Dynamic Bus Dispatching Based On Robust Optimization

With the rapid development of economy and continuous growth of population,traffic congestion is getting worse and worse.Public transport is the main solution to solve the traffic problems.Nowadays,most studies of dynamic scheduling are doing simulation and prediction for the passenger data,and then the data is regarded as accurate to establish the optimization model.However,the actual passenger arrival rate is uncertain.Relying solely on the simulation and prediction of passenger arrival rate is still difficult to cope with the problems caused by the uncertainty of passenger arrival rate.How to determine the departure time of buses in dynamic scheduling cycle with the uncertainty of passenger arrival rate is an important practical problem.This thesis is the application of robust optimization in the field of dynamic scheduling and uses the method to determine the departure time of buses in the dynamic scheduling cycle,and the main work includes the following three aspects:First,there are different passenger arrival rates under different scenarios,and a scenario-based dynamic bus dispatching model for single lines based on robust optimization with the maximum value of regret is established.The goal of the model is to minimize the total passenger waiting time.This thesis compares the results of robust optimization and stochastic optimization and discusses the superiority of the scenario-based robust optimization.Second,an interval-based dynamic bus dispatching model is established for single lines based on robust optimization without considering the bus capacity.The passenger arrival rate at each stop is represented by an interval.The actual passenger arrival rate may be any value in the interval,so the interval can more accurately express the uncertainty of the actual passenger arrival rate.Then genetic algorithm combined Cplex are used to solve the model.This thesis discusses the superiority of the interval-based robust optimization to solve different passenger flow trends and makes a comparison of the results with different degree of conservatism.Third,an interval-based dynamic bus dispatching model is established for single lines based on robust optimization which takes bus capacity limitation and actual operating speed into consideration to make it more practical and realistic.Then,this thesis further verifies the advantages of the interval-based robust optimization through study and analysis of different cases.