Railway data pricing method and system based on multi-dimensional value evaluation

Abstract

The invention discloses a railway data pricing method and system based on multi-dimensional value evaluation. The method comprises the following steps: constructing a railway data value comprehensive evaluation index system considering internal attributes and external attributes of data; determining the normalized weight of each index by adopting an analytic hierarchy process; quantifying a comprehensive utility value of the internal attribute of the data by utilizing a utility function, and quantifying a comprehensive score of the external attribute of the data by utilizing a fuzzy comprehensive evaluation method; and based on the data internal attribute comprehensive utility value and the data external attribute comprehensive score, considering the willingness to pay of the demand side, constructing a willingness to pay of the demand side, establishing a profit function based on the willingness to pay of the demand side, and solving by taking profit maximization as a target to obtain an optimal pricing result. According to the railway data pricing method, the internal attributes and the external attributes of the data are considered, the effectiveness and the willingness to pay of the data demander in the transaction scene are considered, the scientificity of railway data value quantification and the optimization of the pricing strategy are realized, and feasible technical support is provided for railway data transaction.

Description

A railway data pricing method and system based on multidimensional value assessment Technical Field

[0001] This invention relates to the field of railway technology, and specifically to a railway data pricing method and system based on multidimensional value assessment. Background Technology

[0002] With the development of the digital economy, the railway industry has accumulated a vast amount of data in its operations, scheduling, equipment monitoring, and passenger transport organization. This data is diverse in type, massive in scale, and frequently updated, containing enormous potential value. However, to fully realize the value of railway data, the core lies in promoting its efficient circulation and sharing. Data trading, as the core channel for data circulation, is crucial for unlocking the value of railway data and achieving efficient data circulation and sharing, and its reasonable pricing is key.

[0003] Existing data pricing methods are mainly applied to general data trading markets, and research is typically based on traditional economic methods such as the cost approach, revenue approach, and market approach. These methods have clear structures and are easy to implement. Subsequently, with the development of computers, pricing methods based on model utility, privacy loss, and query costs have gradually emerged, which can characterize the actual value contribution of data from multiple perspectives.

[0004] However, in the railway industry, due to the high security, high business sensitivity, and strong scenario dependence of data, general data pricing methods are often difficult to apply directly. Research on railway scenarios typically employs the analytic hierarchy process (AHP) to construct a value evaluation index system for railway data, comprehensively assessing the data from multiple dimensions to reflect industry characteristics and the multidimensional value attributes of the data.

[0005] While previous research has yielded promising results in pricing general data and railway data, the following issues remain: First, general data pricing methods are difficult to directly apply to railway scenarios, as their assumptions differ significantly from the industry attributes of railway data. Traditional methods often build models based on open data markets, neglecting the high security and strong scenario dependence of railway data, thus failing to capture its true value. Second, existing railway data pricing methods largely remain at the level of static value evaluation, lacking pricing mechanisms applicable to actual transactions. They fail to fully reflect the multidimensional value characteristics of railway data, lacking systematic modeling of the quantitative relationship between the internal attributes of railway data (quality, scale, timeliness, etc.) and actual business scenarios, and do not consider the utility and willingness to pay of data demanders in actual transaction scenarios.

[0006] Current railway data pricing methods focus on cost and struggle to directly characterize the impact of internal railway data attributes on actual utility. Furthermore, these methods fail to integrate external attributes such as market demand and scenario constraints into the pricing process, making it difficult to guide pricing in actual transaction scenarios. Therefore, this invention proposes a railway data pricing method that considers both internal and external attributes of the data, combining internal data utility modeling with external market factor analysis, and integrating demand-side utility and willingness to pay. Summary of the Invention

[0007] This invention provides a railway data pricing method and system based on multidimensional value assessment, in order to solve the problems of existing railway data pricing methods failing to systematically combine internal and external attributes of data, making it difficult to accurately quantify the multidimensional value of data, and lacking a pricing optimization mechanism.

[0008] According to the first aspect, one embodiment provides a railway data pricing method based on multidimensional value assessment. The method includes: constructing a comprehensive evaluation index system for railway data value that takes into account both internal and external attributes of the data, including an objective layer, a criterion layer, and an indicator layer; using the analytic hierarchy process (AHP) to determine the normalized weights of each indicator in the comprehensive evaluation index system and performing a consistency check; based on the weights of each indicator, using a utility function to quantify the comprehensive utility value of the internal attributes of the data, and using a fuzzy comprehensive evaluation method to quantify the comprehensive score of the external attributes of the data; based on the comprehensive utility value of the internal attributes of the data and the comprehensive score of the external attributes of the data, considering the willingness of the demand side to pay, constructing a willingness-to-pay model of the demand side, establishing a profit function based on the willingness-to-pay model of the demand side to pay, and solving for the optimal pricing result with profit maximization as the objective.

