simpy-sim

name	simpy-sim
description	Use when driving SimPy (process-based discrete-event simulation framework in pure Python) via Python scripts — queueing systems, manufacturing lines, network protocols, hospital flow, anything modeled as processes that wait on resources / timeouts / events — through sim runtime one-shot execution.

simpy-sim

You are connected to SimPy via sim-cli.

SimPy is the canonical Python DES (discrete-event simulation) framework. Process-based: simulation logic written as Python generator functions that yield simpy events. Pure Python, no compiled deps. Pip-installable (pip install simpy).

Scripts are plain .py:

import simpy

def car(env):
    while True:
        yield env.timeout(5)        # drive 5 time units
        print(f'arrived at {env.now}')

env = simpy.Environment()
env.process(car(env))
env.run(until=20)

Same subprocess driver mode as PyBaMM / PyMFEM.

---

base/ — always relevant

Path	What's there
base/reference/workflow.md	Environment → process generators → resources → run.
base/reference/resources.md	Resource / Container / Store / PriorityResource.
base/reference/events.md	timeout / event / process / all_of / any_of.
base/snippets/01_mm1.py	Verified M/M/1 queue E2E.
base/known_issues.md	Generators not coroutines, env.now is float, statistical noise.

sdk/4/ — SimPy 4.x

• sdk/4/notes.md — version-specific surface notes.

---

Hard constraints

1. Process functions are generators, not coroutines. Use yield env.timeout(...), yield env.process(other(env)), yield resource.request(), etc. NOT await.
2. env.run(until=T) is the entry point. Without until, it runs forever (until no more events are scheduled).
3. All time arithmetic is in arbitrary units — pick consistent units (seconds, hours, minutes) and document them.
4. Acceptance != "ran without error". Always run for enough time that statistics converge (rule of thumb: >= 10000 events) and validate against analytical (M/M/1 L = ρ/(1-ρ)) or against a simpler reference model.
5. Seed random for reproducibility: random.seed(42).
6. Print results as JSON on the last stdout line.

---

Required protocol

1. Gather inputs:
   • Category A: processes / arrival rate / service distribution / resource capacities / acceptance criterion (queue length, throughput, utilization, ...).
   • Category B: simulation horizon, RNG seed, warm-up period.
2. sim check simpy.
3. Write .py per base/reference/workflow.md.
4. sim lint script.py.
5. sim run script.py --solver simpy.
6. Validate JSON.