mtspbc

ALNS-based solver for the multiple Travelling Salesman Problem with Backup Coverage (mTSP-BC).

Supplemental material for:

An Adaptive Large Neighborhood Search for the Multiple Traveling Salesman Problem with Backup Coverage — Cardozo, Dhein and Araújo
DOI: 0.1109/ACCESS.2026.XXXXXXX

Compute capsule with reproductible execution provisioned in Code Ocean platform. DOI: 10.24433/CO.2463253.v2

Installation

Requires a C++17 compiler, Python ≥ 3.9, and pybind11.

pip install pybind11 numpy
pip install -e .

Usage

import mtspbc
mtspbc.solve("path/to/instance.BC")

Or from the command line:

mtspbc path/to/instance.BC

Instance format

m  n  d
x0 y0
x1 y1
...

First line: number of vehicles m, number of clients n, maximum coverage distance d.
Followed by n+1 coordinate pairs (depot first).

Acknowledgements

The ALNS framework used here is adapted from the open-source alns Python package by Wouter Kool, Niels Wouda, and Leon Lan.