On methods for solving pickup and delivery problem with time windows. Part II: heuristic approach

Abstract

This paper focuses on the Pickup and Delivery Problem with Time Windows (PDPTW), an optimization problem arising in numerous logistics and transportation settings. To address this problem, we propose a variant of the Adaptive Large Neighborhood Search (ALNS) metaheuristic, which integrates a diverse set of problem-specific destroy and repair operators. The algorithm is further enhanced with a tabu-based memory mechanism to prevent cycling and improve exploration of the solution space. The proposed method demonstrates a favorable balance between solution quality and computational efficiency. Moreover, its modular structure makes it well-suited for adaptation to various extensions of the PDPTW, including stochastic and dynamic scenarios. Computational experiments on benchmark datasets from Li & Lim and Sartori confirm the effectiveness and robustness of the approach.