Research On High-speed Rail Dynamic Pricing Model Based On Bounded Rational Logit Model

With the rapid development of China's high-speed rail,the improvement of transportation service quality and transportation timeliness does not match the existing ticket prices,and the ticket pricing mechanism is not adapted to the basic needs of the socialist market economy and transportation market competition.Starting in 2018,the China Railway Corporation issued a notice to test the “one price per day” pricing strategy on some highspeed railway lines.Therefore,it is extremely important to create a high-speed railway passenger fare model that is more in line with the competition in the modern transportation market.In this context,in order to establish a dynamic pricing model that is more suitable for the existing high-speed rail network,this paper mainly takes the medium and long-distance transportation market as the research object,and use the ticket stub data of June 2018 and the data obtained from the survey as the data set,and takes the low season of transportation as the research period,and dynamically price the existing high-speed rail trains to enhance the competitiveness of high-speed rail in this market,So start the research from the following points:(1)First,referring to the research results of domestic and foreign scholars in the fields of dynamic pricing strategies,passenger travel behavior analysis,and bi-level planning models,in order to address the shortcomings of existing research,this article focuses on the travel behavior of passengers under limited rationality and proposes highspeed rail Dynamic pricing strategies for passenger travel behavior laws.(2)Secondly,analyze the basic models and theories of common travel behaviors,and summarize the deficiencies of common theories and the factors that affect the travel of high-speed rail passengers.By referring to domestic and foreign references,and combining with the actual situation of China's current high-speed rail operation,this part chose to select age,monthly passenger income,travel purpose,travel costs,travel time,and comfort as parameters to study passenger travel behavior.In addition,the Logit model is innovatively combined with the third-generation prospect theory,and a bounded rational Logit model is constructed to study the travel behavior of high-speed rail passengers.In addition,using the actual travel data of the Beijing-Shanghai line as the data set,the MNL model and the NL model are selected as the comparison model.The model accuracy results show that the model's accuracy results show that the bounded rational Logit model can more accurately analyze the travel behavior of passengers and calculate the passengers' choice of various high-speed railways.The probability of train types sorts out the travel behavior mechanism of Beijing-Shanghai high-speed rail passengers,and at the same time lays a solid foundation for the later construction of a high-speed rail dynamic pricing model based on the bounded rational Logit model.(3)Then analyze the shortcomings of the existing high-speed railway pricing strategy,based on the existing pricing strategy literature,with the single train and multitrain dynamic pricing problem as the research goal,systematically elaborate the dynamic pricing model used in this article.This part mainly refers to the revenue management theory and the bi-level programming model,and expounds the theoretical basis of the two and their respective advantages and disadvantages.On this basis,the stochastic dynamic programming theory is selected to establish a dynamic pricing model suitable for the medium and long-distance transportation market,and the model can be related to the passenger travel behavior of the second part,which provides a theoretical basis for later research.(4)Based on the theories and models of Part 2 and Part 3,the travel behavior model and dynamic pricing theory are combined to build a dynamic ticket pricing model based on travel behavior.The model uses the bounded rational Logit model as the travel behavior analysis.Model,taking the passenger's selection probability for each travel mode after the ticket price changes as the market share,and applying it to a dynamic pricing model based on stochastic dynamic programming theory,and applying the forward update strategy to obtain the model Optimal solution.(5)Finally,the actual data of the Beijing-Shanghai high-speed rail line in 2018 is selected as an example.Based on the travel behavior data questionnaire survey,this paper based on the passenger RP/SP fusion data to statistically obtain the passenger's psychological reference value for the three factors affecting comfort,time and price,and select the actual passenger flow.The data is substituted into the model for calculation,and the results are compared with the traditional bi-level programming model.The results show that the dynamic pricing model optimization based on travel behavior in this paper not only obtains the travel preferences of passengers,improves service quality,but also improves train seating.Rate and the total revenue of the railway sector,the final result is also superior to the pricing strategy using a two-level programming model.