23-Ind-A6 Systems Simulation · December 2013
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
National Exams — December 2013 — 98-Ind-A6 Systems Simulation. Three-hour, closed-book exam; one of two permitted calculators (Sharp or Casio), one 8.5″×11.0″ aid sheet (both sides). Format: 14 sub-questions across four parts — Part A (do 1 of 2, 25 marks), Part B (do 3 of 5, 15 marks), Part C (do 1 of 3, 10 marks), Part D (do 2 of 4, 20 marks); 7 questions, 70 marks constitute a complete paper. All fourteen sub-questions are solved below for completeness (the source restarts its own numbering at 1 within each Part). Statistical tables (Normal, t, chi-square, F) were supplied with the exam; the values below are the same table values obtained by direct computation.
Reference texts: Banks, Carson, Nelson & Nicol, Discrete-Event System Simulation (5th ed., Pearson) — simulation study design, random-number generation, input/output data analysis, variance reduction, verification & validation, queueing simulation; Montgomery, Design and Analysis of Experiments (9th ed., Wiley) — factorial designs and ANOVA.
Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.
Verification asks "did we build the model RIGHT?" — does the computer program correctly implement the intended conceptual model, free of coding errors, logic bugs, and unintended approximations? Angus can verify his model by: structured code walkthroughs with a second engineer; running each sub-model (mode-choice logic, yard capacity check, subcontractor allocation) against hand-traced test cases with known expected output (e.g. force a fixed, non-random input sequence and check the model reproduces the hand calculation exactly); checking that output units and magnitudes are dimensionally sensible; and stress-testing boundary conditions (zero demand, yard at exactly 2000 m³).
Validation asks "did we build the RIGHT model?" — does the model's behaviour adequately represent the real system for the study's intended purpose, regardless of whether the code is bug-free? Since BSL's warship supply chain does not exist yet, Angus cannot validate the FULL model against historical operating data; practical validation available now includes: fitting and testing the transit-time input distributions against the pilot carrier data already collected (Questions 3–7); comparing a simplified sub-model (e.g. the yard as an M/M/2-type queue, Question 12) against known analytical results as a sanity check; sensitivity analysis to confirm the model responds to parameter changes in the expected direction and magnitude; and face validity — structured review of the model's assumptions and behaviour by BSL's own production planners and by the subcontractors themselves, who know the real process even though it hasn't started.
Model credibility is broader than validation: it is whether the model's DECISION-MAKERS (here, Ms. Feist and BSL management) actually TRUST the model enough to act on its recommendations. A model can be statistically validated on every sub-component and still lack credibility if management doesn't understand its assumptions, wasn't involved in defining its objectives, or has no track record with the modelling team. Angus can build credibility by involving management and the subcontractors throughout the study (not just at the final readout), documenting every assumption in language a non-modeller can assess, and presenting validation results (including the sub-model checks above) transparently rather than as a black-box conclusion; credibility is confirmed less by a single statistical test than by whether management is willing to commit capital (the yard-size decision) to the model's recommendation.
