Management capability assessment method, device and equipment

Abstract

The invention relates to a management capability assessment method, device and equipment, and relates to the technical field of computers, and the method comprises the steps: obtaining an original data set corresponding to a plurality of to-be-assessed service lines; each original data set comprises index data corresponding to the corresponding service line and a plurality of evaluation indexes; aiming at each evaluation index, taking the index data, corresponding to the evaluation index, of each service line as input, and screening out target index data corresponding to the evaluation index from the multiple pieces of index data based on a K-nearest neighbor model; for each service line, obtaining a target data set corresponding to the service line, and determining an evaluation index corresponding to each target index data in the target data set as a target evaluation index corresponding to the service line; and factor analysis is performed on the target data set corresponding to each business line, and the comprehensive score value corresponding to each business line is determined based on the factor analysis result, so that the management capability of the coal enterprise is accurately and efficiently evaluated.

Description

Technical Field

[0001] This application relates to the field of enterprise management technology, and in particular to a management capability assessment method, apparatus and equipment. Background Technology

[0002] In the standardized management of the coal industry, there are multiple dimensions of evaluation indicators, including: production capacity evaluation, profitability evaluation, operational capacity evaluation, overall efficiency evaluation, safety management evaluation, cost management evaluation, energy conservation and environmental protection evaluation, innovation management evaluation, and talent quality evaluation. By combining various data generated from the operation of coal enterprises with these multiple indicators for scientific evaluation and prediction, a comprehensive analysis of the enterprise's management capabilities and level can be achieved, enabling precise benchmarking and empowerment for enterprises in the industry.

[0003] In related technologies, when assessing management capabilities, an indicator system is constructed, and evaluation indicators are obtained by processing the indicator system based on sparse factor analysis. An evaluation model is then established based on the evaluation indicators. However, this method has high computational complexity and requires a long computation time. Furthermore, solving sparse optimization problems requires determining multiple intermediate parameters, and the selection of these parameters is uncertain in complex and ever-changing real-world environments, which increases the instability of the solution.

[0004] Therefore, there is an urgent need for a method that can accurately and efficiently assess the management capabilities of coal enterprises. Summary of the Invention

[0005] This invention provides a management capability assessment method for accurately and efficiently evaluating the management capabilities of coal enterprises.

[0006] In a first aspect, embodiments of this application provide a management capability assessment method, the method comprising: In response to the management capability assessment instructions, obtain the original datasets corresponding to the multiple business lines to be assessed; each original dataset includes: the indicator data corresponding to the corresponding business line and multiple evaluation indicators respectively; For each evaluation indicator, the indicator data corresponding to the evaluation indicator for each business line is used as input, and the target indicator data corresponding to the evaluation indicator is selected from multiple indicator data based on the K-nearest neighbor model. For each business line, a target dataset corresponding to the business line is obtained, and the evaluation indicators corresponding to each target indicator data in the target dataset are determined as the target evaluation indicators corresponding to the business line; the target dataset includes: multiple target indicator data corresponding to the corresponding business line; Factor analysis was performed on the target datasets corresponding to each business line, and the comprehensive score value corresponding to each business line was determined based on the factor analysis results.

[0007] The above method, when conducting capability assessment, uses the K-nearest neighbor model to screen indicator data and evaluation indicators, and performs factor analysis on the screened data to obtain a comprehensive score. The K-nearest neighbor model removes outlier data and interference from indicators, while factor analysis improves the accuracy of the comprehensive score calculation.

[0008] In one possible implementation, the parameters corresponding to the K-nearest neighbor model include reference values ​​for each evaluation index and the value of K; the step of filtering target index data corresponding to the evaluation index from multiple index data based on the K-nearest neighbor model includes: Calculate the distance between each of the multiple indicator data and the reference value of the evaluation indicator; Based on the distances corresponding to the multiple indicator data, K indicator data are selected from the multiple indicator data as target indicator data corresponding to the evaluation indicator.

