Research On Three Dimensional Container Loading Problem Of Automobile Parts Logistics Center

With the rapid development of the domestic market economy, the auto industry began to enter the stage of explosive growth. As an important logistics center for automotive manufacturers and service providers, Auto parts logistics center’s distribution scale and frequency is growing, and it’s corresponding logistics costs continue to rise. As an extremely important aspect of logistics, auto parts logistics center’s loading problems are getting more and more attention. Meanwhile, with the increasing number of corporate business, logistics center needs take simultaneous distribution to multiple destinations into consideration, generating reasonable loading scheme can bring huge benefits for the enterprise. So taking advantage of scientific methods of loading optimal allocation, improving transport vehicle loading rate and meeting the needs of different situations is of important significance to reduce operating costs, improve distribution efficiency and promote customer satisfaction.This thesis studies the three dimensional container loading problem of automobile parts logistics center. First, we introduces the function and the operating procedures pf automobile parts logistics center, and makes a thorough analysis of loading assignment processes, respectively, makes a thorough analysis of all stages problems, including box packing problem, pallet loading problem and vehicle loading problem, and conducted the detailed information of the objective functions and various constraints in all loading stages. Then, we establish the model of three dimensional container loading problem considered multi-destination delivery, at the same time, the model established can satisfy the characteristic of adjustable loading structure. A hybrid quantum genetic algorithm is designed to solve this model. The objective of this model is to acquire the maximum loading space utilization, and meet the demands of realistic loading problems through realizing the characteristic of adjustable loading structure by simply parameters initializing. The hybrid optimization algorithm designed is based on quantum genetic algorithm and using the space depict strategy policy based on surfaces as its decoding strategy. For this reason, the algorithm that introduced in this paper proposed a new quantum coding pattern, fixed strategy and quantum update mechanism. Compared to the traditional evolutionary algorithm, quantum genetic algorithm has more powerful capacity in traversal search. Finally, an actual numerical experiment and a comparison test were illustrated to verify the efficient and stable performance of the proposed algorithm. The results have shown that the solution has a satisfactory, stable, compact effect and an adjustable loading structure.