Intelligent Seat Allotment Method for Railway Passenger Train Based on Passenger Flow Forecast

Based on the China Railway Ticketing and Reservation System(TRS)dataset,train ticket data were converted into the time series data of passenger volume.Series analysis method based on improved moving average time was adopted to achieve train daily station-station passenger flow forecast by seat classes.Comprehensively considering seat cleavage factors,ticket protective factors and traffic culture factors,station-station seat adjustment model was established.With the minimum long-distance ticket used by seat cleavage for station-station seat allocation as objective function,seat occupied optimization model was established.The intelligent seat allocation model was composed of the above two models and the pre-allocation scheme was obtained by solving the model.The application results show that seats can be allocated according to passenger needs automatically,fast,reasonably and effectively,and at present the method has been used in China Railway Ticketing and Reservation System.