[0009] The above method improves the accuracy of indicator data selection by calculating the distance between the indicator data and the reference value of the evaluation indicator to filter the target indicator data.

[0010] In one possible implementation, the K-nearest neighbor model is obtained by iteratively training a base model based on the original datasets and comprehensive score values ​​corresponding to multiple business lines in multiple capability management assessment operations prior to the current capability management assessment operation; wherein, the following operations are performed during one iteration of training: In the current capability management assessment operation, obtain the original datasets and comprehensive score values ​​corresponding to multiple business lines respectively, and filter the target business line from the multiple business lines based on the comprehensive score values; Determine the reference outputs for each target business line; The original datasets corresponding to each target business line are input into the K-nearest neighbor model obtained after the previous training cycle, with the reference output corresponding to each target business line as the target, and the K-nearest neighbor model obtained after the previous training cycle is trained.

[0011] The above method avoids interference from abnormal business lines in model training by screening business lines based on comprehensive capability scores, and improves the accuracy of model classification by iteratively training the K-nearest neighbor model.

[0012] In one possible implementation, before filtering the target indicator data corresponding to the evaluation indicator from multiple indicator data based on the K-nearest neighbor model, the method further includes: For each evaluation indicator, calculate the average value and standard deviation of the indicator data corresponding to each business line and the evaluation indicator; For each indicator data corresponding to each business line, the indicator data is standardized based on the indicator data and the average and standard deviation of the evaluation indicators corresponding to the indicator data; The step of selecting target indicator data corresponding to the evaluation indicator from multiple indicator data based on the K-nearest neighbor model includes: The target indicator data corresponding to the evaluation indicator is selected from multiple standardized indicator data based on the K-nearest neighbor model.

[0013] The above method eliminates the influence of dimensions by standardizing the indicator data.

[0014] In one possible implementation, before performing factor analysis on the target datasets corresponding to each business line, the method further includes: A sampling suitability test is performed on the target datasets corresponding to each business line to obtain the first detection result; Bartlett's test of sphericity was performed on the target datasets corresponding to each business line to obtain the second detection result; The first and second detection results are determined to meet the preset conditions.

[0015] The above method determines whether the target dataset can be used for factor analysis by using sampling suitability tests and Bartlett's test of sphericity, thus avoiding waste of computational resources.

[0016] In one possible implementation, the step of performing factor analysis on the target datasets corresponding to each business line, and determining the comprehensive score value corresponding to each business line based on the factor analysis results, includes: Factor analysis was performed on the target datasets corresponding to each business line to obtain multiple principal factors and the loading coefficients of each principal factor and each target evaluation index. For each business line, a comprehensive score value is determined based on the target indicator data corresponding to the business line and the loading coefficient of each main factor.

[0017] The above method, by performing factor analysis on the target dataset and determining the comprehensive score of the business line based on the principal factor loading coefficients, ensures the accuracy of the evaluation.

[0018] In one possible implementation, determining the comprehensive score value corresponding to the business line based on the target indicator data corresponding to the business line and the loading coefficients corresponding to each principal factor includes: Based on the target indicator data corresponding to the business line and the loading coefficient of each principal factor, the score value corresponding to each principal factor is determined. The comprehensive score value corresponding to the business line is determined by weighted calculation based on the weight and score value of each main factor. The weights are determined based on the variance contribution rate of the corresponding principal factors during factor analysis.

[0019] The above method uses the variance contribution rate of the principal factors in the factor analysis process as the weight to perform weighted calculation of the evaluation values ​​corresponding to each principal factor, thereby further ensuring the accuracy of the comprehensive score.

