This paper introduces RouteFinder, a comprehensive foundation model framework to tackle different Vehicle Routing Problem (VRP) variants. Our core idea is that a foundation model for VRPs should be able to represent variants by treating each as a subset of a generalized problem equipped with different attributes. We propose a unified VRP environment capable of efficiently handling any attribute combination. The RouteFinder model leverages a modern transformer-based encoder and global attribute embeddings to improve task representation. Additionally, we introduce two reinforcement learning techniques to enhance multi-task performance: mixed batch training, which enables training on different variants at once, and multi-variant reward normalization to balance different reward scales. Finally, we propose efficient adapter layers that enable fine-tuning for new variants with unseen attributes. Extensive experiments on 48 VRP variants show RouteFinder outperforms recent state-of-the-art learning methods. Code:this https URL.
View on arXiv@article{berto2025_2406.15007,
title={ RouteFinder: Towards Foundation Models for Vehicle Routing Problems },
author={ Federico Berto and Chuanbo Hua and Nayeli Gast Zepeda and André Hottung and Niels Wouda and Leon Lan and Junyoung Park and Kevin Tierney and Jinkyoo Park },
journal={arXiv preprint arXiv:2406.15007},
year={ 2025 }
}