A decision-making model of railway tourist dedicated train operation plan

Abstract

The rapid increase of China’s railway mileage and the effective release of transport capacity provide a basis for the development of tourist-dedicated train. Considering the decision-making between the railway enterprises and the local government on whether the tourist-dedicated train stops at the local station, the length of stay and the market price, the model of its operation plan is designed to solve algorithm problem and provide foundation for railway enterprises to improve the efficiency of operation. Based on the results of experiments, we found that the price sensitivity of consumers is negatively correlated with the pricing and income of tourist-dedicated train. The higher the sensitivity of prices is, the lower the fares and profits of the tourist railways service will realize. In addition, the initial scale of tourism demand of each region along the railway has a direct impact on whether to build a stop in the area. A larger demand scale will increase the possibility of the tourist-dedicated train stopping. In this case, to develop the Mid-and High Tier tourist-dedicated train, the close contact and strategic cooperation with local governments before its opening should be traded-off and considered as important factors for the China National Rail.

Keywords

Tourist dedicated train decision-making model game theory

1 Introduction

Tourist-dedicated train is a kind of tourism product that railway enterprises or National Railways rely on the resources of railway transportation to meet the needs of social tourism. In recent years, with the rapid growth of railway operation mileage, especially the coverage and extension from “four vertical and four horizontal” to “eight vertical and eight horizontal” system of national high-speed railway, the development of tourist-dedicated train is embracing the more and more new opportunities. In terms of function, the special tourism train meets the travel needs of consumers but also promotes the economic development of the scenic spots along the route, meanwhile increasing the size of passenger inflow, brand promotion, consumer demand, and investment opportunities. Therefore, taking the needs of consumers, local governments and railway enterprises into consideration, it has great essence to study how to optimize the operation of tourist-dedicated train.

2 Related works and background

Transportation is an important foundation to the develop tourism. The improvement of transportation technology, such as the expansion and transformation of existing railway lines and the increase of high-speed railways, can effectively reduce the time required for tourism and stimulate tourism demand. Masson and Petiot [13] conducted a furthered study on how European high-speed rail affects tourism growth and economic development based on the new economic geography model, pointing out that high-speed rail will promote the development of urban tourism and commercial tourism to some extent and strengthen the concentration of tourism supply in big cities, making core destinations more attractive and increasing the number of tourists, which is more important for European metropolises, such as Barcelona. But it will also lead to a reduction in the number of surrounding tourist destinations. Pagliara et al. [5] studied the impact of Italian high-speed rail projects on local tourism by using 77 Italian municipal databases between 2006 and 2013. The research showed that high-speed rail has a positive impact on local tourism, and that a denser high-speed rail network increases the potential to attract not only local tourists, but also the global tourists. Guirao and Campa [2] used the proven multi-criteria corridor selection methodology (to assess the impact of tourism on high-speed rail demand) and multiple regression model (to assess the impact of high-speed rail on tourism demand) to study the cross-impact between high-speed rail and tourism in Spain. It was found that tourism destinations play a significant role in promoting the demand for high-speed rail, but the impact of high-speed rail on tourism demand is still worth discussing.

China has the world’s longest high-speed rail network that exceeds the sum of the rest countries’ high-speed rail mileage in the world. The study by Wang et al. [15] shows that the construction of China’s high-speed rail network expands the space of the tourism market and greatly increases the choice of tourism destinations, meanwhile it weakens that of some cities or regions that are not tourism centers. Wang et al. [4] found that high-speed railway strengthens the tourism-based economic relationship between cities, which makes the spatial distribution of external tourism economic relations show a “corridor” effect through the economic relationship model analysis and spatial analysis of 338 national administrative units. Jin et al. [12] compared the tourism situation of ice and snow resorts in northeastern part of China, Heilongjiang, Jilin and Liaoning Province, before and after the construction of high-speed rail in 2008 and 2018 respectively; and stated that the shortening of travel time between cities by high-speed rail changed the spatial distribution of ice and snow resorts and increased the number of tourists to that places. Su and Wall [11] assessed the impact of the Qinghai-Tibet Railway on tourists’ travel decisions and travelling experience from the perspective of tourist, and concluded that the opening of the Qinghai-Tibet Railway provided the transportation convenience and traveling affordability to Tibet. Zhang et al. [1] found that the links between high-speed rail increase the value of listed companies in tourism sector, result from greater number of tourists. Campa et al. [8] used a model of the relationship between high-speed rail network and tourism based on the Chinese background. Through the study of Spanish data from 1999 to 2015, it is found that some tourism revenue growth is positively associated with the construction of high-speed rail network, but the impact is lower than that in China.

