The Integrated Scheduling Plan Problem In Coal Blending Port Based On The Theory Of The Network Flow

In the coal supply chain,the port is not only a hub of sea-land intermodal transportation,but also an important place for coal blending.The scheduling plan strongly affect the port throughput and the efficiency of the supply chain.In the coal blending port considered in this thesis,there are more than 30 types of coal stored in the yard and the number of coal blending schemes is as high as 60.The variety of coal types and coal blending schemes makes coal unloading,storage,reclaiming and loading operations of ports extremely complex and difficult to optimize.On the other hand,the decision-making processes of train dumping,stacking and reclaiming,berthing and ship loading are made separately such that the port resources are underutilized.In this situation,how to make full use of the coal inventory and coordinate the key resources in order to serve more vessels in demand,is an important problem facing the port.The integrated scheduling plan problem of port is a large-scale combinatorial optimization problem.It involves many objects with complex relationships.And there's a large amount of information in the problem solution.The focus of current research perspective is mainly on resource scheduling problems that few studies are devoted to address the integrated ship loading and stockyard planning problem to maximize port throughput as a whole.In this study,we consider port operations i.e.coal unloading,coal blending and berth allocation simultaneously.The integrated coal blending,berthing and ship loading problem is proposed.We show the one-to-one mapping between an optimal plan of the proposed problem and a solution of a maximum flow problem with special constraints.The main researches and innovations are as follows:(1)Just as the coal logistics process,the correlation between the objects is introduced,so are the constraints involved in the process of coal unloading,coal blending,reclaiming,and berth allocation.According to the coal logistics process of the port,a train-yard-berth-vessel network of the coal blending port is established.Also,the conversion between integrated scheduling plan problem and the network maximum flow problem with special constraints is introduced.(2)A network maximum flow model with special constraints is built to maximize the port throughput.By considering the priorities of vessels,the ratios of the coal blending schemes,the coal unloading and berth allocation,the model deals with the edges with priorities and ratios,and the unsplittable flows in nodes.(3)The maximum flow problem is reducible to a weighted flow problem considering the priorities of the edges,but exponentially large weights are necessary in general.Therefore,a dynamic programming algorithm(DP)and tabu search algorithm(TS)are designed to solve the problem.Computational tests are conducted to analyze the performance of the algorithms.(4)Taking the actual data of an export coal blending port as an example,the network maximum flow model for the integrated scheduling plan problem is built.The DP algorithm is used to solve the problem and all the optimal solutions would be gotten.By solving the sub-problem of the berth allocation,the integrated scheduling plan satisfied with the largest port throughput and the shortest waiting time is obtained.