Survey on Neural Routing Solvers

2026-02-25

AI Summary

This study addresses the limitations of traditional vehicle routing problem solvers, which rely on manually designed heuristic rules that are costly to develop and exhibit limited generalization. For the first time, the work establishes a hierarchical taxonomy of neural routing solvers grounded in heuristic principles and introduces a systematic evaluation framework centered on generalization capability. By comparing existing benchmarks with this new evaluation protocol, the study uncovers a significant yet long-overlooked gap in the cross-scenario generalization performance of current neural solvers. This contribution provides the neural combinatorial optimization community with a novel taxonomic perspective and a standardized benchmark for rigorous and meaningful assessment of generalization.

Technology Category

• Search and Optimization: Heuristic Search
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling
• Constraint Satisfaction and Optimization: Solvers and Tools

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

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.

Problem

• Neural Routing Solvers
• Vehicle Routing Problems
• Heuristics
• Generalization
• Evaluation Pipeline

Innovation

• Neural Routing Solvers
• Heuristic Frameworks
• Hierarchical Taxonomy
• Generalization Evaluation
• Vehicle Routing Problems