[0009] Furthermore, a comprehensive evaluation index system for railway data value is constructed, taking into account both internal and external attributes of the data. This system includes an objective layer, a criterion layer, and an indicator layer. Specifically, the objective layer represents the comprehensive value of railway data assets; the criterion layer includes internal and external attributes of the data. Internal attributes include sub-criterion layers for data quality and data scale, while external attributes include sub-criterion layers for data market attributes and data security. In the indicator layer, data quality indicators include accuracy, completeness, consistency, and timeliness; data scale indicators include data volume, growth rate, and update rate; data market attribute indicators include scarcity, number of demanders, and policy impact; and data security indicators include privacy protection level, compliance, and data access control.

[0010] Furthermore, the analytic hierarchy process (AHP) is used to determine the normalized weights of each indicator in the comprehensive evaluation index system, and a consistency test is performed. Specifically, this includes: establishing a hierarchical structure model: clarifying the hierarchical relationships between the target layer, the criterion layer, and the indicator layer; constructing pairwise comparison judgment matrices: using a 1-9 scale to compare the pairwise importance of indicators at the same level to form judgment matrices; and calculating and normalizing eigenvectors: for each judgment matrix, calculating the largest eigenvalue. The corresponding feature vectors are normalized to obtain the initial weights of each indicator; a consistency test is performed: the consistency test is judged by the consistency ratio CR. When CR < 0.1, the weights are valid; otherwise, the judgment matrix is ​​adjusted and recalculated; finally, the normalized weights of each indicator are obtained.

[0011] Furthermore, the utility function is used to quantify the comprehensive utility value of the internal attributes of the data. Specifically, this includes: constructing 10 rating levels for each internal attribute index, with a range of 0.1-1 and a step size of 0.1, to simulate different attribute levels; generating multiple datasets with different attribute levels based on the 10 rating levels of each internal attribute index; predicting the classification accuracy of each dataset using the XGBoost model on the downstream task of the demand side, and using the classification accuracy predicted by the XGBoost model as the actual utility benchmark of the internal attributes of the data; selecting the Gaussian cumulative distribution function as the basic utility function model, and aiming to minimize the sum of squared errors between the fitted value of the utility function and the XGBoost classification accuracy representing the actual utility, solving for the optimal parameters using the least squares method to obtain the optimal utility function model with optimal parameters.

[0012] Furthermore, the utility function is used to quantify the comprehensive utility value of the internal attributes of the data. Specifically, this includes: converting the actual values ​​of each internal attribute indicator into rating scales and then inputting them into the optimal utility function model to calculate the utility value of each internal attribute indicator; and then weighting and summing the utility values ​​of each internal attribute indicator with their respective weights to obtain the comprehensive utility value of the internal attributes of the data. ,in This refers to the number of internal attribute indicators. For the first The weight of each internal attribute indicator, For the first The utility value of each internal attribute indicator.

[0013] Furthermore, the fuzzy comprehensive evaluation method is used to quantify the comprehensive score of the external attributes of the data. Specifically, this includes: first, determining the evaluation set: establishing an indicator set D containing all external attribute indicators of the data, establishing an evaluation level set V including n evaluation levels, and assigning a corresponding quantitative score P to each evaluation level; second, constructing a membership matrix R: for each external attribute indicator of the data, inviting railway experts to conduct a level evaluation, and determining the membership degree of each indicator to each evaluation level. First, an m×n membership matrix R is formed, where m is the number of external attribute indicators and n is the evaluation level. Second, based on the weight vector W of each external attribute indicator determined by the analytic hierarchy process (AHP), weighted fuzzy operations are used to combine the weight vector W with the membership matrix R to obtain the fuzzy comprehensive evaluation result vector. Finally, according to Calculate the comprehensive score of the external attributes of the data. , .

[0014] Furthermore, based on the comprehensive utility value of the data's internal attributes and the comprehensive score of the data's external attributes, and considering the demand side's willingness to pay, a demand side payment willingness model is constructed, specifically including: defining the payment willingness model as:

[0015] in, The highest willingness to pay by the demand side. The comprehensive utility value of the data's internal attributes. This is the willingness-to-pay coefficient. This reflects the overall market purchasing tendency or price sensitivity, making , This is a comprehensive score based on the external attributes of the data.

