Optimization On Dwell Time For Metro Train

The urban railway timetable has great influence on the service quality of daily operation and dwell time setting is the basis of timetable establishment.In the daily operation,the dwell time is mainly affected by boarding and alighting scale.However,the passenger flow volume may change at every second.Too short dwell time setting may lead to the departure delay while too long dwell may extend passengers'travelling time.For this purpose,based on the analysis of the relationship between dwell time and passenger boarding and alighting scale,research on dwell time optimization is of great importance for improving quality of timetable.During the dwell time setting,the standard dwell time can be set according to the average boarding and alighting volume.Based on that,the redundant dwell time can be set considering the fluctuation range of passenger flow volume.Redundant dwell time not only can eliminate the departure delay caused by dwell time extension,but also absorb the slight train section operating extension.Therefore,based on analysis of the correspondence between the actual dwell time and the passenger population,this paper establishes an optimized model on redundant dwell time allocating to minimize the passengers' total travel time.Case studies are conducted to verify the effectiveness of the proposed model.Main contents of this study are as follows:1.This paper studies the relationship between the disturbance and allocation of redundant time.The chapter first analyzes the concept,the classification and the cause of the disturbance.On the basis of the analysis on the concept,the classification and the mechanism of redundant time,the process which redundant time is used to decrease the negative effect caused by the disturbance is described by mathematic formula.2.This paper analyzes the great influence on the actual dwell time by the randomness of passenger flow arrival.Based on preprocessing the historical data of passenger flow and calculating OD matrix,dwell time model is proposed by analyzing the process of the passenger getting on and off the train.Finally,according to dwell time calculated by the historical data of passenger flow,the distribution function fitting and hypothesis testing are also proposed.The expected value is used as the reference for standard dwell time at each stop.3.Considering the constraints which conclude maximum and minimum dwell time,minimum safety interval and upper limit of total redundant time,this paper establishes an optimization model for redundant dwell time allocating,aiming at minimize the total passenger travelling time.Genetic algorithm is designed to solve the proposed model.Case studies indicate that,compared with original timetable,redundant dwell time allocation which fit the passenger flow laws decrease the total travelling time by 9.37%and decrease the total waiting time by 7.77%.At present,with the fixed upper limit of redundant dwell time,the common allocating methods conclude:proportional distribution which based on the proportion of the standard dwell time to total dwell time,average distribution by the number of stations and concentrate distribution guided by specific indicators.Comparing with existing methods,the optimization range of total travelling time of this paper increase 3.08%,5.26%and 6.68%.