Survey on Neural Routing Solvers
This is an incremental survey that addresses evaluation issues for researchers and practitioners in optimization and machine learning.
This survey tackles the problem of evaluating neural routing solvers (NRSs) for vehicle routing by highlighting their heuristic nature and proposing a new generalization-focused evaluation pipeline, revealing gaps in current research through comparative benchmarking.
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.