EVRPGen: A web-based instance generator for the electric vehicle routing problem with road junctions and road types

This paper presents a web-based instance generator for Electric Vehicle Routing Problems (EVRP) with Road Junctions and Road Types, using OpenStreetMap data. Users define an area, specify network components (depots, customers, charging stations, junctions), and customize vehicle parameters. The React-based frontend enables configuration, visualization, and queries, while the Flask backend processes road networks, classifies road types, and assigns demand and service times. A RESTful API ensures real-time instance generation. Generated instances can be downloaded as text-based datasets and interactive visualizations. The tool is open-source and contributes to the area of sustainable transportation by enabling scenario-based EVRP experimentation. • EVRPGen is a web-based instance generator for the Electric Vehicle Routing Problem with Road Junctions and Road Types (EVRP-RJ-RT). • It extracts and processes real-world road network data from OpenStreetMap (OSM) to generate realistic problem instances. • The software provides a user-friendly interface built with React and a backend implemented in Python using Flask, OSMnx, and NetworkX. • Customizable parameters include network components (depots, customers, charging stations, junc- tions) and vehicle specifications (battery capacity, load capacity, recharge rates). • The generated instances can be downloaded as structured text-based datasets and visualized in- teractively via an HTML-based map. • The tool facilitates benchmarking and experimentation for sustainable logistics and EV routing solutions. • EVRPGen is open-source and accessible at https://github.com/manilakbay/InstanceGenerator .