[0020] Secondly, embodiments of this application provide a management capability assessment device, the device comprising: The data acquisition module is used to acquire the original datasets corresponding to multiple business lines to be evaluated in response to the management capability assessment instructions; each original dataset includes: the indicator data corresponding to the corresponding business line and multiple evaluation indicator items respectively; The data filtering module is used to filter out the target indicator data corresponding to the above evaluation indicator from multiple indicator data based on the K-nearest neighbor model for each evaluation indicator, taking the indicator data of each business line corresponding to the above evaluation indicator as input. The indicator determination module is used to obtain the target dataset corresponding to each business line, and determine the evaluation indicators corresponding to each target indicator data in the target dataset as the target evaluation indicators corresponding to the business line. The target dataset includes multiple target indicator data corresponding to the corresponding business line. The scoring determination module is used to perform factor analysis on the target datasets corresponding to each business line, and determine the comprehensive score value corresponding to each business line based on the factor analysis results.

[0021] In one possible implementation, the parameters of the K-nearest neighbor model include reference values ​​for each evaluation index and the value of K; the data filtering module is specifically used for: Calculate the distance between the data of each of the above indicators and the reference values ​​of the above evaluation indicators respectively; Based on the distances corresponding to the above multiple indicator data, select K indicator data as target indicator data corresponding to the above evaluation indicators.

[0022] In one possible implementation, the aforementioned K-nearest neighbor model is obtained by iteratively training a base model based on the original datasets and comprehensive score values ​​corresponding to multiple business lines in multiple capability management assessment operations prior to the current capability management assessment operation; wherein, the following operations are performed during one iteration of training: Obtain the original datasets and comprehensive scores for multiple business lines in the current capability management assessment operation, and filter the target business line from the multiple business lines based on the comprehensive scores; Determine the reference outputs for each target business line; The original datasets corresponding to each target business line are input into the K-nearest neighbor model obtained after the previous training iteration, with the reference output corresponding to each target business line as the objective, and the K-nearest neighbor model obtained after the previous training iteration is trained.

[0023] In one possible implementation, the data filtering module described above is specifically used for: For each evaluation indicator, calculate the average value and standard deviation of the indicator data corresponding to the above evaluation indicators for each business line; For each indicator data corresponding to each business line, the above indicator data is standardized based on the average value and standard deviation of the evaluation indicators corresponding to the above indicator data. Based on the K-nearest neighbor model, target indicator data corresponding to the above evaluation indicators are selected from multiple standardized indicator data.

[0024] In one possible implementation, the scoring determination module described above is specifically used for: A sampling suitability test is performed on the target datasets corresponding to each business line to obtain the first detection result; Bartlett's test of sphericity was performed on the target datasets corresponding to each business line to obtain the second detection result; The first and second test results are determined to meet the preset conditions.

[0025] In one possible implementation, the scoring determination module described above is specifically used for: Factor analysis was performed on the target datasets corresponding to each business line to obtain multiple principal factors and the loading coefficients of each principal factor and each target evaluation index. For each business line, the comprehensive score value is determined based on the target indicator data corresponding to the business line and the loading coefficient of each main factor.

[0026] In one possible implementation, the scoring determination module described above is specifically used for: Based on the target indicator data corresponding to the above business lines and the loading coefficients of each principal factor, the score value corresponding to each principal factor is determined. The comprehensive score for the above business lines is determined by weighted calculation based on the weight and score value of each main factor. The aforementioned weights are determined based on the variance contribution rate of the corresponding principal factors during factor analysis.

[0027] Thirdly, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps in the management capability assessment method.

[0028] Fourthly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the management capability assessment method described above.

[0029] Fifthly, embodiments of this application provide a computer program product, including a computer program stored in a computer-readable storage medium; when a processor of a memory access device reads the computer program from the computer-readable storage medium, the processor executes the computer program, causing the memory access device to perform the steps in the management capability assessment method described above in this application.