[0016] Furthermore, a profit function is established based on the demand-side willingness-to-pay model, and the optimal pricing result is obtained by solving the problem with profit maximization as the objective. Specifically, this includes assuming that there exists a profit function in the market. One potential demander, and the actual willingness of the demander to pay. Constrained by the probability distribution function f(w); if the average cost of data assets is c and the price of data assets is p, then the profit function π corresponding to the data assets is:

[0017] Assuming the actual willingness of the demand side to pay follows If the distribution is uniform, then the profit function π can be expressed as:

[0018]

[0019] F(p) represents the cumulative distribution function when the demander's willingness to pay w ≤ p, i.e., the probability that the demander is willing to buy data at price p or lower; however, when the actual price is higher than the demander's willingness to pay, the demander cannot obtain positive utility and will not make a purchase decision. Only when the actual price is higher than the demander's willingness to pay will the demander not make a purchase decision. Only when p > w is the demand side likely to accept the transaction. ,but:

[0020] The profit function can be expressed as a nonlinear programming problem as follows:

[0021]

[0022]

[0023]

[0024] The objective is to maximize the profit function, with the following constraints: ; For the profit function Take the partial derivative of the data pricing p, and set the partial derivative to 0, that is:

[0025] Get the optimal price :

[0026] That is, the optimal price is positively correlated with the combined utility value of the data's internal attributes and the combined score of the data's external attributes.

[0027] According to the second aspect, one embodiment provides a railway data pricing system based on multidimensional value assessment. The system includes: an evaluation index system construction module, used to construct a comprehensive evaluation index system for railway data value that takes into account both internal and external attributes of the data, including an objective layer, a criterion layer, and an index layer; an index weight determination module, used to determine the normalized weights of each index in the comprehensive evaluation index system using the analytic hierarchy process (AHP) and perform consistency checks; an internal and external attribute evaluation module, used to quantify the comprehensive utility value of the internal attributes of the data using a utility function based on the weights of each index and to quantify the comprehensive score of the external attributes of the data using a fuzzy comprehensive evaluation method; and a pricing solution module, used to construct a demand-side payment willingness model based on the comprehensive utility value of the internal attributes of the data and the comprehensive score of the external attributes of the data, considering the demand-side payment willingness, establish a profit function based on the demand-side payment willingness model, and solve for the optimal pricing result with profit maximization as the objective.

[0028] According to three aspects, one embodiment provides an electronic device, the device comprising: a processor and a memory; the memory for storing one or more program instructions; the processor for executing one or more program instructions to perform steps of a railway data pricing method based on multidimensional value assessment as described in any of the preceding embodiments.

[0029] According to a fourth aspect, one embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of a railway data pricing method based on multidimensional value assessment as described in any of the preceding claims.

[0030] This invention provides a railway data pricing method and system based on multidimensional value assessment, which has the following beneficial effects: 1) This invention realizes unified quantitative modeling of internal and external attributes of railway data, which makes up for the shortcomings of existing methods that only value from the dimensions of cost, resulting in unreasonable prices and failure to adapt to market changes, and makes railway data pricing more comprehensive and reliable.

[0031] 2) This invention improves the scientificity and interpretability of the pricing model by constructing an internal utility function based on a quality perturbation experiment, thus avoiding bias caused by subjective expert assignment.

[0032] 3) This invention introduces a fuzzy comprehensive evaluation model to characterize the external market environment and uses external attributes as payment willingness coefficients, so that the final pricing can dynamically reflect external conditions such as market demand, and realize differentiated pricing of railway data in different scenarios.

[0033] 4) This invention constructs a profit-maximizing pricing model, enabling the optimal price to be calculated quickly and deployed systematically; the pricing process is transparent and the results are traceable, improving the reliability of railway data pricing. Attached Figure Description

[0034] Figure 1 is a flowchart of a railway data pricing method based on multidimensional value assessment according to an embodiment of the present invention; Figure 2 is a schematic diagram of the comprehensive evaluation index system of railway data value in a railway data pricing method based on multidimensional value assessment according to an embodiment of the present invention; Figure 3 is a schematic diagram of profit changing with pricing in a railway data pricing method based on multidimensional value assessment according to an embodiment of the present invention; Figure 4 is a schematic diagram of pricing and profit changing with the internal attribute utility of data in a railway data pricing method based on multidimensional value assessment according to an embodiment of the present invention. Detailed Implementation

[0035] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of the invention. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, certain operations related to the present invention are not shown or described in the specification. This is to avoid obscuring the core parts of the invention with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0036] Furthermore, the features, operations, or characteristics described in the specification can be combined in any suitable manner to form various embodiments. At the same time, the steps or actions in the method description can be rearranged or adjusted in a manner obvious to those skilled in the art. Therefore, the various orders in the specification and drawings are only for the clear description of a particular embodiment and do not imply a necessary order, unless otherwise stated that a particular order must be followed.

