Ticket price system setting method based on multi-objective optimization and dynamic adjustment

Abstract

The invention provides a ticket price system setting method based on multi-objective optimization and dynamic adjustment, and relates to the technical field of asset income management, and the method comprises the steps: collecting order data of all seats starting from one place to another place in a travel interval, and building a cost function; increasing price point density in the key area, filling additional price points by using an interpolation method to obtain a potential execution price set, and calculating transportation income under a price combination by using an income calculation model by using traversal calculation and a fast elite multi-objective genetic algorithm to obtain an optimal price combination; and verifying the setting result, establishing a dynamic adjustment mechanism, and adjusting the ticket price system setting method based on the real-time sales data, the market feedback continuous optimization price point and the verification result so as to ensure the final optimization effect of the ticket price system setting method. According to the method, a group of optimal prices are selected from alternative ticket prices as execution prices through research on a demand price curve.

Description

Technical Field

[0001] The present invention relates to the technical field of asset income management, and in particular to a fare system setting method based on multi-objective optimization and dynamic adjustment. Background Art

[0002] In revenue management, the optimization of the fare system is a complex issue, which involves multiple conflicting goals, such as maximizing revenue, minimizing costs, and customer satisfaction. The traditional fare setting method mainly relies on experience and historical data, lacks quantitative management methods, which greatly limits the scientific nature and flexibility of fare setting. With the rapid development of high-speed railways and the steady improvement of people's living standards, high-speed railways are quietly changing people's lifestyles. While high-speed railways are rising rapidly, the problems of their fare system are gradually emerging. Traditional pricing methods have restricted the development of high-speed railways to some extent. There are problems with the existing fare system, such as the untimely adjustment of the fare of old high-speed railway lines, and the ticket refund and change system does not take into account the preferential ratio of the tickets purchased by passengers, resulting in a waste of high-speed rail capacity. In addition, the existing fare system fails to fully reflect market demand and competition conditions, and lacks flexibility and adaptability. Summary of the invention

[0003] The purpose of the present invention is to provide a fare system setting method based on multi-objective optimization and dynamic adjustment to improve the above problems. In order to achieve the above purpose, the technical solution adopted by the present invention is as follows: In a first aspect, the present application provides a method for setting a fare system based on multi-objective optimization and dynamic adjustment, comprising: Collect the order data of all seats departing from one location to another within the travel range, construct the demand price function through regression analysis, and establish the cost function through cost data analysis; Based on the cost function, the cost coverage price is taken as the lower limit, and a set of optional price points is established within this range. The price point density is increased in key areas, and additional price points are filled using interpolation to ensure that the entire price range is covered, thereby obtaining a set of potential execution prices. The parameters collected in the optional price point set include the mileage from the starting point to the arrival point, transportation time, vehicle operation cost data, passenger flow data, and time value data. According to the potential execution price set, the traversal calculation and fast elite multi-objective genetic algorithm are used to calculate the transportation revenue under the price combination using the revenue calculation model to obtain the expected maximum revenue. The expected maximum revenue is used as the objective function, and the seat inventory constraint corresponding to each price point in the combination is introduced to search and screen the potential execution price combination set on the Pareto frontier to obtain the optimal price combination, and then the optimal price combination is used as the setting result of the fare system. The revenue calculation needs to take into account the price, demand, seat inventory, discount and additional fees. Verify the setting results, establish a dynamic adjustment mechanism, continuously optimize price points based on real-time sales data and market feedback, and adjust the fare system setting method based on the verification results to ensure the ultimate optimization effect of the fare system setting method.

[0004] Preferably, the order data of all seats departing from one location to another location within the travel interval are collected, and a demand price function is constructed through regression analysis, which includes: Collect the order data of all seats departing from one location to another within the travel interval, identify and record all different price levels within the time period, and mark them in order from low to high in price to obtain the number of orders for each price level; Based on the order quantity of each price level, after demand analysis and processing, the first demand data and the second demand data are obtained, wherein the first demand data is the order data corresponding to the price level, and its demand is the sum of the demand of the orders at the price level; wherein the second demand data is the order data corresponding to the price level not less than that, and its demand is the sum of the demand of the orders from the price level to the highest price level; According to the second demand data, the regression analysis method is used to fit the relationship between price and demand, and then the demand price function of the impact of price changes on passenger demand when a fixed fare is implemented is obtained.

[0005] Preferably, the cost function is based on the cost coverage price as the lower limit, and a set of optional price points is established within this range. The price point density is increased in the key area, and the extra price points are filled by interpolation to ensure that the entire price range is covered, so as to obtain a set of potential execution prices, which includes: Use multiple regression to analyze historical sales data and market information to construct a demand price function; Based on the demand price function, use the Monte Carlo simulation method to consider market volatility factors and calculate the risk-adjusted cost coverage price as the lowest acceptable price; Use the linear programming method to consider market capacity and competition factors to determine the highest feasible price; Establish a continuous price range between the lowest acceptable price and the highest feasible price; Set corresponding price points for major customer groups based on continuous price ranges, set corresponding price points for significant seasonal fluctuations based on seasonal analysis, and allocate additional price points around the main competitors' prices taking into account competition; Based on the additional price points, the optional price point set is determined by filling the remaining price points using an interpolation method according to the given number of price points.

[0006] Preferably, the ergodic calculation and fast elite multi-objective genetic algorithm are used to calculate the transportation revenue under the price combination using the revenue calculation model to obtain the expected maximum revenue, which includes: Calculate the passenger demand for each price class when considering a single fare; Based on passenger demand and considering the coexistence of multiple prices in the price combination, the demand forecast value of each price level is calculated from high to low according to the principle of downward compatibility of demand; Based on the demand forecast and the corresponding ticket price, the expected revenue when multiple prices are adopted in parallel is obtained through revenue calculation. Compare the expected revenues under different price combinations, and record the price combination that maximizes the revenue as the expected maximum revenue for the travel segment.

