Adaptive large neighborhood decomposition search algorithm for multi-allocation hub location routing problem

Abstract

In this study, we investigate a multi-allocation hub location routing problem (MAHLRP) for the design of an intra-city express service system, in which flows of mails and parcels are exchanged among the branch offices of the service provider via local tours and hubs. In this application, the pickup and delivery processes are handled simultaneously, and both hub capacity and vehicle capacity are considered. We propose a mixed integer programming formulation for this variant of problem for the first time, followed by a meta-heuristic algorithm, named as adaptive large neighborhood decomposition search, to solve the problem. The proposed model and algorithm are also applied to the single-allocation hub location routing problem (SAHLRP) with minor modifications for comparison reasons. Series of numerical experiments have been conducted on the instances generated from Australian Post data set to test the proposed model and algorithm for both the SAHLRP and the MAHLRP. The results prove that our algorithm outperforms the CPLEX on solving these two problems, as well as that applying the MAHLRP can efficiently reduce the operating cost as compared to the SAHLRP.