[0037] The first embodiment of this invention provides a railway data pricing method based on multidimensional value assessment, demonstrating how to complete the process of constructing an indicator system, quantifying the utility of internal attributes, scoring external attributes, modeling willingness to pay, and solving for optimal pricing in actual railway operations, according to the method of this invention. A detailed description is provided below with reference to Figure 1.

[0038] As shown in Figure 1, in step S100, a comprehensive evaluation index system for railway data value that takes into account both internal and external attributes of the data is constructed, including the target layer, the criterion layer, and the index layer.

[0039] In this embodiment, the main factors affecting the value of railway data are divided into two categories: internal factors and external factors. Internal factors determine the intrinsic use value of the data and affect its utility for specific downstream tasks of the demand side. External factors determine the marketability of the data and affect its tradability.

[0040] Specifically, a comprehensive evaluation index system for railway data value is determined; a three-tiered evaluation index system is constructed, consisting of an objective layer, a criterion layer, and an index layer, comprehensively covering both internal and external attributes of the data. As shown in Figure 2, the specific index system structure is as follows: Objective layer: Comprehensive value of railway data assets.

[0041] Criteria layer: This includes internal data attributes and external data attributes. Internal data attributes determine the value of data in use, while external data attributes determine the value of data in transactions.

[0042] The criteria layer comprises two sub-criteria layers: data quality and data scale, under the internal attributes of data. Data quality indicators include accuracy, completeness, consistency, and timeliness, while data scale indicators include data volume, growth rate, and update rate. The external attributes of data comprise two sub-criteria layers: data market attributes and data security, under the external attributes of data. Data market attribute indicators include scarcity, number of demanders, and policy impact, while data security indicators include privacy protection level, compliance, and data access control.

[0043] As shown in Figure 1, in step S200, the normalized weights of each indicator in the comprehensive evaluation index system are determined by the analytic hierarchy process and a consistency check is performed.

[0044] Specifically, the weights of the indicators are determined based on the Analytic Hierarchy Process (AHP); the relative importance of each indicator in the evaluation system is also determined using AHP. For indicators at any level, pairwise comparisons are made to obtain scores for different levels of importance, and a judgment matrix is ​​constructed accordingly. For each judgment matrix, the eigenvector corresponding to the largest eigenvalue is calculated and normalized to obtain the weights of each indicator.

[0045] S210, Establish a hierarchical structure model: Clarify the hierarchical relationship between the target layer (comprehensive value of railway data assets), the criteria layer (internal attributes of data / external attributes of data), and the indicator layer (the specific indicators mentioned above).

[0046] S220, Constructing a pairwise comparison judgment matrix: Several railway experts were invited to use a 1-9 scale method to compare the importance of indicators at the same level pairwise, forming a judgment matrix. .

[0047] S230, Calculate and normalize the eigenvectors: for each judgment matrix Calculate the largest eigenvalue Corresponding feature vector :

[0048] in, Indicates the first The weight value of each indicator reflects its relative importance in the entire evaluation system; After normalization, the weight vector is obtained:

[0049] This weight vector represents the relative importance of the indicators.

[0050] S240, Consistency Test: Calculate the Consistency Index (CI).

[0051] in, It is the largest eigenvalue of the matrix. It determines the order of a matrix.

[0052] Then calculate the consistency ratio (CR):

[0053] RI is the random consistency index, whose value is related to the matrix order n and is obtained by looking up a table.

[0054] When CR < 0.1, the judgment matrix passes the consistency test and the weights are valid; if CR ≥ 0.1, then the value of the judgment matrix needs to be readjusted and recalculated.

[0055] S250, Weight Summary: The final normalized weights of all indicators are obtained, and the obtained weights will be used for subsequent internal utility calculations and comprehensive scoring of external attributes.

[0056] As shown in Figure 1, in step S300, based on the weights of each indicator, the comprehensive utility value of the internal attributes of the data is quantified using a utility function, and the comprehensive score of the external attributes of the data is quantified using a fuzzy comprehensive evaluation method.

[0057] The above steps specifically include: S310, quantifying the comprehensive utility value of the internal attributes of the data based on the utility function.

