variance (CV) EV - AC Positive = Under budget; Negative = Over budget
Schedule Performance Index (SPI) EV / PV >1 = Good; <1 = Poor
Cost Performance Index (CPI) EV / AC >1 = Efficient; <1 = Inefficient

Worked Example: Ncell App Development

  • Planned Value (PV): Rs. 500,000 for 4 weeks (app backend development).
  • Earned Value (EV): Only 60% of work done after 4 weeks (EV = Rs. 300,000).
  • Actual Cost (AC): Rs. 400,000 spent. Calculations:
  • SV = 300,000 - 500,000 = -200,000 (2 weeks behind schedule).
  • CPI = 300,000 / 400,000 = 0.75 (25% over budget). Action: Hire more developers or extend the deadline.

B. Critical Path Method (CPM)

The longest path in a project network diagram determines the minimum project duration. Any delay on this path delays the entire project.

Example: eSewa Payment Gateway Upgrade

graph TD
    A["Design API"] -->|"3 weeks"| B["Develop Backend"]
    A -->|"2 weeks"| C["Test UI"]
    B -->|"4 weeks"| D["Integrate with Banks"]
    C -->|"3 weeks"| D
    D -->|"2 weeks"| E["Deploy"]

Critical Path: A → B → D → E (Total = 11 weeks). Non-Critical Path: A → C → D (Float = 2 weeks).

Week 1-3Design API (A)Week 3-7Develop Backend(B)Week 2-5Test UI (C)Week 7-9Integrate withBanks (D)Week 9-11Deploy (E)
Critical Path (A→B→D→E = 11 weeks) vs. Non-Critical Path (A→C→D = 9 weeks, 2-week float)

C. Gantt Charts and Pert Charts

  • Gantt Charts: Visualize task timelines, dependencies, and resource allocation.
  • PERT Charts: Focus on probabilistic time estimates (optimistic, pessimistic, most likely).

Comparison Table:

Feature Gantt Chart PERT Chart
Purpose Task scheduling & resource tracking Time estimation & risk analysis
Time Estimate Fixed dates Triangular distribution (O, P, M)
Dependencies Clear arrows Probabilistic paths
Use Case Construction, software sprints R&D projects, high-risk ventures

Gantt chart template**Sample for a Daraz delivery project (Image: Alex Thompson and Derrek Harrison, Public domain, via Wikimedia Commons)


3. Quality Assurance Techniques

A. Preventive Techniques

  1. Reviews and Inspections
    • Fagan Inspection: Formal review with moderators, readers, and authors.
    • Walkthroughs: Informal group review of code/documentation.
  2. Static Testing
    • Code Reviews: Peer checks for bugs, security flaws, or inefficiencies.
    • Static Analysis Tools: SonarQube, Checkstyle (detects coding standard violations).
  3. Process Improvements
    • CMMI (Capability Maturity Model Integration): Levels 1–5 for process maturity.
    • Six Sigma: Reduces defects to 3.4 per million (used by banks like NMB).

B. Corrective Techniques

  1. Dynamic Testing
    • Unit Testing: JUnit, pytest (individual components).
    • Integration Testing: Verify modules work together (e.g., WhatsApp API + Database).
    • System Testing: End-to-end validation (e.g., Pathao fare calculation).
  2. Defect Tracking
    • Bug Tracking Tools: Jira, Bugzilla (prioritize fixes using MoSCoW method).
    • Root Cause Analysis (RCA): Fishbone diagram to identify defect sources.

C. Quality Models

Model Factors Measured Example Application
McCall’s Model Product operation, revision, transition eSewa app usability & scalability
ISO 9001 Process consistency, customer focus NTC network reliability
Boehm’s Model Correctness, reliability, efficiency NEPSE trading platform stability

4. Modern Project Management Practices

A. Agile Monitoring

  • Burndown Charts: Track work remaining in a sprint (e.g., Daraz’s daily standups).
  • Velocity Tracking: Average story points completed per sprint (e.g., Khalti’s 2-week cycles).
  • Kanban Boards: Visualize workflow stages (To Do, In Progress, Done).
Sprint 1: 21 ptsSprint 2: 30 ptsSprint 3: 25 ptsCurrent Velocity
Velocity Tracking for Khalti’s 2-week sprints (Average = 25.3 pts/sprint)
To DoBacklog ItemsIn ProgressActive TasksDoneCompleted Stories
Kanban Board for Pathao Driver App Feature (Work-in-Progress limit = 3 tasks)

B. DevOps and Continuous Monitoring

  • CI/CD Pipelines: Automated testing/deployment (e.g., Google’s SRE practices).
  • Real-Time Dashboards: Grafana, Prometheus (monitor server health, API latency).