[0007] Preferably, the traversal calculation and fast elite multi-objective genetic algorithm are used to calculate the transportation revenue under the price combination using the revenue calculation model to obtain the expected maximum revenue, and the expected maximum revenue is used as the objective function. The seat inventory constraint corresponding to each price point in the combination is introduced, and the potential execution price combination set on the Pareto frontier is searched and screened to obtain the optimal price combination, wherein the calculation steps are as follows: Step 1: Consider different potential price combinations according to the price point set, and use different potential price combinations as encoding schemes for different individual designs. Generate a set of random initial solutions based on the initial population of the fast elite multi-objective genetic algorithm, where each initial solution has a corresponding selection model configuration to ensure that the initial population covers all parts of the solution space. Step 2: Based on the initial solution, evaluate the fitness of each individual, and use the selection model to calculate the passenger selection probability, and calculate the first objective function and the second objective function. The first objective function is to maximize the expected return, and the second objective function is to minimize the number of prices in the price combination; Step 3: According to the first objective function and the second objective function, the population is non-dominated sorted, wherein the non-dominated sorting includes finding all non-dominated solutions and forming a first frontier, and continuing to find non-dominated solutions in the remaining solutions to form the next frontier, and repeating the iteration until all solutions are classified; Step 4: Based on all the classified solutions in step 3, calculate the normalized distance between adjacent solutions, where the total congestion distance of the solution is the sum of its distances on all objective functions; Step 5: Randomly select two individuals and compare their non-dominated sorting levels. If the levels are the same, select the individual with a larger crowding distance. Repeat this process until enough parent individuals are selected. Step 6: Merge the parent and offspring populations, perform non-dominated sorting and crowding distance calculation on the merged population, and select individuals with higher sorting and larger crowding distance to form a new generation population; Step 7: Check whether the maximum number of iterations or other termination conditions have been reached. If not, continue iterating. If reached, output the final set of non-dominated solutions and analyze the trade-offs between different solutions. Step 8. Based on the final set of non-dominated solutions, the Pareto solutions are processed through the analytical tools and models in the decision support system, and the final optimal price combination is selected from the Pareto frontier in combination with the actual operating conditions and decision maker preferences. The optimal price combination maximizes the expected return of the travel segment while satisfying the operating conditions.

[0008] In the second aspect, the present application also provides a fare system setting system based on multi-objective optimization and dynamic adjustment, including: Collection module: used to collect the order data of all seats departing from one location to another within the travel interval, build the demand price function through regression analysis, and establish the cost function through cost data analysis; Establishment module: Based on the cost function, the cost coverage price is used as the lower limit, and a set of optional price points is established within this range. The density of price points is increased in key areas, and additional price points are filled using interpolation to ensure that the entire price range is covered, so as to obtain a set of potential execution prices. The parameters collected in the optional price point set include the mileage from the starting point to the arrival point, the transportation time, the vehicle operation cost data, the passenger flow data, and the time value data; Calculation module: It is used to calculate the transportation revenue under the price combination based on the potential execution price set, using traversal calculation and fast elite multi-objective genetic algorithm to apply the revenue calculation model, and obtain the expected maximum revenue. The expected maximum revenue is used as the objective function, and the seat inventory constraint corresponding to each price point in the combination is introduced to search and screen out the potential execution price combination set on the Pareto frontier to obtain the optimal price combination, and then use the optimal price combination as the setting result of the fare system. The revenue calculation needs to consider the price, demand, seat inventory, discount and additional fees; Adjustment module: used to verify the setting results, establish a dynamic adjustment mechanism, continuously optimize price points based on real-time sales data and market feedback, and adjust the fare system setting method based on the verification results to ensure the final optimization effect of the fare system setting method.

[0009] In a third aspect, the present application also provides a fare system setting device based on multi-objective optimization and dynamic adjustment, including: Memory for storing computer programs; A processor is used to implement the steps of the fare system setting method based on multi-objective optimization and dynamic adjustment when executing the computer program.

[0010] In a fourth aspect, the present application also provides a readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the above-mentioned method for setting a fare system based on multi-objective optimization and dynamic adjustment are implemented.

[0011] The beneficial effects of the present invention are: In view of the limitations of the prior art, the present invention proposes a multi-objective optimization algorithm based on non-dominated sorting, namely, an improved non-dominated sorting genetic algorithm. By introducing fast non-dominated sorting and crowding distance calculation, the present invention effectively solves the problems of identifying non-dominated solutions and maintaining the diversity of solution sets in multi-objective optimization problems. Compared with traditional multi-objective optimization algorithms, the present invention not only improves the computational efficiency of the algorithm, but also does not require additional parameter settings, making the algorithm more concise and easy to implement. It can automatically find a set of balanced solutions, namely, the Pareto optimal solution set, and provides decision makers with a comprehensive selection of optimization solutions, thereby playing a key role in multi-objective decision-making.

[0012] The present invention includes assets such as airline seats, high-speed rail seats or hotel rooms. Through the study of the demand price curve, a set of optimal prices are selected from the alternative ticket prices as the execution price; starting from the demand price curve, the present invention helps the ticket price management personnel select the various levels of ticket prices that need to be executed from a large number of different discounts by maximizing the expected return. The algorithm can be used to compile software, and the software directly recommends executable ticket price combinations.

[0013] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0015] Figure 1 It is a flow chart of a fare system setting method based on multi-objective optimization and dynamic adjustment described in an embodiment of the present invention; Figure 2 It is a schematic diagram of the structure of a fare system setting system based on multi-objective optimization and dynamic adjustment described in an embodiment of the present invention; Figure 3 It is a schematic diagram of the structure of a fare system setting device based on multi-objective optimization and dynamic adjustment described in an embodiment of the present invention; Figure 4 It is a schematic diagram of a price demand curve of a fare system in revenue management described in an embodiment of the present invention.

[0016] In the figure: 701, collection module; 702, establishment module; 703, calculation module; 704, adjustment module; 800, optimization setting device of fare system in revenue management; 801, processor; 802, memory; 803, multimedia component; 804, I / O interface; 805, communication component. DETAILED DESCRIPTION

[0017] To make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. The components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the invention claimed for protection, but only represents the selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present invention.

[0018] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0019] Embodiment 1:

[0020] This embodiment provides a fare system setting method based on multi-objective optimization and dynamic adjustment.

[0021] See also Figure 1 , the figure shows that the method includes step S100, step S200, step S300 and step S400.

[0022] S100, collecting order data of all seats departing from one location to another within the travel range, constructing a demand price function through regression analysis, and establishing a cost function through cost data analysis.