[0030] For details on each of the second to fifth aspects mentioned above, as well as the technical effects that each aspect may achieve, please refer to the above description of the technical effects that can be achieved for the first aspect or the various possible solutions in the first aspect. These details will not be repeated here. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 A flowchart illustrating a management capability assessment method provided in this application embodiment; Figure 2 A schematic diagram of a gravel map provided in an embodiment of this application; Figure 3 A schematic diagram of a heat map provided in an embodiment of this application; Figure 4 A schematic diagram of a factor loading plot provided in an embodiment of this application; Figure 5 A flowchart of a management capability assessment process provided in this application embodiment; Figure 6 A schematic diagram of a management capability assessment device provided in an embodiment of this application; Figure 7 This is a schematic diagram of a possible electronic device provided in an embodiment of this application. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The specific operating methods in the method embodiments can also be applied to the device embodiments or system embodiments.

[0034] In the description of this application, "multiple" is understood as "at least two". "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. A connected to B can represent: A and B directly connected, or A and B connected through C. Furthermore, in the description of this application, terms such as "first" and "second" are used only for descriptive purposes and should not be construed as indicating or implying relative importance or order.

[0035] For ease of understanding, the technical terms used in the embodiments of this application are introduced here: (1) Z-score standardization: also known as standard deviation standardization, is a method of standardizing data based on the mean and standard deviation of the original data. It transforms the original values ​​to obtain standardized values. This method is particularly suitable for attributes whose maximum and minimum values ​​are unknown, or for situations where there are outliers that exceed the range of values.

[0036] (2) Exploratory Factor Analysis (EFA): It is a technique used to find out the essential structure of multivariate observed variables and process them to reduce dimensionality. Therefore, it can integrate variables with complex relationships into a few core factors. The factor extraction method of principal factor analysis does not have outliers, intervals, linear values, normal distribution of multivariates, or orthogonality.

[0037] The coal industry standards conference released evaluation indicators for multiple dimensions of standardized management, including: production capacity evaluation, profitability evaluation, operational capacity evaluation, comprehensive efficiency evaluation, safety management evaluation, cost management evaluation, energy conservation and environmental protection evaluation, innovation management evaluation, and talent quality evaluation.

[0038] In the era of rapid big data development, scientifically evaluating and predicting various data generated by coal enterprise operations, combined with multiple dimensions of indicators, allows for a comprehensive analysis of the enterprise's management capabilities and level. This provides objective self-evaluation for industry enterprises, offering dynamic industry-level benchmarks for precise alignment and empowerment. The "data + indicator" model maximizes the value of key coal mine industry benchmarks. Timely development and implementation of comprehensive and professional index scoring methods, along with targeted efforts and multi-faceted breakthroughs, ensure more practically verifiable benchmarking results. This accumulates strength for coal enterprises to establish a more robust management system, enhance their competitiveness, and drive a significant leap forward in the management level of the coal industry.

[0039] In related technologies, when assessing management capabilities, an indicator system is constructed, and evaluation indicators are obtained by processing the indicator system using sparse factor analysis. An evaluation model is then established based on these indicators. The core of this method is constructing the indicator system and the evaluation model, but it suffers from the following two problems: First, the construction of the indicator system did not incorporate industry guidelines, standards, and data, and its rationality needs to be examined. Secondly, using sparse factor analysis is difficult, has high computational complexity, and requires a long computation time. Furthermore, solving sparse optimization problems requires determining multiple intermediate parameters, such as initial values, iteration step size, and regularization parameters. The selection of these parameters is uncertain in complex and ever-changing real-world environments, which increases the instability of the solution.

[0040] Therefore, there is an urgent need for a method that can accurately and efficiently assess the management capabilities of coal enterprises.

[0041] In response, this application provides a management capability assessment method. After collecting the indicator data to be assessed using big data technology, the method uses a K-nearest neighbor model to filter the indicators and obtain objective and effective target indicator data. Combined with the evaluation indicators of industry benchmarks, the method uses factor analysis to construct the model. Factor analysis can complete the model operation with less resource consumption, and the model has strong adaptability. Even when the source data changes in real time, it can still support the continuous operation of the model and output results, and can obtain real-time and accurate management capability assessment values.

