Mathematical Modeling – CO2011 Symbolic and Algebraic Reasoning in Petri Nets (Toolkit)

Overview

• Parse PNML 1-safe P/T nets
• Explicit reachability (BFS)
• Symbolic reachability with BDD
• Deadlock detection via ILP + BDD
• Linear optimization over reachable markings

Quick start

1. Create a virtual environment (recommended) python3 -m venv .venv
2. Activate the environment macOS/Linux (bash/zsh): . .venv/bin/activate fish: . .venv/bin/activate.fish Windows PowerShell: .venv\Scripts\Activate.ps1
3. Install dependencies python -m pip install --upgrade pip python -m pip install -r requirements.txt
4. Run commands (after activation) Explicit reachability: python -m app.cli explicit --pnml pnml/mutual_exclusion.pnml BDD reachability: python -m app.cli bdd --pnml pnml/mutual_exclusion.pnml Deadlock detection: python -m app.cli deadlock --pnml pnml/deadlock_reachable.pnml Optimization example: python -m app.cli optimize --pnml pnml/choice_conflict.pnml --weights p1=1,p2=2

Structure

• app/: source code
• docs/: short readmes for each task
• pnml/: example models (provided)

Notes

• Models are assumed 1-safe as per assignment.
• If CBC solver is missing, PuLP will try to use the default solver. See docs if needed.

• explicit: enumerate reachable markings with BFS python -m app.cli explicit --pnml
• bdd: symbolic reachability with BDD, reports count python -m app.cli bdd --pnml
• deadlock: ILP+BDD search for a reachable dead marking python -m app.cli deadlock --pnml
• optimize: maximize linear objective over reachable markings python -m app.cli optimize --pnml --weights p1=2,p2=-1

• ModuleNotFoundError (dd or pulp): ensure the venv is activated and run: python -m pip install -r requirements.txt
• Slow or timed-out installs: rerun the install; optionally: python -m pip install --upgrade pip setuptools wheel
• CBC not found (rare): install COIN-OR CBC, e.g. on macOS: brew install coin-or-tools/coinor/cbc or via conda: conda install -c conda-forge coincbc
• PowerShell activation policy (Windows): Set-ExecutionPolicy -Scope Process -ExecutionPolicy Bypass