Efficient Neural Neighborhood Search for Pickup and Delivery Problems

TL;DR

This paper introduces an efficient neural neighborhood search method for pickup and delivery problems, utilizing a novel attention mechanism and specialized decoders to improve solution quality and outperform existing neural and traditional solvers.

Contribution

The paper proposes a novel neural neighborhood search framework with synthesis attention and customized decoders, achieving state-of-the-art results on PDP variants and surpassing LKH3.

Findings

Achieves state-of-the-art results on two PDP variants.

Outperforms the LKH3 solver on constrained PDP.

Demonstrates the effectiveness of synthesis attention and specialized decoders.

Abstract

We present an efficient Neural Neighborhood Search (N2S) approach for pickup and delivery problems (PDPs). In specific, we design a powerful Synthesis Attention that allows the vanilla self-attention to synthesize various types of features regarding a route solution. We also exploit two customized decoders that automatically learn to perform removal and reinsertion of a pickup-delivery node pair to tackle the precedence constraint. Additionally, a diversity enhancement scheme is leveraged to further ameliorate the performance. Our N2S is generic, and extensive experiments on two canonical PDP variants show that it can produce state-of-the-art results among existing neural methods. Moreover, it even outstrips the well-known LKH3 solver on the more constrained PDP variant. Our implementation for N2S is available online.