software project management montecarloscheduleanalysis.pdf
- 1. Monte Carlo Schedule Analysis The Concept, Benefits and Limitations Intaver Institute 400, 7015, Macleod Trail S.W., Calgary, Alberta, T2H 2K6, Canada
- 2. What is Monte Carlo Analysis? Monte Carlo simulation is a mathematical method used in risk analysis. Monte Carlo simulations are used to approximate the distribution of potential results based on probabilistic inputs.
- 3. Monte Carlo Simulations Input Parameters Output Parameters Calculation Engine Critical Path Scheduling Engine ( ) Task duration cost, finish time, etc. cost, finish time, etc. Project duration Monte Carlo simulations use distributions as inputs, which are also the results
- 4. Monte Carlo Schedule Analysis 4 5 6 3 2 1 7 7 8 2 3 6 5 4 1 4 5 6 3 2 7 8 9 10 11 12 13 14 15 16 1 2 3 4 5 6 7 Task 1 Task 2 Task 3 Monte Carlo simulations take multiple distributions and create histograms to depict the results of the analysis
- 5. Two Approaches to Estimating Probabilities • The relative frequency approach, where probability equals the number of occurrences of specific outcome (or event) divided by the total number of possible outcomes. • The subjective approach represents an expert’s degree of belief that a particular outcome will occur.
- 6. Two of Approaches for Defining Uncertainties • Distribution-based approach • Event-based approach • Monte Carlo can be used to simulate the results of discrete risk events with probability and impact on multiple activities
- 7. What Distribution Should Be Used? Normal Triangual Uniform Also useful for Monte Carlo simulations: • Lognornal • Beta
- 8. Ignoring Base-Rate Frequencies • Historically, the probability that a particular component will be defective is 1%. • The component is tested before installation. • The test showed that the component is defective. • The test usually successfully identifies defective components 80% of the time. • What is the probability that a component is defective? The correct answer is close to 4%, however, most people would think that answer is a little bit lower than 80%.
- 9. Role of Emotions Emotions can affect our judgment
- 10. Eliciting Judgment About Probabilities of Single Events • Pose a direct question: “What is the probability that the project will be canceled due to budgetary problems?” • Ask the experts two opposing questions: (1) “What is the probability that the project will be canceled?” and (2) “What is the probability the project will be completed?” The sum of these two assessments should be 100%. • Break compound events into simple events and review them separately.
- 11. Probability Wheel 25% No delay of activity 35% 3 day delay of activity 40% 5 day delay of activity Use of visual aids like a probability wheel can aid in the increasing validity of estimates
- 12. Task Duration 4 8 12 16 20 100% 80% 60% 40% 20% Frequency Probability 2 3 4 5 6 (days) Question: What is the chance that duration is less than 3 days? Eliciting Judgment: Probability Method
- 13. Eliciting Judgment: Method of Relative Heights Task Duration 2 4 6 8 10 2 3 4 5 6 50% 40% 30% 20% 10% Frequency Probability (days) Question: How many times the duration will be between 2 and 3 days? Plotting possible estimates on a histogram can help improve estimatesc
- 14. How Many Trials Are Required? • Huge number of trials (> 1000) usually does not increase accuracy of analysis • Incorporate rare events • Use convergence monitoring
- 15. What Is The Chance That a Project Will Be on Time And Within Budget?
- 16. Analysis of Monte Carlo Results • Sensitivity and Correlations • Critical Indices • Crucial tasks • Critical Risks • Probabilistic Calendars • Deadlines • Conditional Branching • Probabilistic Branching • Chance of Task Existence
- 17. Crucial Tasks Crucial tasks for project duration Monte Carlo analysis identifies task cruciality, how often tasks are on the critical path.
- 18. Critical Risks
- 19. Conditional Branching 6 days If duration <= 6 days If duration > 6 days
- 20. Monte Carlo and Critical Chain Monitoring Project Buffer
- 21. Tracking Chance of Project Meeting a Deadline Project Duration Chance of project meeting a dealine 0% 20% 40% 60% 80% 100% (weeks) 0 2 4 6 8 10 12 14 Chance to meet a deadline is reducing as a results of events Mitigation efforts can increase a chance to meet a deadline
- 22. When Monte Carlo Is Useful • You have reliable historical data • You have tools to track actual data for each phase of the project • You have a group of experts who understand the project, have experience in similar projects, and are trained to avoid cognitive and motivational biases
- 23. Additional Resources 23 Project Think: Why Good Managers Make Poor Project Choices Project Decisions: The Art and Science Introduction to Project Risk Management and Decision Analysis Project Risk Analysis Made Ridiculously Simple
- 24. Questions?