People also pay attention to how to improve the capacity of tourism demand forecasting and the enthusiasm of tourism service-related subjects. For example, in the aspect of tourism demand forecasting, Uwimana [3] uses Hidden Markov Model (HMM) to explore the role of Google trend information in tourism demand forecasting. Dilogini et al. [9] evaluated and revealed the relationship between consumer characteristics and information technology-related marketing activities. As for the cooperation among subjects related to tourism services, Fang et al. [10] considered the influence of tour guides’ efforts on tourists’ tourism utility, and found that performance-based sharing mechanism and reward and punishment mechanism can effectively improve tourists’ utility and satisfaction. Based on the altruistic preference of decision makers, Wan et al. [14] discussed the optimal pricing strategy and cooperation strategy between low-carbon tourism products, service provider (TCP) and travel agency (OTA).

The research on the plan of opening tourist-dedicated train has also attracted more attention. For example, Yang et al. [16] analyzed the influencing factors of tourism train operation plan from the aspects of tourism demand behavior, pick-up and departure capacity of station and reception capacity of tourist attractions, railway transport capacity and so on, and put forward the method of tourism train diagram. Jia [7] constructed the economic benefit evaluation index system of railway tourism special train operation plan from four aspects: cost profit margin, break-even point, market share and train on-line rate. Zeng [17] takes Guiyang-Guangzhou high-speed railway as an example, under the assumption that the tourism demand is known, a multi-objective tourism train operation plan model is constructed to maximize the income of railway enterprises, minimize the waste of train capacity and minimize the unmet demand.

Different from western countries, China’s railway is mainly built and operated by China National Railway Group Co., Ltd., and railway enterprises in China are also accelerating the transformation to market-oriented operation. In this case, from the view of railway enterprises, how to optimize the decision-making of their own tourism special train development plan according to the market principle under the premise of fully considering the market demand, we need to further expand the relevant research on the basis of the paper mentioned above, especially on the basis of the relevant research on the tourist-dedicated train opening plan.

3 Motivation

Tourism special trains usually provide pre-planned travel routes and related services for fixed originating passengers. As a tourism product or service, railway enterprises need to formulate the operation plan of tourism special train based on the market demand, including market price, operation cycle, operation quantity, train type, stops location and the length of stay and so on. When it comes to market demand, the lower the ticket price of tourism special train is, the greater the demand is, which is more beneficial to consumers and the geographical area where the scenic spots are located, yet the income will be reduced if the price is set at too low level. Under the circumstance of the tight railway capacity in the past, the operation of special tourism trains is mainly and exclusively decided by railway enterprises based on their own costs and market demand. However, today, with the rapid development of high-speed railway, the existing lines are fully released and the transport capacity is becoming increasingly sufficient, to achieve win-win results by strengthening cooperation with local governments along the line. For example, the government of the place where the planned stop is located gives fixed subsidies to railway enterprises to reduce the operating costs of tourism special trains, and thus reducing the market price of tourism special trains, increasing the scale and frequency of transport volume, and realizing the common benefits of railway enterprises, local governments and consumers. From the perspectives of railway enterprises and considering the game of cooperation with local governments, this paper studies and formulates the decision-making model of the development plan.

4 Model assumption and problem definition

Suppose that there are stations and road sections in the railway network of a given fixed operation section, shown as G (S, E) with the existence of n stops and m routine, and each station belongs to a different local government, as shown in Fig. 1.

Fig. 1

The decision-making process of the tourism special train opening plan.

Assume: $\begin{matrix} S = & {s_{i} | i = 1, 2, \dots, n} E = {e_{j} | j = 1, 2, \dots, m} \\ L = {l (e_{j}) | j = 1, 2, \dots, m} \\ G = {g_{i} | i = 1, 2, \dots, n} \end{matrix}$ represents the collection of stations, the collection of road sections, the length of road sections, and the collection of local governments respectively.

