Survey on Neural Routing Solvers

TL;DR

This survey reviews neural routing solvers that use deep learning for vehicle routing, highlighting their heuristic nature, taxonomy, and proposing a new evaluation pipeline to identify research gaps.

Contribution

It introduces a hierarchical taxonomy of NRSs based on heuristic principles and proposes a generalization-focused evaluation pipeline.

Findings

Uncovered gaps in current NRS research through benchmarking.

Highlighted the heuristic nature of neural routing solvers.

Proposed a new evaluation pipeline for better generalization assessment.

Abstract

Neural routing solvers (NRSs) that leverage deep learning to tackle vehicle routing problems have demonstrated notable potential for practical applications. By learning implicit heuristic rules from data, NRSs replace the handcrafted counterparts in classic heuristic frameworks, thereby reducing reliance on costly manual design and trial-and-error adjustments. This survey makes two main contributions: (1) The heuristic nature of NRSs is highlighted, and existing NRSs are reviewed from the perspective of heuristics. A hierarchical taxonomy based on heuristic principles is further introduced. (2) A generalization-focused evaluation pipeline is proposed to address limitations of the conventional pipeline. Comparative benchmarking of representative NRSs across both pipelines uncovers a series of previously unreported gaps in current research.