Research On Vehicle Routing Optimization Method Under Multi-school Mix-loaded Target

Since the Nineteenth National Congress,the strategy of rural revitalization,as a major national strategy,has been promoted to the important agenda of the work of the Party and governments at all levels.To revitalize the countryside,education tops the list.In recent years,with the rapid urbanization of our country,rural areas in villages and towns have evolved into the implementation soil for the withdrawal of primary and secondary schools and the consolidation of schools.Generally speaking,after the integration,the quality of education in rural schools has been improved.However,due to the change of the distribution of the original schools in the geographical space,the distance between students in some rural areas has increased,leaving the "school at home",which has become the basic problem of rural education.As far as the current level of economic development,the cost of investment in education and the cost of school bus operation are concerned,it is obviously not practical to equip school buses for every primary and secondary school in township areas.Therefore,this thesis will consider the overall planning of school buses' traveling paths in the whole township area as a unit,so as to fully improve Township schools.Vehicle utilization rate,to achieve the goal of cost savings and a certain operational efficiency.School Bus Routing Problem(SBRP)is actually an extension of Vehicle Routing Problem(VRP).Its basic theory is still based on VRP.It is a research branch under the huge research system of VRP and belongs to VRP in essence.This thesis takes SBRP as the research background.Firstly,the mathematical structure of the VRP model with soft and hard time windows(Vehicle Routing Problem with Simultaneous Delivery and Pickup under Coexistence of Soft and Hard Time Windows,VRPSDPCSHTW)is studied.The similarities and differences between its modeling idea and SBRP scenario are analyzed.By improving its mathematical model,a multi-loaded and mixed-load SBRP model with soft and hard time windows to reach the lowest total cost is proposed,and a genetic algorithm is designed to solve the problem.The reliability of the model is preliminarily tested by a small-scale example,and the analysis results of the actual case show the validity of the multi-loaded mixed-load SBRP model constructed in this thesis.Finally,the model checking results are compared with the results based on the Dijkstra algorithm,which further shows the applicability of this thesis to multi-school mix-loaded SBRP problem modeling.The context of this thesis is as follows:Chapter 1 puts forward the background and technical route of this thesis,focusing on the positive significance and research content of building multi-school mix-loaded SBRP in rural areas of villages and towns in China.Chapter 2 summarizes the VRP theoretical basis and the research progress of SBRP.This thesis focuses on the study of the existing school bus routing problems and the differences between urban and rural areas in implementation in China,and discusses the model theory of VRPB.Besides,it also analyses the similarities and differences between VRPB theory and multi-school mix-loaded SBRP.The key issues to be considered in this thesis are put forward as well.Chapter 3 studies and constructs the model of multi-school mixed-load SBRP.Firstly,the problem of multi-school mix-loaded SBRP studied in this thesis is described.According to the description,the corresponding objective function and constraints are constructed.Finally,the preliminary validity of the model is tested by a small example.Chapter 4 introduces the basic principle and operation steps of genetic algorithm.Combining with the multi-school mix-loaded SBRP model constructed in Chapter 3,the corresponding genetic algorithm is designed.The running process of the algorithm is described in detail,and the example model is solved by genetic algorithm.Compared with the result of ATRIP which relies on the Dijkstra algorithm,the genetic algorithm has better solution accuracy and efficiency.Chapter 5 completes the case analysis of Taitou Town,Shouguang City,Shandong Province,based on the MATLAB platform,compares the path data generated by genetic algorithm in the case of three schools mixed-loaded form and other mixed-load forms.The results show that the total cost of the three schools mixed-loaded form solved by genetic algorithm is the least in this case and the path efficiency is the best.The last chapter summarizes the main work and achievements of this thesis,and puts forward the limitations of this study and the feasible direction of future research.