The train collection of the operation plan is defined as T ={ T_k|k = 1, 2, ⋯ , q } whose the corresponding number of the special train is N (T_k), and the fixed number of personnel in the train is A (T_k). The tourist special train may or may not stay at the station, setting as x_i which equals to 1 when train stops, otherwise equals zero; the stay time at the station is recorded as t_i, considering that the stay time of half a day is generally defined as an integer. Considering that people have certain restrictions on the running period of the tourist dedicated train, it is assumed that the total travel time $\hat{T} = ω τ$ that the tourist dedicated train can stay along the way (excluding the running time of the train on the line), ω is an integer greater than 0, for this reason, the stay time of the train at all stations needs to satisfy $\sum_{i = 1}^{n} x_{i} t_{i} ⩽ \hat{T} = ω τ$ . The annual market demand of the tourist dedicated train is $d (x_{i}, \dots, x_{n}, p) = \sum_{i = 1}^{n} x_{i} (a_{i} - bp)$ , and a_i > 0, the number of the initial demand of it that can be attracted by the location of the station s_i (i = 1, 2, ⋯ , n), b > 0 is the sensitivity of the demand to the price, and p is the price of the tourism special train.

In order to attract special tourist trains to stay locally, local governments g_i (i = 1, 2, ⋯ , n) pay fixed subsidies of φ_i (i = 1, 2, ⋯ , n), φ_i ⩾ 0, to railway enterprises. The benefits that the tourist special train bring to the local government (v_i (d (x_i, ⋯ , x_n, p) , t_i) = μ_it_id (x_i, ⋯ , x_n, p)), including μ_i ⩾ 0, the comprehensive benefits that each consumer can bring to the local area for the time spent in each unit.

The other symbols used in this article are as follows:

θ for the seat utilization rate of the special train, and θ_u and θ_d for the upper and lower limits of the seat utilization rate of the special train respectively.

c_stop (T_k, s_i) for the stop fee of the train at the station s_i, it is assumed that the stop fee of all stations is the same, the figure of c_stop (T_k).

f_i (t_i) = αt_i and α > 0, the per capita tourism expenses incurred by the stop s_i (i = 1, 2, ⋯ , n) of the train, where $\sum_{i = 1}^{n} f_{i} (t_{i}) < p$ .

f_tran (T_k) is the transfer fees for the train stops, assuming that the transfer fees are the same at all stations.

c_fixed (T_k), the fixed cost for special tourism train.

u (T_k), the grade of the special tourism train.

r (e_j) = 1, 2 respectively indicate that the line type of the road section e_j is an existing railway or a passenger dedicated line.

F (u (T_k) , r (e_j)), the cost of train kilometers (depreciation, maintenance, energy consumption, etc.).

δ (T_k, e_j) indicates whether the special train T_k passes through the road section, e_j, when passing through defined it as δ (T_k, e_j) = 1, otherwise δ (T_k, e_j) = 0.

φ (e_j), the passing capacity surplus of section e_j.

φ (s_i), the receiving and dispatching capacity surplus of the station s_i.

Other assumptions in this article are as follows:

Information symmetry between railway enterprises and local governments.

the annual demand is stable, that is, the influence of seasonal factors on tourism demand is not taken into account.

the tourist trains were assumed runs or operates the same type of class.

the pick-up and departure capacity of all stations and the carrying capacity of the section are adequate.

Railway enterprises need to decide the market price p, stop strategy x_i and stop time of tourism special train t_i, or parameter y_i, i = 1, 2, ⋯ , n; and the local government needs the subsidy support given by φ_i, i = 1, 2, ⋯ , n. The decision-making order is that the railway enterprises first decide the market price of the tourism special train, then the local government makes the subsidy support, and finally the railway enterprise decides the parking plan of the tourism special train.