[0023] It can be understood that step S100 includes S101, S102 and S103, wherein: S101, collecting order data of all seats departing from one location to another location within the travel interval, identifying and recording all different price levels within the time period, and marking them in order from low to high in terms of price, and obtaining the order quantity of each price level; S102, based on the order quantity of each price level, after demand analysis and processing, obtain first demand data and second demand data, wherein the first demand data is the order data corresponding to the price level, and its demand is the sum of the order demands of the price level; wherein the second demand data is the order data corresponding to the price level not less than the price level, and its demand is the sum of the order demands from the price level to the highest price level; S103. According to the second demand data, a regression analysis method is used to fit the relationship between price and demand, and then a demand price function is obtained for the effect of price changes on passenger demand when a fixed fare is implemented.

[0024] It should be noted that the process of obtaining the demand price function based on order data is as follows: (1) Select the order data of all seats of the OD within a certain period of time. There are n price levels in total. Count the price levels of all seats ,in , count the number of orders at each price level .

[0025] (2) Assume that passengers with low ticket prices for high-speed rail products do not accept high-priced products, while passengers with high ticket prices can accommodate low-priced products. Then, the lowest price Corresponding needs For all order requirements of this OD, that is: ,in The lowest price level. To execute the lowest level price The passenger demand at time n is the number of price levels, i is the summation variable, For price level order data; by extension, price levels Corresponding needs ,in, Sort the price from low to high by any sequence number k, i is the sum calculation variable, To sort the price levels of k from low to high, Execution price level passenger demand at that time, For price level Order data.

[0026] (3) Price level according to step (2) Corresponding needs ,in, Sort the price from low to high by any sequence number k, i is the sum calculation variable, Execution price level passenger demand at that time, For price level Order data.

[0027] Then, the demand price function is determined by function fitting. ,in is the price, D is the number of passengers required, It's the price is the demand function expression of the variable.

[0028] It should be noted that, in this embodiment, the demand function is a monotonically decreasing function, which can be a linear or nonlinear function. , p is a fixed price, D(p) is the demand when the fixed price p is executed, a and b are parameters. When this function is used to simulate the demand function, the demand for each price level is calculated based on the statistical calculation of the order data: ,in, To sort the price from low to high, sort any sequence number k, is the price level sorted from low to high as k, i is the sum calculation variable, Execution price level passenger demand at that time, For price level Order data.

[0029] When using a nonlinear function, which can be a polynomial to simulate demand, it can be in the following two function forms: 1. Use quadratic function to simulate the demand function form: Set the quadratic function , p is a fixed price, To implement the demand at a fixed price p, a, b, and c are parameters, based on the statistical calculation of the demand for each price level based on the order data: ,in, To sort the price from low to high, sort any sequence number k, is the price level sorted from low to high, i is the sum calculation variable, Execution price level passenger demand at that time, For price level Order data. Use the least squares method to find the parameters a, b, and c.

[0030] 2. Use exponential function to simulate the demand function form: Set the exponential function , p is a fixed price, To implement the demand when the fixed price p is used, a and b are parameters, and the demand for each price level is calculated based on the statistics of the order data: ,in, To sort the price from low to high, sort any sequence number k, is the price level sorted from low to high as k, i is the sum calculation variable, Execution price level passenger demand at that time, For price level The order data is obtained by using the least squares method to obtain the a and b parameters.

[0031] S200. Based on the cost function, the cost coverage price is taken as the lower limit, and a set of optional price points is established within this range. The price point density is increased in key areas, and additional price points are filled using interpolation to ensure that the entire price range is covered, thereby obtaining a set of potential execution prices, wherein the parameters collected in the optional price point set include mileage from the starting point to the arrival point, transportation time, vehicle operation cost data, passenger flow data and time value data, route construction cost, vehicle purchase price, depreciation time, electricity cost and number of seats.

[0032] In this step, the vehicle operation cost data includes fuel consumption, vehicle depreciation, maintenance costs, etc. These data can be obtained through standard data provided by vehicle manufacturers or actual operation data. These cost data are proportional to the mileage and are the basic factors that must be considered in fare setting. The passenger's time value can be obtained through surveys or market research, reflecting the passenger's willingness to pay for time saving. The shortening of transportation time can improve the attractiveness of the service, but it may also increase the cost of the operator, so the time value needs to be reflected in the fare. Passenger flow data can obtain passenger flow data in different time periods through historical operation data and market analysis. These data are helpful to predict passenger demand at different price points and thus affect fare setting. Using historical order data, cost data, mileage, transportation time and other objective physical data and economic parameters, the demand price function and cost function are constructed, that is, the above physical data are collected and analyzed to determine the relationship between them and the fare. Based on the analysis results, a multi-objective optimization model is established, considering the mileage, transportation time and other physical data, and applying multi-objective optimization algorithms such as NSGA-II to search and screen out the potential execution price combination set on the Pareto frontier and dynamically adjust the fare. Therefore, the analysis and processing of these data provide a scientific basis for fare setting.

[0033] It can be understood that step S200 includes S201, S202 and S203, wherein: S201. Use multiple regression to analyze historical sales data and market information to construct a demand price function; based on the demand price function, use the Monte Carlo simulation method to consider market volatility factors and calculate the risk-adjusted cost coverage price as the lowest acceptable price; use the linear programming method to consider market capacity and competition factors to determine the highest feasible price; establish a continuous price range between the lowest acceptable price and the highest feasible price; It should be noted that, first, relevant order data is collected. These data include all order information of different seat types within the travel range, as well as the price and demand corresponding to each order. Before applying the K-means algorithm, it is necessary to determine the number of clusters, that is, the number of price segments, where this number can be set according to business needs, market research or prior knowledge; the price in the order data is input as a feature into the K-means algorithm, and the algorithm will divide the data points into several clusters according to the similarity of the price, each cluster represents a price segment; for each price segment, its center point is calculated, that is, the average of all prices in the segment. This center point represents the typical price level of the price segment and is a potential optional price; price elasticity is an economic concept that measures the sensitivity of demand to price changes. When determining the center point, considering price elasticity can help us better understand the impact of price changes on passenger demand, so as to choose a more attractive price point. Therefore, objective clustering analysis of price distribution through the K-means algorithm can avoid the deviation caused by subjective judgment and make price setting more scientific and reasonable. The center point of each price segment reflects the general acceptance of passengers in the market for different price levels, which helps airlines better adapt to market demand.

[0034] S202. According to the continuous price range, set corresponding price points for major customer groups, set corresponding price points for significant seasonal fluctuations based on seasonal analysis, and allocate additional price points around the prices of major competitors in consideration of competition; S203. Based on the additional price points, according to the given number of price points, an interpolation method is used to fill in the remaining price points to determine a set of optional price points.

