Seat allocation model for high-speed railway passenger transportation based on flexible train composition

Abstract

Seat inventory control is a crucial technique for railway revenue management. In literature, most of the existing seat allocation models assume a fixed capacity like classical revenue management. To extend the literature, this paper relaxes the assumption of fixed capacity by flexible train composition and discusses the effect of flexible train composition. The probabilistic nonlinear programming model is proposed for high-speed railway passenger service network formed by multi trains with different stop schedule plans. The model is transformed into equivalent linear programming which can be solved by ILGO CPLEX quickly. This paper makes seat inventory control and train composition decisions simultaneously considering stochastic demand and passenger choice behaviour. Numerical experiment results show that the policy under flexible train composition is superior to that under fixed train composition. The sensitive analysis shows that demand intensity, fare classes and elasticity of demand have a significant impact on the policy. The proposed model can provide a decision-making basis for discount sales and ticket allocation under flexible train composition in railway passenger operation.