[0042] The present application will now be described in further detail with reference to the accompanying drawings. See also: Figure 1 The diagram shown is a flowchart of a management capability assessment method provided in an embodiment of this application. The specific implementation process is as follows: Step S101: In response to the management capability assessment instruction, obtain the original datasets corresponding to the multiple business lines to be assessed; Each original dataset includes: indicator data corresponding to the relevant business line and multiple evaluation indicators.

[0043] In some embodiments, the above-mentioned evaluation indicators can be set based on needs. Optionally, the above-mentioned evaluation indicators may include at least one of the following: capacity utilization rate, return on total assets, land reclamation rate, R&D investment intensity, degree of mechanization of fully mechanized mining, proportion of employees with technician and senior technician qualifications, degree of mechanization of tunneling, and return on net assets.

[0044] Step S102: For each evaluation indicator, the indicator data corresponding to each business line and the evaluation indicator are used as input, and the target indicator data corresponding to the evaluation indicator is selected from multiple indicator data based on the K-nearest neighbor model. In some embodiments, before inputting the indicator data into the K-nearest neighbor model, this application needs to standardize each indicator data to eliminate the influence of dimensions. The specific standardization method can be set according to the requirements, and this application does not limit it.

[0045] In one possible implementation, before filtering out the target indicator data corresponding to the evaluation indicator from multiple indicator data based on the K-nearest neighbor model, the method further includes: For each evaluation indicator, calculate the average and standard deviation of the indicator data corresponding to each business line and the evaluation indicator; For each indicator data corresponding to each business line, the indicator data is standardized based on the indicator data and the average and standard deviation of the evaluation indicators corresponding to the indicator data; Based on the K-nearest neighbor model, target indicator data corresponding to the evaluation indicators are selected from multiple indicator data, including: The target indicator data corresponding to the evaluation indicator is selected from multiple standardized indicator data based on the K-nearest neighbor model.

[0046] In some embodiments, this application standardizes the data of each indicator based on the Z-score method. Specifically, it obtains the first... The evaluation metric corresponds to multiple metric data points, i.e., metric data points from each business line, and the following formula is used to analyze the data points of the multiple metric data points. Standardize the key performance data for each business line: ; in, For the first The corresponding evaluation index is the first Key performance data for each business line For the first The average value of indicator data from multiple business lines corresponding to each evaluation indicator. For the first The standard deviation of indicator data from multiple business lines corresponding to each evaluation indicator.

[0047] After obtaining the standardized results corresponding to each indicator data based on the above method, since Z-score standardization cannot identify data features and cannot effectively integrate the data after removing the units, this embodiment of the application uses the K-nearest neighbor (KNN) model for optimization to determine the target indicator data corresponding to each business line.

[0048] In one possible implementation, the parameters of the K-nearest neighbor model include reference values ​​for each evaluation index and the value of K; based on the K-nearest neighbor model, target index data corresponding to the evaluation index is selected from multiple index data, including: Calculate the distance between multiple indicator data and the reference value of the evaluation indicator; Based on the distances corresponding to multiple indicator data, select K indicator data as target indicator data corresponding to the evaluation indicators.

[0049] In some embodiments, the reference values ​​corresponding to each of the above evaluation indicators are the values ​​of the center points corresponding to each evaluation indicator stored in the K-nearest neighbor model. The reference values ​​corresponding to the above evaluation indicators and the value of K will be automatically adjusted and optimized during the model training process.

[0050] When selecting target indicator data based on the K-nearest neighbor model, for any evaluation indicator, the distances between multiple indicator data and the reference value of the evaluation indicator are calculated separately, and the K nearest indicator data are selected as the target indicator data corresponding to the evaluation indicator.

[0051] In one possible implementation, the K-nearest neighbor model is obtained by iteratively training a base model based on the original datasets and comprehensive score values ​​corresponding to multiple business lines in multiple capability management assessment operations prior to the current capability management assessment operation; wherein, the following operations are performed during one iteration of training: Obtain the original datasets and comprehensive scores for multiple business lines in the current capability management assessment operation, and filter the target business line from multiple business lines based on the comprehensive score; Determine the reference outputs for each target business line; The original datasets corresponding to each target business line are input into the K-nearest neighbor model obtained after the previous training iteration, with the reference output corresponding to each target business line as the objective, and the K-nearest neighbor model obtained after the previous training iteration is trained.

