Description

This repository contains a collection of scripts and jupyter notebooks to perform semi-automatic analysis of Vehicle Routing Problem Solver algorithm behavior.

How to use

Preconditions

• install python
• install jupyter notebook
• install dependencies from requirements.txt
• install/build VRP solver

Prepare data

• create a new folder for experiment, e.g. experiments/my_experiment, with two subfolders:
  • configs: should contain algorithm configuration files
  • problems: should contain VRP definitions in pragmatic format
• run runner.py script and pass the following parameters:
  • path to vrp-cli executable
  • path to experiment folder
  • sample size: how many times each problem should be run with each config

Please note, that the solver uses routing matrix approximation via great-circle distances and constant speed. Check solver documentation to see how to configure that behaviour.

Analyze results

• run jupyter notebook
• select one of available notebooks
• change path to test data
• run kernel