5 Method and model

5.1 Local government revenue function

The income of the local government is the comprehensive income x_iv_i (d (x_i, ⋯ , x_n, p) , t_i) brought by the special tourism train, and the cost is the fixed expenditure of subsidies, x_iφ_i. The function is known as $\begin{matrix} π_{i}^{1} (φ_{i} | x_{i}, t_{i}, p) = x_{i} [μ_{i} t_{i} d (x_{i}, \dots, x_{n}, p) - φ_{i}] \\ = x_{i} (μ_{i} t_{i} \sum_{z = 1}^{n} x_{z} (a_{z} - bp) - φ_{i}) \\ i = 1, 2, \dots, n \end{matrix}$

5.2 Income function of railway enterprises

(1) the number of trains operating on the tourism special train is $q = ⌈ \frac{d (x_{i}, \dots, x_{n}, p)}{θ A} ⌉ = ⌈ \frac{\sum_{i = 1}^{n} x_{i} (a_{i} - bp)}{θ A} ⌉ .$

(2) income

The income of railway enterprises mainly comes from consumer-oriented income d (x_i, ⋯ , x_n, p) × p and subsidies given by local governments $\sum_{i = 1}^{n} x_{i} φ_{i}$ , that is, $\begin{matrix} V_{inco} = d (x_{i}, \dots, x_{n}, p) \times p + \sum_{i = 1}^{n} x_{i} φ_{i} \\ = \sum_{i = 1}^{n} x_{i} (a_{i} - bp) p + \sum_{i = 1}^{n} x_{i} φ_{i} \\ = \sum_{i = 1}^{n} x_{i} [(a_{i} - bp) p + φ_{i}] \end{matrix}$

(3) Cost

The cost of the railway enterprise includes two parts: the fixed cost and the variable cost of the operation of the special tourism train, among which the fixed cost includes the related investment expenses of non-current assets such as lines, stations, mobile equipment such as EMU, and service facilities and equipment in the train, that is, $\begin{matrix} C_{fixed} = \sum_{i = 1}^{q} c_{fixed} (T_{k}) = q \times c_{fixed} \\ = ⌈ \frac{\sum_{i = 1}^{n} x_{i} (a_{i} - bp)}{θ A} ⌉ c_{fixed} \end{matrix}$

One of the variable costs is electricity, water and other related expenses, that is, $C_{oper} = \sum_{k = 1}^{q} \sum_{j = 1}^{m} F (u (T_{k}), r (e_{j})) δ (T_{k}, e_{j}) l (e_{j})$

The second variable cost is the expenses related to stopping, that is, $C_{stop} = \sum_{k = 1}^{q} \sum_{i = 1}^{n} x_{i} c_{stop} (T_{k})$

The third variable cost is the cost related to transferring, that is, $C_{tran} = \sum_{k = 1}^{q} \sum_{i = 1}^{n} x_{i} f_{tran} (T_{k})$

The fourth variable cost is the tourism consumption expenditure after the stop, that is, $\begin{matrix} C_{trav} = \sum_{i = 1}^{n} x_{i} f_{i} (t_{i}) d (x_{i}, \dots, x_{n}, p) \\ = \sum_{i = 1}^{n} (x_{i} {α t}_{i} \sum_{z = 1}^{n} x_{z} (a_{z} - bp)) \end{matrix}$

To sum up, the income function of railway enterprises is:

$\begin{matrix} π^{2} ({x_{1}, \dots, x_{n}, t_{1}, \dots, t_{n}, p | φ}_{i}) \\ = V_{inco} - C_{fixed} - C_{oper} - C_{stop} - C_{tran} - C_{trav} \\ = \sum_{i = 1}^{n} x_{i} [(a_{i} - bp) p + φ_{i}] \\ - ⌈ \frac{\sum_{i = 1}^{n} x_{i} (a_{i} - bp)}{θ A} ⌉ c_{fixed} \\ - \sum_{k = 1}^{\sum_{i = 1}^{n} x_{i} (a_{i} - bp) / (θ A)} \sum_{j = 1}^{m} F (u (T_{k}), r (e_{j})) δ (T_{k}, e_{j}) l (e_{j}) \\ - \sum_{k} = 1^{\sum_{i = 1}^{n} x_{i} (a_{i} - bp) / (θ A)} \sum_{i = 1}^{n} x_{i} [c_{stop} (T_{k}) + f_{tran} (T_{k})] \\ - \sum_{i = 1}^{n} [x_{i} {α t}_{i} \sum_{z = 1}^{n} x_{z} (a_{z} - bp)] \end{matrix}$

