A unified-adaptive large neighborhood search metaheuristic for periodic location-routing problems |
| |
Affiliation: | 1. CIRRELT, Canada Research Chair in Distribution Management and HEC Montréal, Montréal, Canada;2. Southampton Business School and CORMSIS, University of Southampton, Southampton, United Kingdom;3. CIRRELT and HEC Montréal, Montréal, Canada |
| |
Abstract: | This paper introduces three variants of the Periodic Location-Routing Problem (PLRP): the Heterogeneous PLRP with Time Windows (HPTW), the Heterogeneous PLRP (HP) and the homogeneous PLRP with Time Windows (PTW). These problems extend the well-known location-routing problem by considering a homogeneous or heterogeneous fleet, multiple periods and time windows. The paper develops a powerful Unified-Adaptive Large Neighborhood Search (U-ALNS) metaheuristic for these problems. The U-ALNS successfully uses existing algorithmic procedures and also offers a number of new advanced efficient procedures capable of handling a multi-period horizon, fleet composition and location decisions. Computational experiments on benchmark instances show that the U-ALNS is highly effective on PLRPs. The U-ALNS outperforms previous methods on a set of standard benchmark instances for the PLRP. We also present new benchmark results for the PLRP, HPTW, HP and PTW. |
| |
Keywords: | Location-routing Periodic Heterogeneous fleet Time windows Adaptive large neighborhood search |
本文献已被 ScienceDirect 等数据库收录! |
|