The Optimization Research Of Sea-bound Empty Container Allocation Under Stochastic Conditions

Empty container allocation is important to shipping companies. How to make the allocation decision to minimize the cost is the question this paper aims to answer. The study is from the viewpoint of the shipping company, on the base of the literature review in this field, as well as on the conditions of the current market.The empty container allocation problem has been described systemically, and both of the subjective and objective factors that cause the problem are analyzed. In addition, the characters of the problem have been pointed out. On the basis above, some proposals are suggested. As the quantitive study, a stochastic model has been constructed, on the grounds of Markov Decision Processes (MDP) and dynamic programming theory, paying attention to the stochastic characters of the demand and supply of empty containers. There are three sections in the model. The first is to decide how many containers a port should keep as inventory;the second is to determine the amount and direction the empty containers should flow;the last is to plan how the certain empty containers flow to the destination ports, by integer programming. In the end of the paper, there is the digital simulation.The main contributions of this paper: 1, stochastic factors are paid attention, and the model is based on the combination of MDP and dynamic programming theory. 2,both of the allocation cost and the shipper's satisfaction are considered, and the integer programming in the third section of the model makes the solution of the problem more comprehensive.