[0035] Wherein, based on the determined optional price point set, a certain number of prices are screened out as potential execution prices, the number of all possible price combinations is calculated, and a price combination selection set is obtained. The combination set quantity calculation formula is as follows: In the formula, The number of combinations of selecting s elements from m independent elements, m ! represents the factorial of m, which is the product of all integers from 1 to m. s ! represents the factorial of s, that is, the product of all integers from 1 to s, m is the total number of elements to be selected, and s is the number of elements to be selected.

[0036] It should be noted that in the simulated annealing algorithm, it is usually necessary to define an energy function that evaluates the "energy" or cost of a given price combination. In the pricing strategy, the energy function can be designed to be the opposite of the expected return, that is, the cost is not the return that is expected to be minimized. The algorithm "moves" by randomly selecting price points and calculates the "energy" of the new price combination. If the new combination has lower energy (better return), the move is accepted. If the energy is higher, the move is still accepted with a certain probability, which is to allow the algorithm to jump out of the local optimal solution. As the algorithm proceeds, the "temperature" is gradually lowered, which reduces the probability of accepting high energy states, so that the search process gradually stabilizes. After enough iterations, when the system "cools" to a certain temperature, the algorithm tends to stabilize, and a certain number of optimal price points are screened out as potential execution prices. By accepting a worse solution with a certain probability, the algorithm can effectively avoid falling into the local optimal solution and increase the chance of finding the global optimal solution. Therefore, in actual research, changes in price elasticity can also be considered, and the parameters of the algorithm can be adjusted through real-time data feedback to make the pricing strategy more flexible and responsive to market changes. In addition, the price combinations selected by the algorithm can be actually tested through methods such as A / B testing to evaluate their impact on sales and revenue, thereby continuously optimizing pricing strategies.

[0037] It can be understood that step S300 includes S301, S302, S303 and S304, wherein: S301. Calculate the passenger demand at each price level when considering a single fare. The calculation formula is as follows:

[0038] In the formula, is the price, D is the number of passengers required, For price is the demand function expression of the variable; S302, considering the coexistence of multiple prices in the price combination based on passenger demand, and calculating the demand forecast value of each price level from high to low according to the principle of downward compatibility of demand; It should be noted that the price combination When multiple price levels coexist, the demand of passengers in high price levels is considered to be downward compatible, and the demand forecast value of each price level is calculated from high to low.

[0039] S303. According to the demand forecast value and the corresponding ticket price, the expected revenue when multiple prices are used in parallel is obtained through revenue calculation. The calculation formula is as follows:

[0040] In the formula, is the expected maximum revenue of the travel segment, k is the price combination adopted The lowest price level is sorted, and s is the price combination The number of prices in the equation, i is the price combination Price sorting number, Price combination The sorting in is the price level of i, for The demand corresponding to the price level.

[0041] S304. Compare the expected benefits under different price combinations, and record the price combination that maximizes the benefit as the expected maximum benefit of the travel segment.

[0042] Therefore, for high-speed trains with a high running density, it is assumed that the high-speed trains can meet all demands at all consumption levels, without considering the limitation of the high-speed train stock.

[0043] It should be noted that the maximum potential revenue under the seat capacity constraint is determined while ensuring that the demand at each price point does not exceed the seat inventory. When calculating the maximum potential revenue, the demand at each price point is considered to be constrained as follows:

[0044] In the formula, is the expected maximum revenue of the travel segment, k To adopt price combination Sort by lowest price tier, s Price combination The price quantity in i Price combination Price sorting number, Price combination The sorting in is i Price level, for The demand corresponding to the price level, To adopt price combination The lowest price among all the options, N is the seat stock of the transportation vehicle.

[0045] Consider the number of available seats, compare the flight revenue of each combination, and select the maximum value , the corresponding price combination is the selected optimal price combination.

[0046] Specifically, the present invention is not only applicable to the determination of the flight seat sale fare system, but also applicable to commodities such as high-speed rail tickets, hotel rooms, cruise ships, etc. that require multi-level fare pricing.

[0047] Specifically, if the number of seats provided by a transportation vehicle is huge relative to the demand for a certain interval, for example, the number of seats provided by all trains of the Chengdu-Chongqing high-speed railway is huge relative to the demand for Chengdu-Chongqing, then there is no need to consider the limit of seat inventory and the maximum value can be directly selected. The corresponding price combination is used as the recommended execution fare combination.

[0048] It should be noted that the traversal calculation and fast elite multi-objective genetic algorithm in S300 use the revenue calculation model to calculate the transportation revenue under the price combination, obtain the expected maximum revenue, take the expected maximum revenue as the objective function, introduce the seat inventory constraint corresponding to each price point in the combination, search and screen out the potential execution price combination set on the Pareto frontier, and obtain the optimal price combination. The calculation steps are as follows: Step 1: Consider different potential price combinations according to the price point set, and use different potential price combinations as encoding schemes for different individual designs. Generate a set of random initial solutions based on the initial population of the fast elite multi-objective genetic algorithm, where each initial solution has a corresponding selection model configuration to ensure that the initial population covers all parts of the solution space. Step 2: Based on the initial solution, evaluate the fitness of each individual, and use the selection model to calculate the passenger selection probability, and calculate the first objective function and the second objective function. The first objective function is to maximize the expected return, and the second objective function is to minimize the number of prices in the price combination; Step 3: According to the first objective function and the second objective function, the population is non-dominated sorted, wherein the non-dominated sorting includes finding all non-dominated solutions and forming a first frontier, and continuing to find non-dominated solutions in the remaining solutions to form the next frontier, and repeating the iteration until all solutions are classified; Step 4: Based on all the classified solutions in step 3, calculate the normalized distance between adjacent solutions, where the total congestion distance of the solution is the sum of its distances on all objective functions; Step 5: Randomly select two individuals and compare their non-dominated sorting levels. If the levels are the same, select the individual with a larger crowding distance. Repeat this process until enough parent individuals are selected. Step 6: Merge the parent and offspring populations, perform non-dominated sorting and crowding distance calculation on the merged population, and select individuals with higher sorting and larger crowding distance to form a new generation population; Step 7: Check whether the maximum number of iterations or other termination conditions have been reached. If not, continue iterating. If reached, output the final set of non-dominated solutions and analyze the trade-offs between different solutions. Step 8. Based on the final set of non-dominated solutions, the Pareto solutions are processed through the analytical tools and models in the decision support system, and the final optimal price combination is selected from the Pareto frontier in combination with the actual operating conditions and decision maker preferences. The optimal price combination maximizes the expected return of the travel segment while satisfying the operating conditions.