[0052] In some embodiments, Table 1 below shows a possible value of the model parameters corresponding to the above-mentioned basic model. During the iterative training of the basic model, the corresponding model parameters will be automatically adjusted to improve the accuracy of the model classification in the next iteration.

[0053] Table 1

[0054] It should be noted that since the K-nearest neighbor model is the base model and has not been trained when the capacity management assessment is performed for the first time, it is not necessary to perform the steps described in S103-S103 above when the capacity management assessment is performed for the first time. That is, factor analysis can be performed directly on the original dataset.

[0055] In some embodiments, after each capability management assessment operation, the parameters in the K-nearest neighbor model are optimized based on the capability management assessment results and the original dataset used in the current capability management assessment operation, so as to facilitate subsequent use. For example, when the K-nearest neighbor model is performing classification, the number of leaves in its parameters can be optimized for the measurement error vector, and the number of neighbors can be optimized for the observable variable vector.

[0056] Specifically, during the training process, the above-mentioned selection of target business lines from multiple business lines based on comprehensive score values ​​can be: selecting business lines with comprehensive score values ​​greater than a preset value as target business lines, or selecting the top preset number of business lines as target business lines; when determining the reference output corresponding to each target business line, the reference output is the evaluation index corresponding to the target business line, and the selection rules for the evaluation index can be set according to the requirements, for example, it can be selected based on the difference between the index data of each evaluation index corresponding to the target business line and the average index data of the corresponding evaluation index.

[0057] Step S103: For each business line, obtain the target dataset corresponding to the business line, and determine the evaluation indicators corresponding to each target indicator data in the target dataset as the target evaluation indicators corresponding to the business line. The target dataset includes multiple target indicator data corresponding to the relevant business lines; It should be noted that the target evaluation indicators for each business line may differ.

[0058] Step S104: Perform factor analysis on the target datasets corresponding to each business line, and determine the comprehensive score value corresponding to each business line based on the factor analysis results.

[0059] In some embodiments, to ensure the effectiveness of factor analysis, this application needs to check the target datasets corresponding to each business line before executing the factor analysis algorithm to determine whether they are suitable for factor analysis.

[0060] In one possible implementation, before performing factor analysis on the target datasets corresponding to each business line, the method further includes: A sampling suitability (KMO) test is performed on the target datasets corresponding to each business line to obtain the first detection result; Bartlett's test of sphericity was performed on the target datasets corresponding to each business line to obtain the second detection result; The first and second test results are determined to meet the preset conditions.

[0061] Specifically, using each target evaluation index as a variable, a KMO check is performed on the target dataset corresponding to each business line to determine the partial correlation between the variables. The higher the value of the first check result (i.e., the closer it is to 1), the more suitable it is for factor analysis. In practice, a value greater than 0.9 can be set to indicate that factor analysis is very suitable, a value between 0.7 and 0.9 indicates that factor analysis is suitable, a value between 0.6 and 0.7 indicates that factor analysis can be performed, a value between 0.5 and 0.6 indicates that the effect of factor analysis is poor, and a value less than 0.5 indicates that the effect of factor analysis is extremely poor. In this embodiment, factor analysis can be abandoned if the first check result is not greater than a first threshold. The value of the first threshold can be set based on the requirements.

[0062] Using the target evaluation indicators as variables, Bartlett's test is performed on the target datasets corresponding to each business line to determine whether the correlation matrix between the variables is an identity matrix (i.e., whether the variables are completely uncorrelated). If the corresponding second check result is less than a certain threshold (e.g., 0.05), it indicates that the null hypothesis is rejected, meaning that the variables are significantly correlated and factor analysis can be performed; otherwise, it indicates that the variables are highly independent and not suitable for factor analysis. In this embodiment, factor analysis can be abandoned if the first check result is not less than the second threshold. The value of the second threshold can be set according to requirements.