5.3 Decision making model of railway enterprise operation plan

Given the game theory considered in this paper, the railway enterprise primarily decides the market price of the tourism special train, then the local government determines the subsidy, and finally the railway enterprise decides the stopping plan of the tourist dedicated train. At this time, the decision model of the railway enterprise operation plan is as follows: $max π^{2} (p | x_{1}, \dots, x_{n}, t_{1}, \dots, t_{n}, φ_{i})$ $s . t . {\begin{matrix} max π_{1}^{1} (φ_{1} | x_{1}, t_{1}, p), i = 1, 2, \dots, n . \\ s . t . \max π^{2} (x_{1}, \dots, x_{n}, t_{1}, \dots, t_{n} | p, φ_{1}), \\ θ_{d} ⩽ {d (x}_{1}, \dots, x_{n}, p) / A ⩽ θ_{u}, \\ \sum_{1 - 1}^{n} x_{1} t_{1} ⩽ \hat{T} = ω τ . \\ q is a positive integer . \\ x_{1} = 0 or 1, i = 1, 2, \dots, n . \\ t_{1} is an integer, i = 1, 2, \dots, n . \\ δ (T_{k}, e_{j}) = 0 or 1, j = 1, 2, \dots, m . \end{matrix}$

6 Proposed algorithm

The solution of the model is essentially a combinatorial optimization problem. Because of the complexity of the model and some independent variables are discrete integers, the problem itself belongs to the NP hard problem. In reality, the optimal solution can not be obtained, but the heuristic algorithm is used to solve it. With reference to Givens and Hoeting [6] for the combinatorial optimization problem of the general solution method, random start local search method, combined with the constraints of the model, the following algorithm was developed to solve the problem. The main optimization idea is to randomly select multiple starting points, and then conduct a random search near the starting point, and update our solution when we encounter a better solution than at present. After searching for a certain number of steps, our solution will gradually converge to a better solution, which can control the computing time and meet the actual needs at the same time. Meanwhile, by randomly selecting multiple starting points, the quality of the solution can be effectively improved and the problem of falling into local optimization caused by local search can be alleviated. Since t_i = 0 means x_i = 0, t_i > 0 means x_i = 1, thus we simplify the problem to that of t₁, …, t_n, p and get the algorithm is as follows:

for t₁, …, t_n, p random assignment, for example: t₁, …, t_n is 0 with a probability of 0. 4 and obeys a uniform distribution over 0–5 days with a probability of 0. 6, and p obeys a uniform distribution on (4000-3000 b∼6000-3000 b).

for the current t₁, …, t_n, p The local government gives a subsidy scheme based on the length of stay allocated by tourism enterprises and the initial demand that can be attracted by the local government.

for the current t₁, …, t_n, p and the government subsidy program, calculate $π_{i}^{2}$ , and then apply random-perturbation to t₁, …, t_n, p. At the same time, it is necessary to ensure t₁, …, t_n, p satisfy the constraint, t₁, …, t_n, p > 0, and then calculate the $(π_{i}^{2})^{*}$ for the perturbed t₁, …, t_n, p . The solution is updated if $π_{i}^{2} < (π_{i}^{2})^{*}$ . Otherwise, the solution is kept unchanged and a new random disturbance is carried out, and this step is cycled several times until it converges to a better decision for the railway enterprise.

cycle the step 2-3 for several times and record the final solution.

Select several new starting points, cycle step 1–4, and finally select the optimal solution of all solutions as the final solution of our algorithm.

7 Sensitivity analysis and evaluation

Suppose that there are 8 stations which are serial in sequence in the railway network G (S, E) of a given fixed operation section, so there are 7 road sections (m = 7), the length of which is (146, 145, 225, 747, 501, 223, 298), and the train with 600 personnel is considered. The special tour train is at the station s_i (i = 1, …, n) may or may not stay, the length of stay at the station s_i is recorded, t_i = 0 means no stay. The initial demand of the special tourism train that can be attracted by the location of the station s_i (i = 1, 2, ⋯ , n) is a = (1300, 1400, 1000, 1900, 1500, 1800, 1400, 1800), b = 0.2. is the sensitivity of price, the cost of tourism per capita is assumed as α = 800. Next, we need to optimize the parameters for the residence time t₁, …, t₈, government subsidy φ₁, …, φ₈. According to the algorithm in section 5, 100 random starting points are selected, and each random point is iterated and optimized 150 times, and the result is shown in Fig. 2.