[0049] It should be noted that, in this embodiment, it also includes: According to the price combination selection set, after comparative analysis, the expected benefits of each price combination at different price levels are evaluated, and after linear weighting method, a weighted benefit value is assigned to each price combination, and the weighted benefit value is calculated by assigning different weights to the expected benefits of each price level; It should be noted that we first need to evaluate the expected revenue of each price combination at different price levels. This process involves predicting the potential revenue of each price level. The prediction can be based on historical sales data, market research, passenger purchasing behavior analysis, etc. Evaluating the expected revenue of each price combination means that we need to estimate the demand and corresponding revenue of each price point. The linear weighted method is used to process these expected revenues. According to the importance weight of each indicator (here, the expected revenue of different price levels), the evaluation value of each indicator is weighted and summed to obtain the final comprehensive evaluation result. Specifically, it can be expressed as:

[0050] In the formula, S is the comprehensive evaluation result, that is, the weighted benefit value, , ,… is the weight of expected return at each price level, is the expected return of each price level. The method of determining weights can be subjective weighting method; it can also be objective weighting method, such as determining weights through mathematical analysis or statistical methods. After being processed by linear weighting method, the weighted return value obtained can be used in subsequent optimization algorithms as a basis for evaluating the pros and cons of different price combinations. At the same time, it reflects the comprehensive performance of different price combinations after considering their respective importance, providing decision makers with a quantitative decision support.

[0051] Understandably, in practical applications, this approach can help airlines find a balance between multiple price points to maximize revenue. For example, a high price point may bring higher single-trip revenue but lower demand; while a low price point may attract more passengers, thereby increasing overall revenue, although the single-trip revenue is lower. With this approach, airlines can formulate pricing strategies more scientifically to adapt to market changes and passenger demand.

[0052] According to the expected return and weighted return value, the non-dominated sorting and crowding distance of the NSGA-II algorithm are used to calculate and select a set of Pareto optimal solutions from multiple price combinations, which need to consider the objectives of maximizing return and minimizing cost. It should be noted that each price combination will be evaluated whether it dominates other combinations. If a price combination is not inferior to another combination in all objectives (such as maximizing revenue and minimizing costs), then it is considered non-dominated. Therefore, this process will produce multiple non-dominated layers, each of which contains a set of non-dominated price combinations. For all price combinations in the same layer, the crowding distance between the combinations is calculated, where the crowding distance is an indicator that measures the distribution density of price combinations in the target space. Specifically, for each price combination, its nearest neighbor combination on each target needs to be found and the distance between them is calculated. The cumulative value of these distances is the crowding distance, which reflects the degree of "crowding" around the price combination. Based on the non-dominated sorting and crowding distance calculation, price combinations are selected from each non-dominated layer. The selection is based on their non-dominated level (that is, the level they are in) and crowding distance. First, combinations with lower levels tend to be selected because these combinations are closer to the ideal Pareto frontier. If two combinations have the same rank, the combination with a larger crowding distance is selected to maintain the diversity of solutions, thereby obtaining a set of Pareto optimal solutions that maximize the expected benefits of the travel segment while satisfying the operating conditions, providing decision makers with a variety of options for weighing different objectives, and helping to formulate more comprehensive and effective pricing strategies.

[0053] The Pareto optimal solution is processed by the analytical tools and models in the decision support system, and the final optimal price combination is selected from the Pareto optimal solution in combination with the actual operating conditions and decision maker preferences. The optimal price combination maximizes the expected return of the travel segment while meeting the operating conditions.

[0054] It should be noted that after the analysis tools and model processing in the decision support system, combined with the actual operating conditions and decision maker preferences, the final optimal price combination is selected from the Pareto optimal solution. Among them, the decision support system may include data analysis, model building, prediction tools and visualization components. These tools help decision makers understand the characteristics of each solution and make choices based on business goals and constraints, thereby making the selection process more scientific and systematic, improving the accuracy and efficiency of decision-making. The optimal price combination finally selected can maximize the expected revenue of the travel segment while meeting the operating conditions, achieving the goal of revenue management.

[0055] According to the Pareto optimal solutions obtained from the NSGA-II algorithm, these solutions have theoretically met the goals of maximizing revenue and minimizing costs. Before practical application, the effectiveness of these solutions needs to be verified through actual data or simulation tests. The verification process may include A / B testing of selected price combinations, or implementing new price strategies for some flights within a specific time period, and collecting relevant sales data and customer feedback. After verification, if it is found that some price combinations fail to achieve the expected revenue effect, or customer feedback shows that the price strategy is not reasonable enough, it is necessary to adjust the fare system setting method. The adjustment may involve re-evaluating the parameters of the demand price function, or modifying certain parameters in the NSGA-II algorithm, including but not limited to the implementation methods of the selection operator, crossover operator, and mutation operator, to better adapt to market changes. Then, through continuous verification and adjustment, the ultimate goal is to find a fare system that can both maximize the revenue of airlines and be accepted by the market. In this process, multiple iterations are required, and each optimization is based on the previous results until a stable state is reached, in which the fare system can produce the expected revenue effect in actual operation. Therefore, the optimized fare system can better compete with competitors’ pricing strategies, improve the market competitiveness of airlines, and ensure that the fare system is more in line with market demand and passengers’ willingness to pay through actual verification and adjustment.

[0056] S400. Verify the setting results, establish a dynamic adjustment mechanism, continuously optimize price points based on real-time sales data and market feedback, and adjust the fare system setting method based on the verification results to ensure the final optimization effect of the fare system setting method.

[0057] Among them, select the maximum value ,in is the maximum expected return value among all price combinations, To adopt price combination Medium price, multiple prices coexist effectively, the expected total revenue of this travel segment, the corresponding price combination That is the best price combination selected.

[0058] In some embodiments of the present application, based on the aforementioned scheme, such as Figure 4 As shown, for example: on a certain date, a certain passenger vehicle has 500 seats and needs to implement three levels of ticket prices. There are a total of five price levels (p1=600 yuan, p2=700 yuan, p3=800 yuan, p4=900 yuan, p5=1000 yuan) to choose from. The demand price function is D=1000-0.7p. The following calculates the prices of the three selected levels that can be implemented.

