Research On Finished Vehicle Intermodal Optimization

With the rapid development of the automobile industry, China is the largest country of automobile production and consumption. The larger volume of automobile production and consumption bring the larger the automobile delivery demand. At present, the automobile delivery in China adopts mainly road transportation which is not safe enough, and easily causes transportation congestion, increasing environmental pollution. Furthermore, as the more intensively competing, the capacity of the flatbed is too small to satisfy the larger automobile delivery demands. The limitation of transportation capacity of flatbeds decides that it is difficult for automobile manufacturers to decrease automobile delivery costs by road transportation, which make automobile manufacturers not hold competition advantages under the situation of low sale price of automobile product. Intermodal transportation which is economies of scale, more environmentally-friendly and safe can effectively reduce the transportation cost of automotive products. Therefore, intermodal transportation is expected by automobile manufacturers in automobile industry. In addition, the Chinese government is constructing the integrated transport system, actively building intermodal Infrastructure, solving intermodal transportation bottlenecks and impelling the development of intermodal transportation. Facing with the pressures of intensive competition among automobile manufacturers and the good development chances of intermodal transportation, the automobile manufacturers and automobile logistics enterprises actively carry out intermodal transportation. Under the existed complex automobile intermodal networks, how best to carry out intermodal transportation is the key problem needed to be solved by the automobile manufacturers. Hence, the research of automobile intermodal optimization is conducted, including transportation optimization problems from automobile production bases to intermodal terminals in the first stage of automobile intermodal transportation and those from intermodal terminals to dealers in the second stage of automobile intermodal transportation, which can guide the automobile enterprises’ practices and provide the theoretical and technical supports.The contents of this paper are as follows:(1) Automobile multimodal routing problem is studied. Firstly, the different characteristics between the existing automobile intermodal network and general logistics network are analyzed, which include road, rail and waterway transportation, weekly fixed transportation frequency, multiple batches daily delivery, the demanded delivery time and other structural and operational characteristics. For the above mentioned special requirements from automobile intermodal transportation, the automobile intermodal routing optimization model is developed, which is efficiently solved by optimization solver such as CPLEX. A case study is carried out to assess applicability and performance of the model developed in this study.(2) Under the uncertain demands, the problem of the optimal carrying capacities of rail and shipping services to book in the automobile intermodal networks with complex characteristics is studied. Firstly, to address this practical problem, a two-stage stochastic programming model is formulated, which includes two types of decision variables. The first type of decision variables is the booked carrying capacity of rail and shipping services, and the second type of those is flow in feasible routes. Secondly, based on history data of vehicle demands of automobile enterprises, aiming at combinatorial optimization features of the problem, the sample average approximation method is improved, which is used for the simulation of uncertain automobile demand. Next, by combining the dual decomposition and Lagrangian relaxation method to solve the two-stage stochastic optimization model, the optimal solution that is the booked carrying capacity of rail and shipping services is obtained, which is the basis of solving the optimal flow in each automobile delivery route by the optimization solver in the case of stochastic demand. Finally, a case study to validate the established model and the solving method is put forward, which shows those effectively reduce the adverse effects of uncertain demand on the reasonable capacity allocation of train or ship during automobile intermodal transportation.(3) Based on the existed multimodal automobile transportation network, an integrated optimization problem consisting of deciding reasonable rail and shipping services frequency and the optimal forward and reverse transportation routing is studied. First of all, in order to make full use of existing intermodal network resources, from four aspects including the automobile intermodal network, the customer demands with sensitive delivery time, reverse transportation of old auto parts, transportation costs from forward and reverse transportation to describe the integrated planning problem, it is formulated as an integer linear programming model, whose goal is to maximize the profits of the forward automobile delivery and the reverse transportation of old auto parts. Finally, the model is applied to the instance, which is solved by the optimization software and validated that it can effectively coordinate the use of automobile intermodal transportation resources and reverse logistics decision improves the utilization of transportation resources.(4) The problem of short-distance transportation from automobile intermodal terminals to dealers is studied. Firstly, to solve the conflict caused by the inaccurate data of customer demands or the temporary change of customer demand, the delivery strategy that the delivery routes of vehicles from a vehicle storage center to dealers are dynamically adjusted in real time to meet changing customer delivery demands and then automobiles are allocated between the dealers to flexibly meet customer demands is proposed. Next, according to the strategy, two vehicle schedule models are developed, which include the first model of static vehicle routing problem and the second model of dynamic vehicle routing problem. The clustering-saving algorithm is used for solving the first-stage model and then the initial routes are obtained, furthermore, the improved genetic algorithm is adopted to solve the second model built after the changes of the customer demands and the dynamic adjusted routes are obtained. Finally, through case studies the models and solving methods are validated to be effective for easing the conflict caused by the mismatch between the upstream automobile delivery and the downstream customer demands.