[0058] The internal attributes of railway data reflect the quality and scale characteristics of the data itself, and are the core factors determining the value of the data. Quality indicators such as accuracy, completeness, consistency, and timeliness, as well as scale indicators such as data volume, growth rate, and update rate, jointly determine the data's ability to train and analyze models. To quantitatively express the contribution of these internal attributes to the value of the data, this embodiment introduces a utility function modeling method to describe the relationship between data value and the specific values ​​of indicators from an economic perspective. In economics, utility functions are used to characterize consumers' preferences for commodity attributes. When the quality of a commodity increases, the utility gained by consumers increases, but the magnitude of the increase decreases; this law is called diminishing marginal utility. Railway data also exhibits similar characteristics. This embodiment focuses on the contribution of data to downstream tasks of the demand side, quantifying the comprehensive utility value of the data's internal attributes through utility functions. The specific steps are as follows: S311, for each data internal attribute indicator, construct 10 rating levels in the range of 0.1-1 with a step size of 0.1 to simulate different attribute levels. Table 1 shows the simulation methods for different indicators.

[0059] Table 1. Definitions, Simulation Methods, and Calculation Formulas for Internal Data Attribute Indicators

[0060] S312 generates multiple datasets with different attribute levels by combining 10 rating levels of the internal attribute indicators of each data point.

[0061] In this embodiment, 70 datasets with different attribute levels are generated by combining the 10 rating levels of each internal attribute indicator.

[0062] S313, in the downstream task on the demand side, the classification accuracy of each dataset is predicted using the XGBoost model, with the category labels representing railway business labels (such as train operation status, passenger flow category, etc.). The classification accuracy predicted by the XGBoost model is used as the actual utility benchmark of the data's internal attributes.

[0063] S314. The Gaussian cumulative distribution function is selected as the basic utility function model. The goal is to minimize the sum of squared errors between the fitted value of the utility function and the XGBoost classification accuracy representing the actual utility. The optimal parameters are solved by the least squares method to obtain the optimal utility function model with the optimal parameters.

[0064] The Gaussian cumulative distribution function has the following form:

[0065] in Let be the cumulative distribution function of the standard normal distribution. , , , These are parameters to be determined.

[0066] The optimal parameters are obtained by minimizing the sum of squared errors using the least squares method, as shown in the formula:

[0067] Here For parameter vectors m is the sample size. It is the predicted classification accuracy, used to simulate the actual utility of the demand side. It is the result of fitting the utility function.

[0068] S315, the utility value of each data internal attribute indicator is calculated by converting the actual values ​​of each data internal attribute indicator into a grade score and then substituting it into the optimal utility function model.

[0069] S316, the comprehensive utility value of the internal attributes of the data is obtained by weighted summation of the utility values ​​of each internal attribute indicator and the weight of each indicator:

[0070] in This refers to the number of internal attribute indicators. For the first The weight of each internal attribute indicator, For the first The utility values ​​of each internal attribute indicator are calculated. Finally, the combined utility value of the internal attributes, ranging from 0 to 1, is obtained and used in subsequent pricing models.

[0071] S320 is a quantitative comprehensive score of external attributes of data based on the fuzzy comprehensive evaluation method.

[0072] In this embodiment, the scores of external attributes of the data are quantified based on the Fuzzy Comprehensive Evaluation (FCE) method. The Fuzzy Comprehensive Evaluation method is used to quantify external attributes that are fuzzy and subjective. The specific steps are as follows: S321, First, determine the evaluation set: establish an index set D that includes all external attribute indicators of the data, establish an evaluation level set V that includes n evaluation levels, and set a corresponding quantitative score P for each evaluation level.

[0073] Specifically, the indicator set D: specific indicators of external attributes, D={ (Scarcity) (Number of demanders) (Policy impact) (Privacy protection level) (Compliance) (Data Access Control)}; Evaluation level set V: divided into 5 levels, V={ (Very high) (Higher) (generally), (Lower) (Very low)}, corresponding to a quantitative score P=(1.0,0.8,0.6,0.4,0.2).

[0074] S322, Construct the membership matrix R: For each external attribute indicator of the data, invite railway experts to conduct a grade evaluation and determine the membership degree of each indicator to each evaluation grade. This forms an m×n membership matrix R, where m is the number of external attribute indicators and n is the evaluation level.

[0075] Specifically, for each indicator Experts in the railway field were invited to conduct a grading evaluation to determine the degree to which each indicator belongs to each evaluation grade. (i is the index number, j is the level number), satisfying This forms a 6×5 membership matrix R.

[0076] S323, Secondly, based on the weight vector W of each external attribute index determined using the analytic hierarchy process (AHP), weighted fuzzy operations are used to combine the weight vector W with the membership matrix R to obtain the fuzzy comprehensive evaluation result vector. :

[0077] in This represents the degree of membership of the comprehensive evaluation result to the j-th evaluation level, reflecting the relative extent to which the overall evaluation object belongs to that level.

