Simulations Practice Test

A transportation agency uses a traffic flow simulation to test a new bus-only lane downtown. The real-world problem is predicting whether reallocating one lane improves total travel time or just shifts congestion. The simulation represents vehicles as agents that follow basic rules: maintain safe distance, accelerate when space opens, and stop at red lights. Variables include vehicleDensity (15–90 vehicles/km/lane), lightCycleTime (50–150 seconds), and peakHourFactor (1.0–3.0) that increases arrivals near 8 a.m. The algorithm iterates each second: it spawns vehicles based on peakHourFactor, updates positions and speeds, and records queue lengths at intersections. After a full simulated hour, it computes average delay per vehicle. To refine results, it repeats the hour-long simulation across many random arrival patterns, then modifies lightCycleTime to reduce the worst delays while keeping cross-street waits within a fairness limit. Scenario-specific example: at very high vehicleDensity, even a well-tuned lightCycleTime may not prevent gridlock if too many vehicles enter the network.

A county health department runs an epidemiological simulation to decide whether to reopen indoor sports. The real-world problem is balancing normal activities with preventing a surge that overwhelms clinics. The model uses daily steps and tracks Susceptible, Infected, and Recovered people. Variables include infectionRate (0.10–0.50 per close contact), recoveryRate (0.05–0.25 per day), and populationDensity (low/medium/high contact levels). Each day, the algorithm estimates how many susceptible people become infected based on contacts and infectionRate, then estimates recoveries based on recoveryRate. It repeats for 120 days. To refine results iteratively, analysts run 200 simulations per setting, average the infection curves, and then recalibrate infectionRate if the first two weeks of simulated cases do not match observed case counts. Scenario-specific example: if recoveryRate increases from 0.10 to 0.20, infected people leave the infected state faster, often reducing the peak number of simultaneous infections even if infectionRate remains unchanged.