Skip to content

README.md

Latest commit

 

History

History
76 lines (49 loc) · 3.02 KB

README.md

File metadata and controls

76 lines (49 loc) · 3.02 KB

VRP Algorithms

A collection of algorithms and tools to solve Vehicle Routing Problems (VRP). This repository contains implementations, datasets and result exports for multiple heuristics and metaheuristics.

Key features

  • Implementations of classic VRP algorithms and heuristics (e.g. nearest neighbor, Clarke-Wright savings, genetic algorithms, GRASP)
  • Dataset collection with different sizes (datasets/small, datasets/medium, datasets/large, datasets/xlarge, datasets/raw)
  • Scripts and utilities under src/ for running algorithms and analysing results
  • Precomputed results and charts in results/

Requirements

  • Python 3.11
  • Recommended: Poetry to manage dependencies

Main dependencies (from pyproject.toml): tsplib95, logger. Development tools listed: black, isort, pylint.

Installation (Windows)

  1. Clone the repository:

    • git clone https://github.com/Guiners/VRP-Algorithms
    • cd VRP-Algorithms

  2. Install dependencies with Poetry:

    • poetry install
    • Optionally enter the virtual environment: poetry shell

If you do not use Poetry, create and activate a venv and install required packages manually.

Usage

  • Entry point: src/main.py

Example (using Poetry):

  • poetry run python src\main.py --input datasets\small\vpr101_3_1.csv --algorithm nearest_neighbor

Notes:

  • Available algorithms correspond to folders in src\algorithms\ and to result folders in results\ (e.g. nearest_neighbor_algorithm, clarke_wright_savings, genetic_algorithm, grasp).
  • Many scripts accept dataset file paths from the datasets/ directory. Adjust CLI arguments according to the implemented options in src/main.py.

Project structure (high level)

  • src/ - source code
    • src\main.py - main entry point
    • src\algorithms\ - algorithm implementations
    • src\data_analyst\ - result analysis and plotting
    • src\utils\ - utility helpers
  • datasets/ - problem instances (small / medium / large / xlarge / raw)
  • results/ - saved outputs, charts and tables
  • pyproject.toml - project metadata and dependencies
  • README.md - this file

Datasets

Datasets are provided in the datasets/ folder and grouped by size:

  • datasets/small, datasets/medium, datasets/large, datasets/xlarge, datasets/raw

Each dataset typically has a CSV and a JSON description (for example datasets/small/vpr101_3_1.csv and datasets/small/vpr101_3_1.json).

Results and charts

Precomputed results and visualizations are available under results/. Subfolders group outputs per algorithm and per dataset size.

Contributing

  • Open issues for bugs or feature requests.
  • Pull requests are welcome. Follow repository code style and run linters/formatters if configured.

Tests

  • No automated tests included in the repository root. Add a tests folder and CI configuration to enable test runs.

License

  • No license file found. Add a LICENSE file to define terms for reuse.