Study On The Locomotive Routing Problem In Molten Iron Transportation

Molten iron logistic has a very important influence on the continuous production of steel enterprises,and it is a major problem that steel enterprises need to solve urgently.The main task of locomotive routing problem in molten iron transportation is to determine the route of the locomotive,so as to ensure that the molten iron produced by the blast furnace can be timely transported to the converter in steel plant and carry out subsequent production.The process of molten iron transportation involves a series of complex production processes,and its routing is restricted by the railway track network in the plant area.Therefore,the study of the routing problem of locomotive has great challenges.This thesis studies the locomotive routing problem in molten iron transportation,and has established a mixed integer programming(MIP)model based on problem features.For the practical engineering problem,based on the idea of time-space network,the thesis establishes the time-space network flow model.On this basis,the column generation algorithm is developed to obtain the lower bound of the problem,and then heuristic procedures are proposed to obtain near-optimal solution for larger scale instances.Finally the optimization and dispatching system of molten iron logistics is designed and developed based on the models and approaches proposed above.The main contents of this paper are as follows:(1)A mathematical formulation for locomotive routing problem in molten iron transportation is established.Given the actual situation of molten iron transportation process and the railway line layout in the plant,the thesis establishes a mixed integer programming model(MIP)with consideration of the possible conflicts to minimize the weighted task completion time.For small scale instances,the standard solver CPLEX is used to solve the mixed integer programming model(MIP)to verify its feasibility.(2)A space-time network flow model is established.In order to solve the larger scale problem,based on the idea of time-space network flow.we propose to regard the sections as nodes and regard the locomotive movement as the flow.and establish the time-space network flow model for locomotive routing problem.The standard solution software CPLEX is used to solve the space-time network flow model,and verify its feasibility.(3)A column generation algorithm for the locomotive routing problem in molten iron transportation is designed.In order to solve large-scale problems,column generation algorithm is employed to obtain the lower bound of the problem,and then heuristic approaches based on greedy strategy from different priority angles of road and task are designed to obtain near-optimal solution.The approaches are compared with the results of CLPEX,and the optimal gap and time are used to evaluate the effectiveness of the algorithm.(4)The optimization and dispatching system of molten iron transportation is designed and developed.Taking the transportation process of molten iron in iron and steel production as the research background,based on the above mathematical models and optimization algorithm,we design and develop the optimal scheduling system of molten iron logistics,so as to optimize the scheduling of locomotive in the molten iron transportation.