Research On Inbound Logistics Scheduling Of Bulk Materials In Railway Transfer Station Of Hydropower Projects

Railway transfer station is the material supply hub of large hydropower project.The key that affects the operation efficiency and operating cost of the railway transfer station is the inbound logistics scheduling of bulk materials.To optimize the inbound logistics scheduling of bulk materials in a railway transfer station is of great practical significance,because it is benefit for speeding material movement and decreasing operating cost of the railway transfer station.The goal of inbound logistics scheduling problem for bulk materials in railway transfer station of large hydropower projects is to design a scheduling scheme of inbound logistics including train marshalling and locomotive scheduling,which has the lowest total dwell cost of bulk tankers.Classified by the different usage methods of locomotives,there are two scheduling modes of inbound logistics in the railway transfer station,a general locomotive scheduling mode and a special locomotive scheduling mode.Two mixed integer programming models are formulated for the two scheduling modes,both considering train marshalling and locomotive scheduling.Scheduling algorithms of inbound logistics for bulk materials in the railway transfer station are designed for the two scheduling modes.At first,a novel encoding scheme is proposed for randomly regrouping freight trains;a decoding method is developed to determine train marshalling and locomotive scheduling successively.For the decoding method,the decision of train marshalling contains tanker regrouping and tanker sorting for each regrouped train,and the decision of locomotive scheduling is to design the locomotive scheduling plans according to the operation characteristics of different locomotive usage methods,to allocate the locomotives and arrange job times.Based on the encoding scheme and the decoding method,differential evolution algorithm(DEA)and simulated annealing algorithm(SA)are developed for the two scheduling modes of inbound logistics.With real cases of a railway transfer station in Yalong River Hydropower Project,numerical experiments verify the effectiveness of the designed algorithms.Illustrative cases show the scheduling scheme of inbound logistics,and reveal that one of the main reasons about the big gap of optimal results among different algorithms is owing to the different strategy of tanker regrouping.For the two scheduling modes of inbound logistics,algorithm comparison experiments show that DEA and SA are much better than the three-stage heuristic algorithm(TSA)from another literature on the optimization performance;for small-scale problems,DEA and SA perform similarly;when the problem scale enlarges,the solutions of DEA are better and more stable than those of SA.On the whole,DEA performs better than SA.Comparing the two scheduling modes of inbound logistics,the general locomotive scheduling mode is much better than the special one for decreasing the total dwell cost of bulk tankers.