Research On The Location And Assignment Of Emergency Bus Reserve Bases For Urban Rail Transit

With the expansion of urban rail transit(URT)network and the rapid growth of passenger flow,the operating pressure and safety issues brought by the complex network structure,diversity traffic organization,poor facilities and equipment as well as large passenger flow cannot be underestimated.Making use of the ground bus to evacuate stranded passenger is one of the effective means of URT emergency disposal.Based on the analysis of emergency bus transit organization theory and classical location model,this paper proposes a method about the location and assignment of emergency bus reserve bases(EBRB)comprehensively considering the risk of passengers stranded at URT stations and the delay of emergency bus bridging.In the first stage,back-propagation neural network(BPNN)based on an analysis of passenger stranded risk is proposed.In the second stage,the EBRB location and assignment model of minimum bridging delay and the number of emergency bus reservation is established considering the station risk weight,and the solution algorithm was designed.The main research work includes the following aspects:Firstly,this paper analyzes the relevant strategies of emergency bus bridging combining URT emergency plans of different cities,including the emergency bus response organization and process,the conditions for starting and ending the emergency bus bridging,the way of emergency bus bridging and the choice of the bridging start and end point.Based on the analysis of the classic location model,the characteristics of the location and assignment of EBRB are further summarized.Secondly,this paper considers the risk of passengers stranded at URT stations being included in the location of EBRB.Based on the analysis of the passengers stranded risk,a risk evaluation index system is constructed and quantified rationally.The BPNN model based on the stranded risk is used to estimate the station passenger stranded risk value,which is used as the weight of demand points of an improved P-center location model.The research shows that the risk weight has an influence on the relative importance among stations during location process,which decreases the average distance between high-risk stations and the emergency bus reserve bases(EBRB)by 0.8%-8.4%,but has little impact on the service coverage effect of the whole study area,so it’s beneficial to the reasonable and efficient utilization of emergency resources.Thirdly,based on the analysis of the impact of the stranded risk on the location of EBRB and the analysis of the emergency bus bridging delay,this paper uses the stranded risk as the weight of the URT stations to establish an EBRB location and assignment model that minimizes the average bridging delay and minimizes the maximum number of reserve buses.The model is solved by an improved binary particle swarm optimization algorithm using MATLAB programming.The results of the case study shows that only a small part of the emergency buses during non-rush hours allows the average delay reaching the lowest level.During the peak hours,the average delay is 12.9 minutes when 100%of the emergency buses is used.In addition,after some of the EBRB in the peak hours reducing its bridging services,the location plan that considering the stranded risk ensures that the stations with high risks do not increase delays.The results show that a reasonable EBRB location and emergency bus assignment plan is the precondition and guarantee for the emergency bus to complete the bridging task efficiently,and it is particularly important for the emergency bridging under stressful bus operation during peak hours.