[0059] Step 1: Demand price function is known: D = 1000-0.8p It can be obtained that if a single fare is adopted, the demand for each price level is: D1=520 D2=440 D3=360 D4=280 D5=200 Step 2: Select 3 fare discounts and list all possible price combinations: 1={600, 700, 800} 2={600, 700, 900} 3={600, 700, 1000} 4={600, 800, 900} 5={600, 800, 1000} 6={600, 900, 1000} 7={700, 800, 900} 8={700, 800, 1000} 9={700, 900, 1000} 10={800, 900, 1000} Step 3: Calculate the expected return of the flight under various combinations.

[0060] for 1={600, 700, 800}, without considering the location restrictions, according to the profit calculation formula , the total expected revenue is 80*600+80*700+360*800=392,000, and there are 520 seats required. The expected revenue of 500 seats is V1=380,000. 2={600, 700, 900}. Without considering the position restriction, the total expected revenue is: 80*600+160*700+280*900=412,000. A total of 520 seats are required. Therefore, the expected revenue of 500 positions is V2=400,000.

[0061] Therefore, by analogy, the expected returns of 500 positions of each price combination can be calculated, as shown in Table 1: Table 1 Expected returns of 500 positions for each price combination From the table above, we can see that this transportation tool needs to choose a combination from 5 price levels. 5={600, 800, 1000}, as the implemented fare level, can maximize the expected return value.

[0062] Embodiment 2: like Figure 2 As shown, this embodiment provides a fare system setting system based on multi-objective optimization and dynamic adjustment, see Figure 2 The system comprises: Collection module 701: used to collect order data of all seats departing from one location to another location within the travel interval, construct a demand price function through regression analysis, and establish a cost function through cost data analysis; Establishing module 702: used for establishing a set of optional price points within the range based on the cost function, taking the cost coverage price as the lower limit, increasing the price point density in the key area, and filling in the additional price points using the interpolation method to ensure that the entire price range is covered, and obtaining a set of potential execution prices, wherein the parameters collected in the set of optional price points include the mileage from the starting point to the arrival point, the transportation time, the vehicle operation cost data, the passenger flow data and the time value data, the route construction cost, the vehicle purchase price, the depreciation time, the electricity cost and the number of seats; Calculation module 703: used to calculate the transportation revenue under the price combination according to the potential execution price set, using traversal calculation and fast elite multi-objective genetic algorithm and using the revenue calculation model to obtain the expected maximum revenue, taking the expected maximum revenue as the objective function, introducing the seat inventory constraint corresponding to each price point in the combination, searching and screening out the potential execution price combination set on the Pareto frontier, obtaining the optimal price combination, and then using the optimal price combination as the setting result of the fare system, wherein the revenue calculation needs to consider the price, demand, seat inventory, discount and additional fees; Adjustment module 704: used to verify the setting results, establish a dynamic adjustment mechanism, continuously optimize price points based on real-time sales data and market feedback, and adjust the fare system setting method based on the verification results to ensure the final optimization effect of the fare system setting method.

[0063] Specifically, the collection module 701 includes: Identification and marking unit: used to collect the order data of all seats departing from one location to another within the travel interval, identify and record all different price levels within the time period, and mark them in order from low to high prices to obtain the order quantity of each price level; The first processing unit is used to obtain first demand data and second demand data based on the order quantity of each price level through demand analysis processing, wherein the first demand data is the order data corresponding to the price level, and its demand is the sum of the demand of orders at the price; wherein the second demand data is the order data corresponding to the price level not less than the price level, and its demand is the sum of the demand of orders from the price level to the highest price level; The second processing unit is used to fit the relationship between price and demand using a regression analysis method according to the second demand data, and then obtain a demand price function of the impact of price changes on passenger demand when a fixed fare is implemented.

[0064] Specifically, the establishment module 702 includes: Construction unit: used to use multiple regression to analyze historical sales data and market information to construct a demand price function; based on the demand price function, using the Monte Carlo simulation method, taking into account market volatility factors, calculate the risk-adjusted cost coverage price as the lowest acceptable price; using the linear programming method, taking into account market capacity and competition factors, determine the highest feasible price; and establish a continuous price range between the lowest acceptable price and the highest feasible price; Allocation unit: used to set corresponding price points for major customer groups based on continuous price ranges, set corresponding price points for significant seasonal fluctuations based on seasonal analysis, and allocate additional price points around the main competitors' prices taking into account competition; Determination unit: used to determine the set of optional price points by filling the remaining price points using an interpolation method based on the additional price points according to the given number of price points.

[0065] Specifically, the calculation module 703 includes: Demand calculation unit: used to calculate the passenger demand at each price level when considering a single fare. The calculation formula is as follows:

[0066] In the formula, is the price, D is the number of passengers required, For price is the demand function expression of the variable; Sorting unit: It is used to consider the coexistence of multiple prices in the price combination based on passenger demand, and calculate the demand forecast value of each price level from high to low according to the principle of downward compatibility of demand; Revenue calculation unit: It is used to calculate the expected revenue when multiple prices are used in parallel based on the demand forecast value and the corresponding ticket price. The calculation formula is as follows:

[0067] In the formula, is the expected maximum revenue of the travel segment, k is the price combination adopted The lowest price level is sorted, and s is the price combination The number of prices in the equation, i is the price combination Price sorting number, Price combination The sorting in is the price level of i, for Demand corresponding to the price level; Comparison unit: used to compare the expected benefits under different price combinations, and record the price combination that maximizes the benefit as the expected maximum benefit of the travel segment.

[0068] It should be noted that, regarding the system in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0069] Embodiment 3: Corresponding to the above method embodiment, this embodiment also provides an optimization setting device for the fare system in revenue management. The optimization setting device for the fare system in revenue management described below and the fare system setting method based on multi-objective optimization and dynamic adjustment described above can refer to each other.

[0070] Figure 3 FIG. 8 is a block diagram of a device 800 for optimizing and setting a fare system in revenue management according to an exemplary embodiment. Figure 3 As shown, the device 800 for optimizing the fare system in revenue management includes: a processor 801 and a memory 802. The device 800 for optimizing the fare system in revenue management also includes one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.

