Study On The Optimization Of Online Bus-hailing Routing In Beijing Capital International Airport

China’s air transport industry is developing rapidly.The huge passenger flow to the airport makes the airport landside traffic face great pressure.According to the spatial and temporal distribution characteristics of travel demands of passengers arriving at the airport,passenger "booking" travel can enable the operating enterprises to grasp the travel demand in advance,rationally allocate transportation capacity resources and balance the supply and demand of transportation.Therefore,this paper takes Beijing capital international airport as an example,and presets stations in each terminal and living area of the airport,analyzes the public transport mode of online bus-hailing for large airports based on passenger demand.Under the operation mode of single-type vehicle and multi-type vehicle,the optimization of online bus-hailing routing is studied to provide new ideas for passengers arriving at the airport in evacuation of large transportation hubs.The main research contents and innovations of this paper are as follows:(1)Analysis of passenger travel demandsThe service mode of online bus-hailing is similar to that of taxis and online car-hailing.OD data of Beijing’s taxi from and to the airport are analyzed based on the characteristics of online bus-hailing,proved that the online bus-hailing is suitable to serve passengers arriving at the airport with uneven distribution of time and space.In order to determine destination of online bus-hailing,DBSCAN and k-means clustering algorithms were used to cluster the destination data of departing vehicles in the airport area,and the results were used as the basis for the destination region division of passengers on the landside of the airport,laying a foundation for the later case study.(2)Study on the optimization of single-type online bus-hailing routingPassengers can submit the reservation information online,and they can take the bus from multiple preset stations.However,passengers who take the bus from the same preset station choose different times and travel destinations.In order to accurately describe the passengers served and not served by the vehicle,and to calculate the time of the vehicle arriving at the preset station,this paper conducts a " splitting the preset station" operation,regarding each reservation travel demand of the preset station as a reservation station,and then builds a mathematical model based on this.In the single-type vehicle operation mode,the enterprises that fail to fully meet the reservation demand of passengers will be punished to some extent.Therefore,a mathematical model for the optimization of online bus-hailing routing is established to maximize the economic benefits of the enterprises.In order to guarantee the travel quality of passengers,the maximum travel time of each passenger is limited in the constraint conditions.In order to avoid the waste of transportation resources,the minimum full load rate of vehicles is set.An experimental case was designed in Beijing,and the numerical results were analyzed by means of grouping genetic algorithm.The results showed that the longest travel time,the full load rate of vehicles,the ticket price and the unit penalty cost of passengers’ failed reservation all had an impact on the results.(3)Study on the optimization of multi-type online bus-hailing routingThe multi-type vehicle operation mode makes up for the deficiency that the single-type operation mode may not fully meet the travel demands of passengers.In order to maximize the economic benefits of the enterprise and achieve a more balanced service level,an optimization model of online bus-hailing routing is established.The genetic algorithm was used to solve the study cases and the numerical results were analyzed.The genetic algorithm was used to solve the study cases and the numerical results were analyzed.The results showed that setting the appropriate service level equilibrium reward and punishment function could effectively reduce the difference in full load rate between vehicles,distribute the limited transport capacity resources to each passenger more fairly,and improve the service level equilibrium of the online bus-hailing.