Definition
Discrete Event Simulation (DES) is a simulation methodology where the model’s state changes at discrete points in time triggered by a chronological sequence of instantaneous events. Between events, the state variables of the system remain constant.
Why It Matters
Most of the time, nothing of analytical interest happens in a complex system; DES allows us to warp through these idle intervals by jumping between events. This makes it computationally possible to simulate years of high-stakes operation, such as hospital throughput or data center traffic, in just seconds.
Core Concepts
- Events: Instantaneous occurrences that cause a change in the system’s state (e.g., arrival of a customer, completion of service).
- State Variables: One or more variables that completely describe the system at any given moment (e.g., number of items in a queue).
- Future Event List (FEL): A sorted list (by timestamp) of all events scheduled to occur in the future.
- Clock Advancing: The simulation clock “jumps” from one event to the next, skipping the idle periods in between. This distinguishes it from fixed-increment time-step simulations.
- Combined Event Models: Systems that integrate both discrete jumps and continuous state changes (e.g., a coffee manufacturing plant where roasting is continuous by weight, but packaging is discrete by individual cases).
- Event Thresholds: In combined models, a continuous variable passing a threshold (e.g., bin level hitting 90%) triggers a discrete event (e.g., stopping a grinder).