Problem
Research questions and friction points this paper is trying to address.
Vehicle Routing Problem
Neural Constructive Solvers
Route-by-Route Construction
Global Coordination
Depot-Closed Multi-Component Construction for Neural Vehicle Routing
🤖 AI Summary
This study addresses the difficulty of global coordination inherent in traditional neural solvers that construct solutions route by route, proposing instead a multi-component construction paradigm. Methodologically, it achieves implicit tour closure by removing depot-return tokens, thereby enabling arbitrary-order merging while preserving intermediate-state feasibility. Furthermore, the approach integrates neural policies with the Clarke-Wright savings algorithm and a Ruin-and-Reconstruct mechanism to learn efficient merging strategies. Experimental results demonstrate that the proposed method significantly outperforms existing representative neural solvers on Capacitated Vehicle Routing Problem (CVRP) instances ranging from 100 to 1000 customers, as well as in zero-shot constraint generalization scenarios.
📝 Abstract
Most neural constructive solvers for the vehicle routing problem (VRP) use route-by-route construction, extending one route until completion before starting the next. This commits route membership early and hinders global coordination across routes. We propose multi-component construction, which maintains many route components simultaneously and merges them in an arbitrary order. This removes the depot-return cue that route-by-route construction obtains from the remaining capacity; to compensate, we introduce an interpretation in which every component is treated as an implicitly depot-closed route. Under this depot-closed interpretation, every intermediate state of standard CVRP construction is a complete feasible solution, and the exact cost reduction of a merge is the Clarke-Wright saving. The neural policy combines this CW-saving signal with the evolving component state to learn what to connect and when to connect. A policy trained only on CVRP100 outperforms the reported results of representative neural solvers on CVRP100-500 with greedy inference and, reused for ruin-and-reconstruct, performs strongly at all evaluated sizes up to CVRP1000. In a zero-shot Constraint Tightness evaluation with capacities from $C=10$ to $500$, it outperforms the reported neural solvers at every capacity. Controlled analyses show that robustness persists without CW grounding and point to learned route-closing behavior as a plausible contributor to the tight-regime degradation of learned route-by-route solvers.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Multi-Component Construction
Depot-Closed Interpretation
Clarke-Wright Saving
Neural Vehicle Routing
Zero-Shot Generalization
🔎 Similar Papers
No similar papers found.