C. Risk-Based Monitoring

  • Risk Register: Track probability/impact of risks (e.g., Ncell network outage during Diwali).
  • Monte Carlo Simulation: Predict project outcomes under uncertainty.

5. Tools for Monitoring and Control

Tool Category Tools Examples Use Case
Scheduling Microsoft Project, GanttProject Timeline planning for NEPSE apps
Issue Tracking Jira, Trello Bug fixes in eSewa
Collaboration Slack, Microsoft Teams Daily standups for Daraz teams
Quality Metrics SonarQube, Splunk Code quality in Ncell apps
Financial QuickBooks, Excel ROI calculation for Pathao

In the Real World

  1. eSewa’s Quality Gates

    • Before every transaction, eSewa runs multi-layered testing (unit → integration → security) to prevent fraud. Their critical path includes bank API integration, which must be 100% reliable during Dashain sales peaks.
    • Real Example: In 2022, eSewa’s burndown chart showed a sprint delay due to a third-party bank API failure. They rerouted tasks to their backup payment processor (Khalti) and recovered on time.
  2. Pathao’s Driver App Monitoring

    • Pathao uses real-time dashboards to track:
      • Driver availability (SPI: % of drivers online vs. planned).
      • Ride completion time (CPI: cost per ride vs. budgeted driver earnings).
    • Their critical path is ride matching algorithm → GPS tracking → payment processing, where a delay in any step causes user churn.
  3. NTC’s Network Reliability (ISO 9001)

    • NTC monitors packet loss and latency using control charts to ensure 99.9% uptime. If metrics deviate (e.g., SV = -2 hours during a festival), they reroute traffic via backup fiber cables.
    • IMAGE: fiber optic cable | NTC’s backup fiber route in Kathmandu

Exam Tip

  1. For EVM Questions:

    • Always calculate SV, CV, SPI, CPI and interpret them (e.g., "The project is 15% over budget").
    • Example: If PV = Rs. 200K, EV = Rs. 150K, AC = Rs. 180K, then:
      • SV = -50K (behind schedule).
      • CPI = 0.83 (over budget).
      • Action: "Reallocate resources to Task C or extend deadline by 2 weeks."
  2. For Critical Path:

    • Draw the AOA (Activity-on-Arrow) or AON (Activity-on-Node) diagram.
    • Example: Given:
      Task A (2w) → Task B (3w)
      Task A → Task C (4w) → Task D (1w)
      
      Critical Path: A → C → D (7 weeks).
  3. For Quality Techniques:

    • Preventive: Mention reviews, inspections, CMMI.
    • Corrective: Mention testing (unit → system), defect tracking (Jira), RCA.
    • Models: Link McCall’s to product quality, ISO 9001 to process quality.
  4. For Agile vs. Traditional:

    • Agile: Burndown charts, velocity, Kanban.
    • Traditional: Gantt charts, milestone tracking, risk registers.
  5. Numerical Questions:

    • ROI: (Net Profit / Cost) × 100.
      • Example: Initial investment = Rs. 5L, benefit = Rs. 10L over 5 years. ROI = (10L - 5L) / 5L × 100 = 100%.
    • NPV: Use formula NPV = Σ [CFt / (1 + r)^t] - Initial Investment.
      • Example: If CF = Rs. 2L/year for 5 years, r = 10%: NPV = 2/(1.1)^1 + 2/(1.1)^2 + ... - 5L ≈ Rs. 3.7L (positive → invest).

Final Note: Always relate answers to Nepalese contexts (e.g., Daraz logistics, Ncell apps, NEPSE trading). Use real numbers (Rs. values, weeks) in examples to score full marks.

Based on the TU BIT syllabus for Software Project Management (BIT402), unit 7.

Discussion

Loading…