VRPBench: A Vehicle Routing Benchmark Tool

TL;DR

VRPBench is a new benchmarking tool for the Vehicle Routing Problem that generates realistic instances from real-world maps, aiding the validation and comparison of optimization algorithms.

Contribution

It introduces VRPBench, a tool for creating and visualizing VRP instances based on real city maps, filling a gap in benchmark resources for routing problems.

Findings

Created a challenging VRP benchmark from a real city map

Generated instances modeling a full day of mail delivery

Facilitated comparison and validation of routing algorithms

Abstract

The number of optimization techniques in the combinatorial domain is large and diversified. Nevertheless, there is still a lack of real benchmarks to validate optimization algorithms. In this work we introduce VRPBench, a tool to create instances and visualize solutions to the Vehicle Routing Problem (VRP) in a planar graph embedded in the Euclidean 2D space. We use VRPBench to model a real-world mail delivery case of the city of Artur Nogueira. Such scenarios were characterized as a multi-objective optimization of the VRP. We extracted a weighted graph from a digital map of the city to create a challenging benchmark for the VRP. Each instance models one generic day of mail delivery with hundreds to thousands of delivery points, thus allowing both the comparison and validation of optimization algorithms for routing problems.