A Hybrid Genetic Algorithm for Logistics Network Design with Flexible Multistage Model

Logistics network plays a key role in building an efficient and flexible logistics system for companies in the global business environment. A lot of research has been conducted in this field. While the researchers treat logistics networks design problem as a traditional multistage logistics network model, in which arcs should connect the two adjoining echelons in the network and there are no arcs striding over any abutting echelons, thereby the problem can be solved stage by stage. However, in practice this kind of traditional multistage logistics network (tMLN) model sometime causes problems, such as too-long delivery path, slow response etc. In this paper, we address flexible multistage logistics network (fMLN) design problem with nonadjacent structure, i.e. in this problem some non-neighboring echelons are connected with arcs (nonadjacent connecting arcs). In some practical cases, the nonadjacent connecting arcs make the logistics networks cost-effective and adaptable to changes in situation. On the other hand, the existence of them makes the problem much more difficult by traditional optimization methods. We formulate this problem as location-allocation model, and propose an effective hybrid genetic algorithm to solve this problem. Moreover, numerical analysis of case study is carried out to show the effectiveness of the proposed approach.