Fig. 2

Iterative optimization results.

The horizontal axis in Fig. 2 is the ordinal number of iterations, and the vertical axis is the profit of the tourism enterprise, and each trajectory represents the optimization starting from each random starting point. After optimizing the parameter space, we choose the optimal combination that makes the highest profit (the search path represented by the red line). In the end, the government subsidy is (303, 0, 0, 443, 0, 420, 0, 0), and the stay time is (1, 0, 0, 1, 0, 1, 0, 0), the ticket price is RMB 4241, corresponding to the highest profit is RMB 476571, the total number of tourists is 2455, the per capita profit is RMB 194, profit margin is about 4.58%. The Fig. 2 shows the final optimal decision of the government for the five cities that do not stay (stay time t_i = 0) is not to subsidize. The cities that subsidize and eventually win the tourism special train to stay are generally the cities with high tourism attractiveness.

Furthermore, we observe the corresponding changes of the optimization results by adjusting the initial demand of the tourist area where the station is located in the above example. We increased the initial demand of the third station from 1000 to 2000, and the initial demand of the fourth station from 1900 to 1100, that is, a = (1300, 1400, 2000, 1100, linebreak 1500, 1800, 1400, 1800). After the same optimization algorithm, the result is shown in Fig. 3. In Fig. 3, the final government subsidy is (0, 0, 933, 0, 0, linebreak 0, 327, 420), the length of stay is (0, 0, 2, 0, 0, 0, linebreak 1, 1), the ticket price is RMB 5364, the final profit is RMB 364728, the total number of tourists is RMB 1982, the per capita profit is RMB 184, and the profit margin is about 3.43%. It is obvious that when the number of people at the third station increases, the tourism special train will change from not stopping before to allocating two days’ stay. After the tourism demand of the fourth station is reduced to 1100 people, the tourist dedicated train will no longer stay at the fourth station in the optimal scheme.

Fig. 3

Iterative optimization results after initial demand changes.

Then, we examine the impact of price-sensitive parameters on the tourism decision-making of railway enterprises, focusing on the optimal price and the optimal income of railway enterprises. Keeping the remaining parameters in the above example unchanged, the coefficient of price elasticity b of passengers is gradually increased from 0.1 to 0.25. Through the same optimization algorithm, the changes of optimal price and optimal income of railway enterprises with the decrease of b value are shown in Figs. 4 and 5. As can be seen from Figs. 4 and 5, the optimal price and optimal return decrease with the increase of the coefficient of price elasticity, that is to say, the higher the price sensitivity of consumers, the smaller the optimal price and optimal income of railway enterprises. In practice, people usually think that the consumers who choose railway tourism belong to the market group with weak tourism consumption power, and this part of the consumer group is often sensitive to the price. The conclusion of Fig. 4 enlightens us that in order to increase income, railway enterprises need to expand the middle and high-tier market groups with strong consumption power. In order to attract this part of the market, we need relatively high-end tourism products or services. Therefore, the development of medium-and high-end railway tourism products or services is the business strategy that railway enterprises need to focus on.

Fig. 4

The effect of price elasticity coefficient on the optimal price of railway enterprises.

Fig. 5

The effect of price elasticity coefficient on the optimal profit of railway enterprises.