[0071] The processor 801 is used to control the overall operation of the device 800 for optimizing the fare system in revenue management to complete all or part of the steps in the above-mentioned method for setting the fare system based on multi-objective optimization and dynamic adjustment. The memory 802 is used to store various types of data to support the operation of the device 800 for optimizing the fare system in revenue management, and these data may include, for example, instructions for any application or method operating on the device 800 for optimizing the fare system in revenue management, and application-related data, such as contact data, sent and received messages, pictures, audio, video, etc. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk. The multimedia component 803 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may be further stored in the memory 802 or sent via the communication component 805. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules, and the above-mentioned other interface modules can be a keyboard, a mouse or buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 805 is used for wired or wireless communication between the optimization setting device 800 of the fare system in the revenue management and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 805 can include: a Wi-Fi module, a Bluetooth module or an NFC module.

[0072] In an exemplary embodiment, the optimization setting device 800 for the fare system in revenue management can be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned fare system setting method based on multi-objective optimization and dynamic adjustment.

[0073] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned method for setting a fare system based on multi-objective optimization and dynamic adjustment are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 802 including program instructions, and the above-mentioned program instructions can be executed by the processor 801 of the device 800 for optimizing and setting a fare system in revenue management to complete the above-mentioned method for setting a fare system based on multi-objective optimization and dynamic adjustment.

[0074] Embodiment 4: Corresponding to the above method embodiment, a readable storage medium is also provided in this embodiment. The readable storage medium described below and the fare system setting method based on multi-objective optimization and dynamic adjustment described above can correspond to each other.

[0075] A computer program is stored on the readable storage medium, and when the computer program is executed by the processor, the steps of the fare system setting method based on multi-objective optimization and dynamic adjustment of the above method embodiment are implemented.

[0076] The readable storage medium may specifically be a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or other readable storage medium that can store program codes.

[0077] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0078] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A fare system setting method based on multi-objective optimization and dynamic adjustment, characterized in that: include: Collect the order data of all seats departing from one location to another within the travel range, construct the demand price function through regression analysis, and establish the cost function through cost data analysis; Based on the cost function, the cost coverage price is taken as the lower limit, and a set of optional price points is established within this range. The price point density is increased in key areas, and additional price points are filled using interpolation to ensure that the entire price range is covered, thereby obtaining a set of potential execution prices. The parameters collected in the optional price point set include the mileage from the starting point to the arrival point, transportation time, vehicle operation cost data, passenger flow data, and time value data. According to the potential execution price set, the traversal calculation and fast elite multi-objective genetic algorithm are used to calculate the transportation revenue under the price combination using the revenue calculation model to obtain the expected maximum revenue. The expected maximum revenue is used as the objective function, and the seat inventory constraint corresponding to each price point in the combination is introduced to search and screen the potential execution price combination set on the Pareto frontier to obtain the optimal price combination, and then the optimal price combination is used as the setting result of the fare system. The revenue calculation needs to take into account the price, demand, seat inventory, discount and additional fees. Verify the setting results, establish a dynamic adjustment mechanism, continuously optimize price points based on real-time sales data and market feedback, and adjust the fare system setting method based on the verification results to ensure the ultimate optimization effect of the fare system setting method.

2. The fare system setting method based on multi-objective optimization and dynamic adjustment according to claim 1 is characterized in that: The order data of all seats departing from one location to another within the travel interval are collected, and a demand price function is constructed through regression analysis, including: Collect the order data of all seats departing from one location to another within the travel interval, identify and record all different price levels within the time period, and mark them in order from low to high in price to obtain the number of orders for each price level; Based on the order quantity of each price level, after demand analysis and processing, the first demand data and the second demand data are obtained, wherein the first demand data is the order data corresponding to the price level, and its demand is the sum of the demand of the orders at the price level; wherein the second demand data is the order data corresponding to the price level not less than that, and its demand is the sum of the demand of the orders from the price level to the highest price level; According to the second demand data, the regression analysis method is used to fit the relationship between price and demand, and then the demand price function of the impact of price changes on passenger demand when a fixed fare is implemented is obtained.

3. The fare system setting method based on multi-objective optimization and dynamic adjustment according to claim 1 is characterized in that: Based on the cost function, the cost coverage price is taken as the lower limit, a set of optional price points is established within this range, the price point density is increased in key areas, and extra price points are filled using interpolation to ensure that the entire price range is covered, and a set of potential execution prices is obtained, including: Use multiple regression to analyze historical sales data and market information to construct a demand price function; Based on the demand price function, use the Monte Carlo simulation method to consider market volatility factors and calculate the risk-adjusted cost coverage price as the lowest acceptable price; Use the linear programming method to consider market capacity and competition factors to determine the highest feasible price; Establish a continuous price range between the lowest acceptable price and the highest feasible price; Set corresponding price points for major customer groups based on continuous price ranges, set corresponding price points for significant seasonal fluctuations based on seasonal analysis, and allocate additional price points around the main competitors' prices taking into account competition; Based on the additional price points, the optional price point set is determined by filling the remaining price points using an interpolation method according to the given number of price points.

4. The fare system setting method based on multi-objective optimization and dynamic adjustment according to claim 3 is characterized in that: Based on the determined optional price point set, a certain number of prices are screened out as potential execution prices, the number of all possible price combinations is calculated, and a price combination selection set is obtained. The combination set quantity calculation formula is as follows: In the formula, The number of combinations of selecting s elements from m independent elements, m ! represents the factorial of m, which is the product of all integers from 1 to m. s ! represents the factorial of s, that is, the product of all integers from 1 to s, m is the total number of elements to be selected, and s is the number of elements to be selected.

5. The fare system setting method based on multi-objective optimization and dynamic adjustment according to claim 1 is characterized in that: The ergodic calculation and fast elite multi-objective genetic algorithm are used to calculate the transportation revenue under the price combination using the revenue calculation model to obtain the expected maximum revenue, which includes: When considering a single fare, the passenger demand for each price level is calculated as follows: In the formula, is the price, D is the number of passengers required, For price is the demand function expression of the variable; Based on passenger demand and considering the coexistence of multiple prices in the price combination, the demand forecast value of each price level is calculated from high to low according to the principle of downward compatibility of demand; According to the demand forecast value and the corresponding ticket price, after the revenue calculation, the expected revenue when multiple prices are used in parallel is obtained. The calculation formula is as follows: In the formula, is the expected maximum revenue of the travel segment, k is the price combination adopted The lowest price level is sorted, and s is the price combination The number of prices in the equation, i is the price combination Price sorting number, Price combination The sorting in is the price level of i, for Demand corresponding to the price level; Compare the expected revenues under different price combinations, and record the price combination that maximizes the revenue as the expected maximum revenue for the travel segment.