[0078] S324, finally, according to Calculate the comprehensive score of the external attributes of the data. , .

[0079]

[0080] in, This represents the quantitative score corresponding to the i-th evaluation level.

[0081] As shown in Figure 1, in step S400, based on the comprehensive utility value of the internal attributes of the data and the comprehensive score of the external attributes of the data, and taking into account the willingness of the demand side to pay, a willingness-to-pay model of the demand side is constructed. Based on the willingness-to-pay model of the demand side to pay, a profit function is established, and the optimal pricing result is obtained by solving with the goal of maximizing profit.

[0082] The above steps specifically include: S410, defining the willingness-to-pay model as follows:

[0083] in, The highest willingness to pay by the demand side. The comprehensive utility value of the data's internal attributes. This is the willingness-to-pay coefficient. This reflects the overall market purchasing tendency or price sensitivity, making , This is a comprehensive score based on the external attributes of the data.

[0084] S420, assuming there is a market One potential demander, and the actual willingness of the demander to pay. This is constrained by the probability distribution function f(w). If the average cost per unit of data asset is... The price of data assets is Then the profit function corresponding to the platform's data assets for:

[0085] Assuming the actual willingness of the demand side to pay follows If the distribution between them is uniform, then the profit function Represented as:

[0086]

[0087] Indicates the willingness of the demand side to pay. The cumulative distribution function over time, i.e., the price at which demanders are willing to pay. The probability of purchasing data at a lower price. However, when the actual price exceeds the demander's willingness to pay, the demander cannot obtain positive utility and therefore will not make a purchase decision; only when... Only when demand is met will the buyer be willing to accept the transaction. Therefore, in pricing models, when... From time to time .therefore:

[0088] The profit function can be expressed as the following nonlinear programming problem:

[0089]

[0090]

[0091]

[0092] The goal is to maximize the profit function, given the condition that the data has intrinsic utility. Not less than 0, external attribute score of data Not less than 0, data price Not less than 0.

[0093] Higher data quality and greater market scarcity lead to higher optimal pricing; however, excessively high prices can cause demand to decline and profits to decrease. Therefore, there is a diminishing marginal return between internal utility and external score and the optimal price, requiring a balance between revenue and demand.

[0094] S430, regarding the profit function Take the partial derivative of the data pricing p, and set the partial derivative to 0, that is:

[0095] Solving for the optimal price :

[0096] That is, the optimal price is positively correlated with the combined utility value of the data's internal attributes and the combined score of the data's external attributes.

[0097] Because of the optimal price Only with Related, given Observe as prices change Changes in profits The changes are shown in Figure 3, and there is only one maximum value when the price... When the set value is too large ( ),profit This is because a transaction will not occur if the price is higher than the demander's maximum willingness to pay.

[0098] Next analysis On profit and price The impact, assuming , The relationship between the changes in profit and price is shown in Figure 4. If the data is too small, its utility to the demand side is too low, and the demand side will not buy it. As prices gradually increase, profits and the optimal price will also increase. Similarly, That's also true.

[0099] In summary, the utility of internal data attributes Too low or external attribute score If the value is too low, the data has little value to the demand side, and the demand side will not buy the data. Only when the internal and external value of the data is reasonable will data transactions be promoted.

[0100] Corresponding to the aforementioned railway data pricing method based on multidimensional value assessment, this invention also discloses a railway data pricing system based on multidimensional value assessment, which specifically includes: an evaluation index system construction module, used to construct a comprehensive evaluation index system for railway data value that takes into account both internal and external attributes of the data, including an objective layer, a criterion layer, and an index layer; an index weight determination module, used to determine the normalized weight of each index in the comprehensive evaluation index system using the analytic hierarchy process (AHP) and perform consistency checks; an internal and external attribute evaluation module, used to quantify the comprehensive utility value of the internal attributes of the data using a utility function based on the weights of each index, and quantify the comprehensive score of the external attributes of the data using a fuzzy comprehensive evaluation method; and a pricing solution module, used to construct a demander's willingness to pay model based on the comprehensive utility value of the internal attributes of the data and the comprehensive score of the external attributes of the data, considering the demander's willingness to pay, establish a profit function based on the demander's willingness to pay model, and solve for the optimal pricing result with profit maximization as the objective.

[0101] It should be noted that for a detailed description of the railway data pricing system based on multidimensional value assessment provided in the embodiments of the present invention, please refer to the relevant description of the railway data pricing method based on multidimensional value assessment provided in the embodiments of the present invention, which will not be repeated here.

