Module 07
Verification, Validation, and Decision Support
Build credibility, design experiments, and communicate results — the final module before the integrative capstone.
§ Overview
This module closes the loop. Verification asks whether the implementation matches the specification; validation asks whether the model represents the real system well enough for its purpose. Around those two activities sit sensitivity analysis, uncertainty quantification, robustness testing, and the communication skills that make a model useful to someone else.
Unit 7.1
Model credibility.
Unit 7.2
Integrative simulation practice.
Laboratory 13
Verification and validation.
Capstone
Integrated computational modelling project.
§ Learning outcomes
By the end of Module 7 you will be able to
- Verify model implementation against its specification using systematic strategies.
- Validate results against data, analytical solutions, and limiting cases.
- Conduct sensitivity, uncertainty, and robustness analyses and interpret their implications.
- Design scenario experiments rather than running one-off trials.
- Create reproducible computational workflows and archive them.
- Communicate simulation assumptions, results, limitations, and recommendations clearly to technical and non-technical audiences.
7.1 Model Credibility
Credibility is earned through documented evidence, not asserted. This unit covers verification, validation, and the sensitivity and uncertainty analyses that establish how much weight a model can bear.
Verification strategies
Test the implementation against the specification, not the data.
Lesson 2Validation evidence and model comparison
Compare against data, analytical solutions, and limiting cases.
Lesson 3Sensitivity, uncertainty, and robustness
Identify which inputs actually drive the outputs, and by how much.
7.2 Integrative Simulation Practice
The capstone is a modelling project from end to end. This unit covers the experimental planning, communication, and integration skills that make it work.
Scenario design and experimental planning
Design experiments rather than running one-off trials.
Lesson 2Technical and non-technical communication
Write for the decision-maker as well as the analyst.
Lesson 3Simulation capstone integration
Assemble the full modelling cycle into one coherent project.
§ Laboratory
Verification and validation
Apply verification and validation strategies to a model of your choice and document the evidence for each claim.
§ Simulation capstone
Integrated computational modelling project · 45% + 5%
Construct, verify, validate, analyze, and communicate a substantial computational model. The capstone carries the largest single weight in the course because it demonstrates integrated modelling competence across every module.
Suggested application areas
- Health-service queues
- Inventory systems
- Financial risk
- Energy storage
- Population or epidemic dynamics
- Diffusion or heat transfer
- Transportation flows
- Supply chains
- Reliability systems
- Resource allocation
- Environmental uncertainty
- Emergency-service operations
Core deliverables
- Project proposal and problem statement
- Conceptual model diagram
- Mathematical formulation
- Implementation and verification evidence
- Validation evidence
- Numerical error analysis
- Sensitivity and uncertainty analysis
- Scenario experiments and visualizations
- Limitations and ethical considerations
- Reproducible model files
- Technical report and executive summary
- Accessible presentation
§ Common pitfalls
Watch for these
- Verification and validation are different activities. A model that is verified is not thereby validated, and a model that is validated is not thereby verified.
- A simulation model should support decision-making without concealing uncertainty. Reporting a single number without its uncertainty undermines the purpose of the model.