Skip to content

Simulation Model

Understanding how Waremax simulates warehouse operations.

Overview

Waremax uses Discrete Event Simulation (DES), a modeling paradigm where the system state changes only at discrete points in time called events.

Topic Description
Discrete Event Simulation DES fundamentals
Time Model How time advances
Event Types Types of events
Determinism & Seeds Reproducibility

Why Discrete Event Simulation?

Compared to Time-Stepping

Time-Stepping (not used):

  • Advances time in fixed increments (e.g., 1 second)
  • Checks for changes every step
  • Wastes computation when nothing happens

Discrete Event Simulation (used):

  • Jumps directly to next meaningful event
  • No wasted computation
  • Efficient for sparse events

Example

If a robot takes 30 seconds to travel:

Approach Steps
Time-stepping (1s) 30 steps
DES 2 events (depart, arrive)

Core Components

Event Queue

A priority queue ordered by event time:

Event Queue:
  t=10.5s: RobotArrive(robot_0, node_5)
  t=12.3s: OrderArrival(order_42)
  t=15.0s: ServiceEnd(robot_2, station_1)
  ...

Simulation Loop

while queue not empty and time < end_time:
    event = queue.pop_next()
    advance_time(event.time)
    process(event)
    # Processing may schedule new events

State Updates

Events modify simulation state:

  • Robot positions
  • Queue contents
  • Inventory levels
  • Metrics counters

Event Flow Example

Order fulfillment flow:

graph TD
    A[OrderArrival] --> B[TaskAssignment]
    B --> C[RobotDepart]
    C --> D[RobotArrive at pickup]
    D --> E[PickupComplete]
    E --> F[RobotDepart to station]
    F --> G[RobotArrive at station]
    G --> H[QueueJoin]
    H --> I[ServiceStart]
    I --> J[ServiceEnd]
    J --> K[TaskComplete]

Key Properties

Determinism

Same seed + same configuration = identical results.

This enables:

  • Reproducible experiments
  • Fair comparisons
  • Debugging

Efficiency

  • O(log n) event scheduling
  • No idle computation
  • Scales to complex scenarios

Accuracy

  • Events happen at exact times
  • No time discretization error
  • Realistic sequencing

Next Steps