[0102] In addition, embodiments of the present invention also provide an electronic device, the device comprising: a processor and a memory; the memory for storing one or more program instructions; the processor for executing one or more program instructions to perform the steps of a railway data pricing method based on multidimensional value assessment as described in any of the preceding embodiments.

[0103] It should be noted that for a detailed description of an electronic device provided in the embodiments of the present invention, please refer to the relevant description of a railway data pricing method based on multidimensional value assessment provided in the embodiments of this application, which will not be repeated here.

[0104] In addition, embodiments of the present invention also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of a railway data pricing method based on multidimensional value assessment as described in any of the preceding claims.

[0105] It should be noted that for a detailed description of the computer-readable storage medium provided in the embodiments of the present invention, please refer to the relevant description of the railway data pricing method based on multidimensional value assessment provided in the embodiments of this application, which will not be repeated here.

[0106] The above examples illustrate the present invention only to aid in understanding it and are not intended to limit the scope of the invention. Those skilled in the art can make various simple deductions, modifications, or substitutions based on the principles of this invention.

Claims

1. A railway data pricing method based on multidimensional value assessment, characterized in that, The method includes: constructing a comprehensive evaluation index system for railway data value that takes into account both internal and external attributes of the data, including an objective layer, a criterion layer, and an indicator layer; using the analytic hierarchy process (AHP) to determine the normalized weights of each indicator in the comprehensive evaluation index system and performing a consistency check; based on the weights of each indicator, using a utility function to quantify the comprehensive utility value of the internal attributes of the data, and using a fuzzy comprehensive evaluation method to quantify the comprehensive score of the external attributes of the data; based on the comprehensive utility value of the internal attributes of the data and the comprehensive score of the external attributes of the data, considering the willingness of the demand side to pay, constructing a willingness-to-pay model of the demand side, establishing a profit function based on the willingness-to-pay model of the demand side to pay, and solving for the optimal pricing result with the goal of maximizing profit.

2. The railway data pricing method based on multidimensional value assessment as described in claim 1, characterized in that, A comprehensive evaluation index system for railway data value is constructed, taking into account both internal and external attributes of the data. This system comprises an objective layer, a criterion layer, and an indicator layer. Specifically, the objective layer represents the comprehensive value of railway data assets; the criterion layer includes internal and external data attributes, with internal attributes comprising sub-criterions for data quality and data scale, and external attributes comprising sub-criterions for data market attributes and data security; and the indicator layer includes data quality indicators such as accuracy, completeness, consistency, and timeliness; data scale indicators such as data volume, growth rate, and update rate; data market attribute indicators such as scarcity, number of demanders, and policy impact; and data security indicators such as privacy protection level, compliance, and data access control.

3. The railway data pricing method based on multidimensional value assessment as described in claim 1, characterized in that, The analytic hierarchy process (AHP) is used to determine the normalized weights of each indicator in the comprehensive evaluation index system and to perform consistency checks. Specifically, this includes: establishing a hierarchical structure model: clarifying the hierarchical relationships between the target layer, criterion layer, and indicator layer; constructing pairwise comparison judgment matrices: using a 1-9 scale to compare the importance of indicators at the same level pairwise to form judgment matrices; and calculating and normalizing eigenvectors: for each judgment matrix, calculating the largest eigenvalue. The corresponding feature vectors are normalized to obtain the initial weights of each indicator; a consistency test is performed: the consistency test is judged by the consistency ratio CR. When CR < 0.1, the weights are valid; otherwise, the judgment matrix is ​​adjusted and recalculated; finally, the normalized weights of each indicator are obtained.

4. The railway data pricing method based on multidimensional value assessment as described in claim 1, characterized in that, The method utilizes utility functions to quantify the comprehensive utility value of internal attributes of data. Specifically, it includes: constructing 10 rating levels for each internal attribute indicator, with a range of 0.1-1 and a step size of 0.1, to simulate different attribute levels; generating multiple datasets with different attribute levels based on the 10 rating levels of each internal attribute indicator; predicting the classification accuracy of each dataset using the XGBoost model on the downstream task of the demand side, and using the classification accuracy predicted by the XGBoost model as the actual utility benchmark of the internal attributes of the data; selecting the Gaussian cumulative distribution function as the basic utility function model, and aiming to minimize the sum of squared errors between the fitted value of the utility function and the XGBoost classification accuracy representing the actual utility, solving for the optimal parameters using the least squares method to obtain the optimal utility function model with optimal parameters.