6. The method for setting a fare system based on multi-objective optimization and dynamic adjustment according to claim 5, characterized in that: The calculation formula of the expected maximum return is as follows: In the formula, is the expected maximum revenue of the travel segment, k To adopt price combination Sort by lowest price tier, s Price combination The price quantity in i Price combination Price sorting number, Price combination The sorting in is i Price level, for The demand corresponding to the price level, To adopt price combination The lowest price among all the options, N is the seat stock of the transportation vehicle.

7. The fare system setting method based on multi-objective optimization and dynamic adjustment according to claim 1 is characterized in that: The traversal calculation and fast elite multi-objective genetic algorithm are used to calculate the transportation revenue under the price combination using the revenue calculation model to obtain the expected maximum revenue. The expected maximum revenue is used as the objective function, and the seat inventory constraint corresponding to each price point in the combination is introduced to search and screen out the potential execution price combination set on the Pareto frontier to obtain the optimal price combination, wherein the calculation steps are as follows: Step 1: Consider different potential price combinations according to the price point set, and use different potential price combinations as encoding schemes for different individual designs. Generate a set of random initial solutions based on the initial population of the fast elite multi-objective genetic algorithm, where each initial solution has a corresponding selection model configuration to ensure that the initial population covers all parts of the solution space. Step 2: Based on the initial solution, evaluate the fitness of each individual, and use the selection model to calculate the passenger selection probability, and calculate the first objective function and the second objective function. The first objective function is to maximize the expected return, and the second objective function is to minimize the number of prices in the price combination; Step 3: According to the first objective function and the second objective function, the population is non-dominated sorted, wherein the non-dominated sorting includes finding all non-dominated solutions and forming a first frontier, and continuing to find non-dominated solutions in the remaining solutions to form the next frontier, and repeating the iteration until all solutions are classified; Step 4: Based on all the classified solutions in step 3, calculate the normalized distance between adjacent solutions, where the total congestion distance of the solution is the sum of its distances on all objective functions; Step 5: Randomly select two individuals and compare their non-dominated sorting levels. If the levels are the same, select the individual with a larger crowding distance. Repeat this process until enough parent individuals are selected. Step 6: Merge the parent and offspring populations, perform non-dominated sorting and crowding distance calculation on the merged population, and select individuals with higher sorting and larger crowding distance to form a new generation population; Step 7: Check whether the maximum number of iterations or other termination conditions have been reached. If not, continue iterating. If reached, output the final set of non-dominated solutions and analyze the trade-offs between different solutions. Step 8. Based on the final set of non-dominated solutions, the Pareto solutions are processed through the analytical tools and models in the decision support system, and the final optimal price combination is selected from the Pareto frontier in combination with the actual operating conditions and decision maker preferences. The optimal price combination maximizes the expected return of the travel segment while satisfying the operating conditions.

8. A fare system setting system based on multi-objective optimization and dynamic adjustment, based on the fare system setting method based on multi-objective optimization and dynamic adjustment according to claim 1, characterized in that: include: Collection module: used to collect the order data of all seats departing from one location to another within the travel interval, build the demand price function through regression analysis, and establish the cost function through cost data analysis; Establishment module: Based on the cost function, the cost coverage price is used as the lower limit, and a set of optional price points is established within this range. The density of price points is increased in key areas, and additional price points are filled using interpolation to ensure that the entire price range is covered, so as to obtain a set of potential execution prices. The parameters collected in the optional price point set include the mileage from the starting point to the arrival point, the transportation time, the vehicle operation cost data, the passenger flow data, and the time value data; Calculation module: It is used to calculate the transportation revenue under the price combination based on the potential execution price set, using traversal calculation and fast elite multi-objective genetic algorithm to apply the revenue calculation model, and obtain the expected maximum revenue. The expected maximum revenue is used as the objective function, and the seat inventory constraint corresponding to each price point in the combination is introduced to search and screen out the potential execution price combination set on the Pareto frontier to obtain the optimal price combination, and then use the optimal price combination as the setting result of the fare system. The revenue calculation needs to consider the price, demand, seat inventory, discount and additional fees; Adjustment module: used to verify the setting results, establish a dynamic adjustment mechanism, continuously optimize price points based on real-time sales data and market feedback, and adjust the fare system setting method based on the verification results to ensure the final optimization effect of the fare system setting method.

9. The fare system setting system based on multi-objective optimization and dynamic adjustment according to claim 8 is characterized in that: The collection module includes: Identification and marking unit: used to collect the order data of all seats departing from one location to another within the travel interval, identify and record all different price levels within the time period, and mark them in order from low to high prices to obtain the order quantity of each price level; The first processing unit is used to obtain first demand data and second demand data based on the order quantity of each price level through demand analysis processing, wherein the first demand data is the order data corresponding to the price level, and its demand is the sum of the demand of orders at the price; wherein the second demand data is the order data corresponding to the price level not less than the price level, and its demand is the sum of the demand of orders from the price level to the highest price level; The second processing unit is used to fit the relationship between price and demand using a regression analysis method according to the second demand data, and then obtain a demand price function of the impact of price changes on passenger demand when a fixed fare is implemented.

10. The fare system setting system based on multi-objective optimization and dynamic adjustment according to claim 8, characterized in that: The establishment module includes: Construction unit: used to use multiple regression to analyze historical sales data and market information to construct a demand price function; based on the demand price function, using the Monte Carlo simulation method, taking into account market volatility factors, calculate the risk-adjusted cost coverage price as the lowest acceptable price; using the linear programming method, taking into account market capacity and competition factors, determine the highest feasible price; and establish a continuous price range between the lowest acceptable price and the highest feasible price; Allocation unit: used to set corresponding price points for major customer groups based on continuous price ranges, set corresponding price points for significant seasonal fluctuations based on seasonal analysis, and allocate additional price points around the main competitors' prices taking into account competition; Determination unit: used to determine the set of optional price points by filling the remaining price points using an interpolation method based on the additional price points according to the given number of price points.

Images