InteractiveMap of Operations Research
Back to Uncertainty & Stochastic Systems
MethodIntermediate03.03
Uncertainty & Stochastic Systems

Simulation

Imitating system behavior when analysis is too hard.

Overview

Simulation is the OR tool for systems whose operational detail is too rich for a clean closed-form model. Instead of solving equations directly, it imitates system behavior under policies and random inputs.

A simulation study is strongest when it compares policies under common random scenarios, reports uncertainty intervals, and validates the model against known behavior.

Core ideas

Monte Carlo

Uses repeated random sampling to estimate distributions, tail risks, and expected performance.

Discrete-event

Advances the clock from event to event, which is natural for queues, factories, clinics, and logistics.

Agent-based

Agent-based is a core checkpoint for Simulation: define it concretely, attach units or rules where possible, and test whether stakeholders interpret it the same way.

Variance reduction

Variance reduction is a core checkpoint for Simulation: define it concretely, attach units or rules where possible, and test whether stakeholders interpret it the same way.

Validation

Validation is a core checkpoint for Simulation: define it concretely, attach units or rules where possible, and test whether stakeholders interpret it the same way.

How to use it

  1. 1Start with Operations: write the decision, time horizon, actors, and objective in operational language.
  2. 2Translate the problem into Monte Carlo, Discrete-event, and Agent-based; define units and data sources for each one.
  3. 3Build a small instance of Simulation that can be solved or simulated by hand inspection before using full production data.
  4. 4Compare the recommendation against a baseline policy, not just against mathematical optimality.
  5. 5Document assumptions, sensitivity results, and the conditions under which the recommendation should be revisited.

Applications

OperationsRiskHealthcareLogisticsNetworks
  • Operations: compare feasible policies, quantify the operating tradeoffs, and make the assumptions behind the recommendation visible.
  • Risk: compare feasible policies, quantify the operating tradeoffs, and make the assumptions behind the recommendation visible.
  • Healthcare: compare feasible policies, quantify the operating tradeoffs, and make the assumptions behind the recommendation visible.
  • Logistics: compare feasible policies, quantify the operating tradeoffs, and make the assumptions behind the recommendation visible.
  • Networks: compare feasible policies, quantify the operating tradeoffs, and make the assumptions behind the recommendation visible.

Common pitfalls

  • Applying Simulation because the label sounds appropriate while leaving the actual decision boundary vague.
  • Treating Monte Carlo as a technical detail instead of a modeling choice that affects the recommendation.
  • Reporting one answer without showing sensitivity to demand, capacity, costs, or behavioral assumptions.
  • Ignoring implementation details such as data quality, explainability, ownership, and how users will override bad recommendations.

Resources