5. A railway data pricing method based on multidimensional value assessment as described in claim 4, characterized in that, The method of quantifying the comprehensive utility value of internal attributes of data using utility functions includes: converting the actual values ​​of each internal attribute indicator into ranking scores and then inputting them into the optimal utility function model to calculate the utility value of each internal attribute indicator; and finally, weighted summing of the utility values ​​of each internal attribute indicator with their respective weights to obtain the comprehensive utility value of the internal attributes of the data. ,in This refers to the number of internal attribute indicators. For the first The weight of each internal attribute indicator, For the first The utility value of each internal attribute indicator.

6. The railway data pricing method based on multidimensional value assessment as described in claim 1, characterized in that, The fuzzy comprehensive evaluation method is used to quantify the comprehensive score of external attributes of data. Specifically, this includes: first, determining the evaluation set: establishing an index set D that includes all external attribute indicators of the data, and establishing... A set of rating levels for each rating level Assign a corresponding quantitative score P to each evaluation level; construct a membership matrix. For each external attribute indicator of the data, railway experts were invited to conduct a rating evaluation to determine the degree of membership of each indicator to each rating level. This forms an m×n membership matrix R, where... The number of metrics representing external attributes of the data. First, an evaluation level is established. Second, based on the weight vector W of each external attribute indicator determined using the analytic hierarchy process (AHP), a weighted fuzzy operation is used to combine the weight vector W with the membership matrix R to obtain the fuzzy comprehensive evaluation result vector. Finally, according to Calculate the comprehensive score of the external attributes of the data. ， 。 7. A railway data pricing method based on multidimensional value assessment as described in claim 1, characterized in that, Based on the comprehensive utility value of internal data attributes and the comprehensive score of external data attributes, and considering the willingness of demanders to pay, a willingness-to-pay model is constructed, specifically including: defining the willingness-to-pay model as follows: in, The highest willingness to pay by the demand side. The comprehensive utility value of the data's internal attributes. This is the willingness-to-pay coefficient. This reflects the overall market purchasing tendency or price sensitivity, making ， This is a comprehensive score based on the external attributes of the data.

8. A railway data pricing method based on multidimensional value assessment as described in claim 7, characterized in that, A profit function is established based on the demand-side willingness-to-pay model, and the optimal pricing result is obtained by solving the problem with the goal of maximizing profit. Specifically, this includes assuming that there exists a profit function in the market. One potential demander, and the actual willingness of the demander to pay. Subject to probability distribution function Constraints; if the average cost of data assets is The price of data assets is Then the profit function corresponding to data assets for: Assuming the actual willingness of the demand side to pay follows If the distribution is uniform, then the profit function π can be expressed as: Indicates the willingness of the demand side to pay. The cumulative distribution function represents the probability that the demander is willing to purchase data at price p or lower. However, when the actual price exceeds the demander's willingness to pay, the demander cannot obtain positive utility and will not make a purchase decision. Only when the actual price is higher than the demander's willingness to pay will the demander make a purchase decision. Only when the demand side is in a certain state is it possible for them to accept the transaction. Therefore, when From time to time ,but: The profit function can be expressed as a nonlinear programming problem as follows: The objective is to maximize the profit function, with the following constraints: ; For the profit function Take the partial derivative of the data pricing p, and set the partial derivative to 0, that is: Get the optimal price ： That is, the optimal price is positively correlated with the combined utility value of the data's internal attributes and the combined score of the data's external attributes.

9. A railway data pricing system based on multidimensional value assessment, characterized in that, The system includes: an evaluation index system construction module, used to construct a comprehensive evaluation index system for railway data value that takes into account both internal and external attributes of the data, including an objective layer, a criterion layer, and an index layer; an index weight determination module, used to determine the normalized weights of each index in the comprehensive evaluation index system using the analytic hierarchy process (AHP) and to perform consistency checks; an internal and external attribute evaluation module, used to quantify the comprehensive utility value of the internal attributes of the data using a utility function based on the weights of each index, and to quantify the comprehensive score of the external attributes of the data using a fuzzy comprehensive evaluation method; and a pricing solution module, used to construct a demand-side payment willingness model based on the comprehensive utility value of the internal attributes of the data and the comprehensive score of the external attributes of the data, taking into account the demand-side payment willingness, establish a profit function based on the demand-side payment willingness model, and solve for the optimal pricing result with profit maximization as the objective.

10. An electronic device, characterized in that, The device includes: a processor and a memory; the memory is used to store one or more program instructions; the processor is used to run one or more program instructions to perform the steps of a railway data pricing method based on multidimensional value assessment as described in any one of claims 1 to 8.

Images