[0063] In some embodiments, the aforementioned preset conditions of this application may be that the first inspection result is greater than the first threshold and the second inspection result is less than the second threshold.

[0064] In some embodiments, if the preset conditions are met based on the first and second inspection results, it indicates that the target dataset corresponding to the current business line is not suitable for factor analysis. In this case, the application can increase the number of data to be evaluated, such as obtaining the original datasets corresponding to each business line multiple times at different time periods, and uniformly processing the original datasets of each business line obtained multiple times before performing KMO and Bartlett's test of sphericity. Alternatively, the target evaluation indicators corresponding to each business line can be adjusted before performing KMO and Bartlett's test of sphericity, such as deleting unsuitable target evaluation indicators (when deleting, the MSA value in the correlation matrix can be referenced), merging highly correlated target evaluation indicators, or adding / replacing target evaluation indicators, etc.

[0065] In some embodiments, this application does not limit the specific algorithm of the above factor analysis. Optionally, the comprehensive score value corresponding to each business line can be determined based on the principal factor analysis algorithm.

[0066] In one possible implementation, factor analysis is performed on the target dataset corresponding to each business line, and a comprehensive score value corresponding to each business line is determined based on the factor analysis results, including: Factor analysis was performed on the target datasets corresponding to each business line to obtain multiple principal factors and the loading coefficients of each principal factor and each target evaluation index. For each business line, a comprehensive score is determined based on the target indicator data corresponding to the business line and the loading coefficient of each main factor.

[0067] In some embodiments, this application takes the target dataset corresponding to each business line as input, where the variables are the set of target evaluation indicators corresponding to each business line and the index item is the business line. It analyzes the dataset using a principal factor analysis algorithm to obtain a variance interpretation table, scree plot, component matrix table, factor loading coefficient table, heatmap corresponding to the factor loading matrix, and factor loading plot. The component matrix table includes the loading coefficients corresponding to each principal factor and each target evaluation indicator. The factor loading coefficient table is obtained by factor rotation (such as the maximum variance method) on the component matrix in the component matrix table. The heatmap and factor loading plot are generated based on the factor loading coefficient table.

[0068] In practice, the number of principal factors can be determined by analyzing the variance explanation table and scree plot. The variance explanation table includes the variance contribution rate of each factor, which characterizes the contribution rate of each factor to the explanation of the variable (which can be understood as how many factors are needed to express the variable as 100%). If it is too low (e.g., below 60%), the factor data needs to be adjusted. The scree plot is a graph drawn based on the degree of explanation of the data variation by each principal component (principal factor). Its function is to determine the number of principal factors to be selected based on the slope of the eigenvalue descent. Combined with the variance explanation table, it can be used to confirm or adjust the number of principal factors. Each principal component is represented by a point, and the number of principal components to be extracted is determined by the position where the "slope tends to flatten". The scree plot is used to determine the number of factors to be selected based on the slope of the eigenvalue descent. The combination of these two methods can be used to confirm or adjust the number of principal factors.

[0069] In some embodiments, by analyzing the factor loading coefficient table and heatmap, the importance of latent variables in each factor can be analyzed. For example, when studying the distribution patterns of 25 useful elements in a "polymetallic ore body," each element is considered as an indicator. Assuming five factors are determined, among factor 1, SO, SO2, Na2S, HS, and H2S have relatively large factor loading coefficients. Therefore, factor 1 can be identified as the sulfide component, and so on. Latent variable analysis of each factor can also be performed in conjunction with specific business operations. The factor loading diagram presents the spatial distribution of factors in a quadrant diagram manner. It should be noted that if there are two factors, a three-dimensional factor loading scatter plot cannot be presented; if there is only one factor, the factor loading diagram cannot be displayed.