8 Conclusion

The rapid increase in the operating mileage of China’s high-speed railway not only relies the foundation for the development of tourist dedicated trains, but also provides more opportunities for railway enterprises to depend on railway transport resources to meet the needs of social tourism. By comprehensively considering the game theory between railway enterprises and local governments, and under the condition that railway enterprises optimize the market price decision and the stopping plan along the tourism special train at the same time, this paper constructs the decision-making model of the operation plan of tourist dedicated train, and designs the model solving algorithm, which provides theoretical support for the railway enterprises to formulate the plan more systematically and scientifically. Through numerical experiments, it is found that the sensitivity of consumers to prices is negatively correlated with the pricing and revenue of tourist dedicated train. The lower the sensitivity of prices is, the higher the fares and profits of the trains of railway enterprises are. The initial scale of tourism demand along the railway has a direct impact on the decision-making of whether the special tourism train stops or not, and the larger the demand scale is, the greater the possibility of special train to stop in its corresponding geographic area. As far as railway enterprises are concerned, to vigorously develop tourist dedicated train, we need to fully investigate and grasp the market demand, and also need to cooperate with local governments to expand the marginal effect of tourist dedicated train, in order that railway tourism can play its fully role in poverty alleviation and economic growth of local governments. At the same time, it not only maximizes its own economic benefits, but also plays a greater role in promoting local development along the line.

In the future, the research can be further expanded from two aspects: one is to use AI Agent to study the game theory in dynamic between many local governments and railway enterprises; while the other is to comprehensively consider the tourist dedicated train, tougher with high-speed passenger cars and general-speed trains, and build the decision-making model of operation plan in the complex railway operation network environment.

References

Zhang

, Liu

and Liu

, High-speed rail, tourist mobil-ity, and firm value[J], Economic Modelling90 (2020), 108–116.

Guirao

and Campa

J.L.

, Cross Effects between High Speed Rail Lines and Tourism: Looking for Empirical Evi-dence Using the Spanish Case Study[J], Transportation Research Procedia14 (2016), 392–401.

Uwimana

, Predicting Tourism Demands by Google Trends: A Hidden Markov Models Based Study[J], Journal of System and Management Sciences10(1) (2020), 106–120.

Wang

D-g.

, Niu

and Qian

, Evolution and optimization of China’s urban tourism spatial structure: A high speed rail perspective[J], Tourism Management64 (2018), 218–232.

Pagliara

, Mauriello

and Garofalo

, Exploring the interdependences between High Speed Rail systems and tourism: Some evidence from Italy[J], Transportation Research Part A: Policy and Practice106 (2017), 300–308.

Givens

G.H.

and Hoeting

J.Á.

, Computational statistics[M], John Wiley & Sons (2012).

Jia

, Benefit-Cost Study on Evaluating Tour-Train Circulating-Plan[J], Journal of Northern Jiaotong University26(2) (2002), 87–90.

Campa

J.L.

, López-Lambas

M.E.

and Guirao

, High speed rail effects on tourism: Spanish empirical evidence derived from China’s modelling experience[J], Journal of Transport Geography57 (2016), 44–54.

Dilogini

, Shivany

and Kumara

, Analysis of Relation Between Customer Behavior and Information Technology Market[J], Journal of System and Management Sciences9(1) (2019), 87–104.

10.

Fang

, Jiangshun

and Jijun

, Design of Tour Guide Outsourcing Service Contract under Dual Principal-Agent[J], Journal of System and Management Sciences10(1) (2020), 32–50.

11.

and Wall

, The Qinghai-Tibet Railway and Tibet Tourism: Travelers’ Perspectives[J], Tourism Management30(5) (2009), 650–657.

12.

Jin

, Yang

, Wang

and Liu

, The influence of high-speed rail on ice–snow tourism in northeastern China[J], Tourism Management78 (2020), 104070.

13.

Masson

and Petiot

, Can the high speed rail reinforce tourism attractiveness? The case of the high speed rail between Perpignan (France) and Barcelona (Spain)[J], Technovation29(9) (2009), 611–617.

14.

Wan

, Jiang

, Qin

and Du

, Pricing Decision And Coordination Contract In Low-Carbon Tourism Supply Chains Based On Altruism Preference[J], Environmental Engineering and Management Journal18(11) (2019), 2501–2518.

15.

Wang

, Huang

, Zou

and Hui

, Effects of the high speed rail network on China’s regional tourism development[J], Tourism Management Perspectives1 (2012), 34–38.

16.

Yang

, Sun

and Song

, Research on Tourist-dedicated Train Operation Scheme and Related Issues, Railway Transport and Economy6 (2001), 39–41.

17.

Zeng

, Research on Tourist-dedicated Train Operation Scheme of Cuiyang-Guangzhou High-speed Railway[D], Beijing Jiaotong University (2015).