Models And Algorithms For Rail-road Intermodal Transportation Of Hazardous Materials

Rail-road intermodal transportation is composed of three key operations,namely rail haul operation,drayage operation and transfer operation,which combines the advantage of economies of scale of railway transporation and the flexibility of road transporation,is a common mode for long-distance land transporation of hazardous materials.Hazardous materials rail-road intermodal transportation optimization problems mainly include the rail-road transit station location problem,the service network design problem and the route optimization problem.The first two problems are from the standpoint of the rail-road transportation operator which is primarily oriented towards the rail transportation management sector,while the last one is from the rail-road transportation shipper's point of view.This thesis uses modern mathematical optimization theory and method to solve these problems,and the main research works are as follows:Chapter 2 investigates the risk assessment problem of rail-road intermodal transportation of hazardous materials.The key point of this problem is to establish a risk assessment model suitable for hazardous materials rail-road intermodal transportation optimization problems with the comprehensive consideration of the risk characteristics of the hazardous materials rail-road intermodal transportation processes.After the systematically analysis of the rail-road intermodal transportation processes,we discuss the characteristics of the processes and the risk assessment models.And then we analyze the accident probability and consequence of each process.Based on the expected risk,we first build a risk assessment model for calculating the risk of given path.Taking into account of the limitation of the above model,we then respectively bulid a uni-commodity flow routing optimization model and a multi-commodity flow routing optimization model for hazardous materials rail-road intermodal transportation based on expected risk,in order to provide the modeling ideas for the following optimization problems.Finally,we give an algorithm to solve the model.A numerical example is used to test the effectiveness and efficiency of the proposed approaches.Computional results show that rail-road transportation is a competitive mode for hazardous materials transportation to reduce the total risk.Chapter 3 discusses the location problem of the transit station of rail-road intermodal transportation for hazardous materials.The problem is to locate rail-road transit station from various alternative stations and allocate of all kinds of hazardous shippments to these selected stations,so that all shipments' demand can be met,and the balance of total cost and total risk could be achieved.Firstly,a p-hub location model is built.This model is easy to be understood,but is hard to be solved due to its huge size caused by the weakness of the modeling stratege.An arc-based multi-commodity flow network design model is then built to improve the above model.Aiming at minimizing the total cost and total risk,the new model contains both the lead time and the capacity constraints.The structure of the multi-commodity flow network design model is compact,and it can greatly reduce the number of variables and constraints compared to the p-hub location model.Based on that,three multi-objective algorithms,including a weighted-sum algorithm,an augmented weighted Tchebycheff algorithm and an augmented ?-constraint algorithm are customized to approximate the non-dominated frontier,respectively.At last,a numerical example coming from the realistic situation is used to test the proposed approaches.Computational results show that the augmented s-constraint algorithm is the more suitable multi-objective optimization approach to handle the hazardous materials rail-road intermodal transportation location-allocation problem in terms of the number and uniqueness of distinct non-dominated solutions.And the total cost and risk could simultaneously be reduced to a certain extent if the capacity of the transit station is expand.Chapter 4 discusses the rail-road intennodal transportation service network desigh problem.The proble lies on determining the frequency of the trains and the allocation of both hazardous and common shipments to these trains as well as the rail-road connecting plan of all shipments on the basis of the established train formation plan with the consideration of the actual situation of China.Firstly,by skillfully using unified method,the rail-road intermodal transportation service network is built.We define the the car service arcs and different types of train service arcs between service nodes as different transporation services arcs,and the various transshipment or transfer operation arcs in the same node as transfer service arcs.Secondly,aiming at minimizing the total cost and total risk,a multi-objective mixed integer linear programming model takes into account of both hazardous and common materials is formulated with the constraints of the compatibility of goods and services,the capacity of the transit station,and lead time of each shipment.The above service-network-based model is simple and compact in structure,which enables the solving of real-world problem with large size.By exploiting the characteristics of the model,three multi-objective algorithms are customized.At last,a numerical example coming from the realistic situation is used to test the proposed approaches.Computational results show that the augmented ?-constraint algorithm is the more suitable multi-objective optimization approach to handle the hazardous materials rail-road intermodal transportation service network design problem in terms of the number and uniqueness of distinct non-dominated solutions.In addition,the total cost and risk could simultaneously be reduced if the lead time constrains is relaxed or if the transfer time is shorten.Chapter 5 discusses the rail-road intermodal transportation route optimization problem of hazardous materials.The key point of the problem is to determine the route for all shipments with the consideration of the economies of scale and the fixed schedule-based space-time as well as the service time windows constrains of railway transportation,so that all shipments' delivery time demand can be met,and the balance of total cost and total risk could be achieved.We first describe the cost of railway transportation as a piecewise linear function.And the rail-road intermodal transportation service network takes into account of time is then constructed on the basis of the analysis of the fixed schedule-based space-time as well as the service time windows constrains of railway transportation.Then a well-understood multi-objective nonlinear model aims at minimizing the total cost and total risk is then built,which includes all operation and capacity constraints,lead time constraints and so on.Then the nonlinear model is linearized,and 3 general multi-objective optimization algorithms are also customized to solve the linear model.Finally,computational results of a large-scale realistic example show that the augmented s-constraint algorithm is the more suitable multi-objective optimization approach to handle the hazardous materials rail-road intermodal transportation route optimization problem in terms of the number and uniqueness of distinct non-dominated solutions,and the effectively avoidance of the redundant calculation,and its good performance when focuses more on total cost.What's more,the sensitive analysis of some parameters shows that the total cost and risk could simultaneously be reduced if the lead time constrains is relaxed or if the transfer time between rail and road is shorten.