VRPy

VRPy is a python framework for solving Vehicle Routing Problems (VRP) including:

• the Capacitated VRP (CVRP),
• the CVRP with resource constraints,
• the CVRP with time windows (CVRPTW),
• the CVRP with simultaneous distribution and collection (CVRPSDC),
• the CVRP with heterogeneous fleet (HFCVRP).

Check out the docs to find more variants and options.

Simple example

from networkx import DiGraph
from vrpy import VehicleRoutingProblem

# Define the network
G = DiGraph()
G.add_edge("Source",1,cost=1,time=2)
G.add_edge("Source",2,cost=2,time=1)
G.add_edge(1,"Sink",cost=0,time=2)
G.add_edge(2,"Sink",cost=2,time=3)
G.add_edge(1,2,cost=1,time=1)
G.add_edge(2,1,cost=1,time=1)

# Define the customers demands
G.nodes[1]["demand"] = 5
G.nodes[2]["demand"] = 4

# Define the Vehicle Routing Problem
prob = VehicleRoutingProblem(G, load_capacity=10, duration=5)

# Solve and display solution value
prob.solve()
print(prob.best_value)
3
print(prob.best_routes)
{1: ["Source",2,1,"Sink"]}

Install

pip install vrpy

Requirements

cspy

NetworkX

numpy

PuLP

Documentation

Documentation is found here.

Running the tests

Unit Tests

python3 -m pytest tests/

Benchmarks

To run some non-regression tests on some benchmarks instances (Solomon and Augerat) do

python3 -m pytest benchmarks/

Note that running the benchmarks requires pandas and that it takes a while.

For more information and to run more instances, see the benchmarks.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Bugs

Please report any bugs that you find here. Or, even better, fork the repository on GitHub and create a pull request. Please read the Community Guidelines before contributing. Any contributions are welcome.