Research On Modelling And Optimisation Of Sea-rail Intermodal Transportation System Under Stochastic Circumstance

With the great strategic concept of "One Belt and One Road" and the deepening of economic globalisation,sea-rail intermodal transportation has become an important boost for the Maritime Silk Road with its advantages of high efficiency,safety,convenience and reliability.Therefore,it is particularly important to study the development planning,transportation organisation and operational management issues in the sea-rail intermodal transportation system.Sea-rail intermodal transportation is a multimodal transportation system integrating multiple transportation modes,which can not only weaken the disadvantages and limitations of unimodal transportation in transportation timeliness,transportation distance and transportation cost,but also compelement the advantages of multiple transportation modes,reduce transportation cost and improve transportation efficiency under deterministic conditions.However,under stochastic conditions with a large amount of uncertain disturbance,the reliability of transportation system is chanllenged.Thus,it is difficult to ensure the ideal performance of multimodal transportation by using traditional deterministic transportation organisation methods.Therefore,in the context of competitive freight market,how to construct stochastic programming models and design corresponding algorithms for multimodal transportation system based on stochastic factors,which can yield a reliable and efficient transportation organisation and operational management plan,is an important frontier research topic.This dissertation conducts a research on modelling and optimisation of sea-rail intermodal transportation system under stochastic circumstance.This dissertation takes the loaded container flow formed by export goods and the empty container flow in the hinterland network of coastal ports as the research object,and conducts an in-depth research on the models and algorithms of container routing problem in the sea-rail intermodal transportation system,empty container repositioning problem in the sea-rail intermodal transportation system and service network design problem in the sea-rail intermodal transportation system under stochastic circumstance,providing the theory and policy support for decision makers to develop a reasonable,efficient and reliable sea-rail transportation plan under stochastic circumstance.The main research contents are as follows:(1)Analysis of the characteristics and stochastic factors of sea-rail intermodal transportation system.Firstly,on the basis of existing research,this dissertation summarises the definition,system composition and organisation process of sea-rail intermodal transportation system,and elaborates the development status of sea-rail intermodal transportation based on the domestic and abroad ports.Secondly,this dissertation conducts an in-depth analysis for the organisation form of railway container transportation,the transfer mode of container ports and the organisation form of maritime container transportation.Finally,the uncertainty of the sea-rail intermodal transportation system is analysed from the aspects of travel time by train,transfer time at ports,travel time by ship,as well as container demand and supply,and the various stochastic factors are further studied from the perspective of reasons and forms.(2)Research on stochastic container routing problem in the sea-rail intermodal transportation system.Based on the analysis of time-sensitivity,reliability and component costs for container routing problem in the sea-rail intermodal transportation system,this dissertation takes loaded container flow formed by export goods as the research object,taking into account stochastic travel time by train,stochastic travel time by ship,and stochastic transfer time.A binary integer chance constrained programming model for stochastic container routing problem is formulated.On-time delivery chance constraints are proposed to limit the arrival time for each container demand,on the basis of stochastic time parameters.With respect to solution methodology,this dissertation designs a hybrid heuristic algorithm to solve the binary integer chance-constrained programming model,incorporating stochastic simulation,multilayer feedforward neural network and genetic algorithm.A realistic sea-rail intermodal network from the hinterland of eastern coast in China to Singapore is conducted to demonstrate the feasibility of the proposed model and analyse the impact of stochastic time parameters and confidence level of chance constraints on late delivery cost,total cost and punctuality,from the perspective of economy and service level.(3)Research on stochastic empty container repositioning problem with multistage decision in the sea-rail intermodal transportation system.Based on the analysis of reasons,characteristics and component costs for empty container repositioning problem in the sea-rail intermodal transportation system,this dissertation takes the empty container flow in the hinterland network of coastal ports as the research object,and constructs a chance-constrained nonlinear integer programming model for stochastic empty container repositioning problem.This model considers the number of repositioned empty containers,repositioning routing and the number of leased empty containers simultaneously,minimising the expected value of total weighted cost,which is the weighted sum of repositioning cost and CO2 emission-related cost.Empty container flow chance constraints are proposed regarding stochastic demand and supply of empty containers.With respect to solution methodology,this dissertation designs a hybrid heuristic algorithm based on dynamic programming method to solve the mathematical model,by integrating sample average approximation method and two-phase tabu search algorithm.A realistic sea-rail intermodal hinterland network of eastern coastal ports in China is studied to demonstrate the feasibility of the proposed model and explore the effect of weight combinations,unit leasing cost,stochastic demand of empty containers and stochastic supply of empty containers on repositioning plan and relevant costs.(4)Research on stochastic service network design problem in the sea-rail intermodal transportation system.According to the characteristics of service network design in the sea-rail intermodal transportation system,this dissertation takes the loaded container flow formed by export goods and the empty container flow in the hinterland network of coastal ports as the research object,considering stochastic travel time,stochastic transfer time and stochastic container demand.A two-stage chance constrained programming model for service network design problem is developed.Capacity chance constraints and on-time delivery chance constraints are proposed to optimise the service selection,service route and service frequency under stochastic conditions.With respect to solution methodology,this dissertation designs a hybrid heuristic algorithm incorporating sample average approximation and ant colony optimisation.In addition,the influence of stochastic travel time,stochastic transfer time,stochastic container demand,confidence levels and sample size on service network design plan and relevant costs.