Research On Dynamic Pricing Of High Speed Railway Based On Passenger Ticketing Behavior Analysis

China high speed railway(HSR)have developed vigorously in recent years.By the end of2021,China has more than 40,000 kilometers of HSR.And today,HSR has become the primary mode of public traffic system to meet residents’ cross-city travel needs by virtue of rapidly growing capacity,efficiency,convenience,safety and comfort.However,in the medium and long distance passenger transport,there exists a significant competitive relationship between HSR and aviation.Compared with the dynamic and differentiated price strategy adopted by aviation,the fixed pricing strategy of HSR will affect passenger resource utilization,passenger ticket revenue,and passenger demand response capability,resulting in the unbalanced passenger flow of the HSR parallel trains,idle seat resources,insufficient release of profitability,and lower passenger travel utility.Therefore,it is necessary to study scientific and reasonable dynamic pricing optimization method for HSR in the context of market-oriented reform.Use pricing levers to adjust passenger travel demand and ticket-purchasing decision-making behavior,and HSR could attract more passengers to take HSR,increase ticket revenue.At the same time,traveler could enjoy higher satisfaction when choosing HSR.This paper takes the HSR parallel trains running on a passenger section as the study object,divides the pre-sale period into several pre-sale time frame,aims to optimize the pricing strategies of all high-speed train in every pre-sale time frame,and realize the dynamic pricing process of HSR.The main work are as follows:(1)First of all,we refer to domestic and foreign literatures on passenger revenue management,passenger ticketing decision-making behavior,and dynamic pricing of HSR,and summarize the HSR dynamic pricing theories.After summarizing the deficiencies of existing studies and teasing out related theories,this paper would focus on studying passenger’s ticketing behavior under bounded rationality,and propose a HSR dynamic pricing methodology,which could not only meet the “enhancing revenue” expectations of the railway enterprise,but also conform to the travel characteristics of travelers.(2)For each pre-sale time frame,the average price theory is used to simulate the pricedemand function,and the prospect theory is used to analyze the bounded rationality characteristics of passengers’ ticketing behavior.What’s more,we simulate the arrival sequence of passenger ticketing requests by means of the homogeneous Poisson process.Combining the above three,we can see intuitively the transfer process of passengers’ travel demands between alternative high speed trains,and get the relationship between the comprehensive benefit of the pre-sale time frame and the pricing strategy.(3)With the goal of maximizing the comprehensive gain of the pre-sale period,including the total passenger ticket revenue and the comprehensive prospect utility of passengers during the pre-sale period,we take into account constraints such as train seat restrictions and ticket adjustment intervals,build a HSR dynamic pricing model on upon work(2),and transform the pricing model into a Markov decision process.Then,a robust solution method is designed based on the Deep Deterministic Policy Gradient(DDPG)algorithm of deep reinforcement learning(DRL)to control the randomness of passenger booking requests.(4)Take the passenger section of the HSR from Wuhan Station to Guiyang North Station as the case background,and we carry out the case study using the research data and OD data.Results show that after implementing an optimal dynamic and differentiated pricing strategy for the HSR parallel trains,the comprehensive gain of the pre-sale period increases by 131%,the overall attendance increases by 5 percentage points,the total ticket revenue of the pre-sale period increases by 10549 RMB,and the ticket revenue of the third pre-sale time frame drives the total ticket revenue to increase by 6.8 percentage points.Meanwhile,the comprehensive prospect utility of all travelers increases by 38.71%,and the growth rate of business travelers’ prospect utility is as high as 70.24%.All of these show that the optimized dynamic pricing strategy not only boosts enormously the comprehensive gain,better fits the heterogeneity of passenger travel,and meets the travel needs of different traveler groups,but also increases ticket revenue while reducing idle seat resources,which is in line with all optimization expectations.