Vehicle-as-Prompt: A Unified Deep Reinforcement Learning Framework for Heterogeneous Fleet Vehicle Routing Problem

TL;DR

This paper introduces Vehicle-as-Prompt, a unified deep reinforcement learning framework that effectively solves complex heterogeneous fleet vehicle routing problems with diverse constraints, outperforming existing methods in solution quality and speed.

Contribution

The paper proposes the Vehicle-as-Prompt mechanism and VaP-CSMV framework, enabling a single model to handle various HFVRP variants with improved accuracy and efficiency.

Findings

VaP-CSMV outperforms existing DRL-based solvers.

Achieves solution quality comparable to traditional heuristics.

Reduces inference time to seconds and generalizes zero-shot.

Abstract

Unlike traditional homogeneous routing problems, the Heterogeneous Fleet Vehicle Routing Problem (HFVRP) involves heterogeneous fixed costs, variable travel costs, and capacity constraints, rendering solution quality highly sensitive to vehicle selection. Furthermore, real-world logistics applications often impose additional complex constraints, markedly increasing computational complexity. However, most existing Deep Reinforcement Learning (DRL)-based methods are restricted to homogeneous scenarios, leading to suboptimal performance when applied to HFVRP and its complex variants. To bridge this gap, we investigate HFVRP under complex constraints and develop a unified DRL framework capable of solving the problem across various variant settings. We introduce the Vehicle-as-Prompt (VaP) mechanism, which formulates the problem as a single-stage autoregressive decision process. Building on this, we propose VaP-CSMV, a framework featuring a cross-semantic encoder and a multi-view decoder that effectively addresses various problem variants and captures the complex mapping relationships between vehicle heterogeneity and customer node attributes. Extensive experimental results demonstrate that VaP-CSMV significantly outperforms existing state-of-the-art DRL-based neural solvers and achieves competitive solution quality compared to traditional heuristic solvers, while reducing inference time to mere seconds. Furthermore, the framework exhibits strong zero-shot generalization capabilities on large-scale and previously unseen problem variants, while ablation studies validate the vital contribution of each component.