Pricing Optimization for High-Speed Railway Considering Hybrid Choice Behavior of Heterogeneous Passengers Under Stochastic Demand

Passenger heterogeneity in loyalty fundamentally influences their choice behaviors, and is pivotal to railway differentiated pricing. Thus, travelers are categorized into loyal passengers and non-loyal passengers. According to the generalized cost minimization, we identify a train priority sequence reflecting consistent preferences of loyal passengers and establish a train selection probability model based on stochastic preferences of non-loyal passengers. Then, a hybrid choice model resulting from the distinct decision-making processes of these two passenger categories is formulated. A nonlinear pricing optimization model in the scenario of multiple train categories with multiple trains is established. An improved Particle Swarm Optimization algorithm based on the Sampling Fitness Strategy (SFS-PSO) is proposed to improve the solution accuracy. The SFS-PSO enhances the search diversity for the personal historical best positions and global best position without expanding the size of the particle swarm as much as possible. The case analysis demonstrates that the proposed pricing optimization approach can increase the expected revenue by 1.7%, validating the rationality of considering the hybrid choice behavior of passenger loyalty heterogeneity for the railway pricing optimization problem. Meanwhile, the case results highlighted the significant impact of the proportion of loyal passengers on revenue improvement.