[0070] In one possible implementation, the comprehensive score value for each business line is determined based on the target indicator data corresponding to the business line and the loading coefficients corresponding to each principal factor, including: Based on the target indicator data corresponding to the business line and the loading coefficient of each principal factor, determine the score value of each principal factor. The comprehensive score for each business line is determined by weighted calculation based on the weight and score value of each main factor. The weights are determined based on the variance contribution rate of the corresponding principal factors during factor analysis.

[0071] In some embodiments, by analyzing the component matrix, the component formula of the principal factors (i.e., the sum of the target indicator data corresponding to each target evaluation indicator and the loading coefficient of the principal factors) and the weight of each principal factor can be obtained, and then the comprehensive score value corresponding to the business line can be calculated based on the component formula and weight.

[0072] In practice, for each principal factor, the target indicator data corresponding to each evaluation indicator in the target dataset corresponding to the above business line is weighted and summed with the principal factor and the loading coefficients corresponding to each principal factor and each target evaluation indicator, respectively, to obtain the score value corresponding to each principal factor. Based on the weights corresponding to each principal factor, the score values ​​corresponding to each principal factor are weighted and summed to obtain the comprehensive score value corresponding to the business line.

[0073] The above method, when conducting capability assessment, uses the K-nearest neighbor model to screen indicator data and evaluation indicators, and performs factor analysis on the screened data to obtain a comprehensive score. The K-nearest neighbor model removes outlier data and interference from indicators, while factor analysis improves the accuracy of the comprehensive score calculation.

[0074] For example, Table 2 below provides a schematic diagram of the target indicator data for each business line. Assume the variables (target evaluation indicators) are: capacity utilization rate (indicator 1), return on net assets (indicator 2), return on total assets (indicator 3), degree of mechanization of fully mechanized mining (indicator 4), degree of mechanization of fully mechanized tunneling (indicator 5), land reclamation rate (indicator 6), R&D investment intensity (indicator 7), and the proportion of employees with technician and senior technician qualifications (indicator 8). The index item is the business line (a total of 10 business lines, i.e., business line AJ). The factor rotation method is the maximum variance method, and the number of principal components is 8. The specific process of factor analysis is described below.

[0075] Table 2

[0076] By analyzing the data described in Table 1 using the principal factor analysis algorithm, the corresponding variance explanation table (not shown here) and scree plot (such as...) can be obtained. Figure 2 As shown in Table 3, the component matrix table (which describes the factor score coefficients contained in each component, i.e., the principal factor loading coefficients, used to calculate the component scores and derive the factor formula) is obtained by performing factor rotation on the component matrix table using the maximum variance method. The factor loading coefficient table shown in Table 4 is obtained as shown in Table 4.

[0077] Table 3

[0078] Table 4

[0079] Based on the above factor loading coefficient table, we can obtain the following: Figure 3 The heat map shown and as Figure 4 The factor loading plot is shown.

[0080] Table 4 shows the loadings and communality of each variable on the factors. The communality is close to or reaches 1, indicating that the extracted factors can fully explain the variance of the variables. Factor 1 has high loadings on capacity utilization (0.934) and the degree of mechanization in fully mechanized mining (0.793), reflecting the characteristics of production efficiency. The dominant variables of Factor 2 are R&D investment intensity (0.941) and the degree of mechanization in tunneling (-0.789), reflecting the inverse correlation between technology investment and equipment. Factor 3 is mainly driven by the proportion of employees with technician qualifications (0.956), representing the quality of human resources. Factor 4 is centered on the return on total assets (0.98), representing the asset profitability dimension. Factor 5 has a prominent loading on land reclamation rate (0.803), pointing to environmental restoration capacity. Factor 6 has a strong explanatory power for return on net assets (0.847), and is related to capital return performance. Factors 7 and 8 have no significant loading variables, and further combination... Figure 2The scree map assessment shown indicates that this factor should be retained.

[0081] After obtaining the principal factors (i.e., factors 1-8 above), the factor formulas can be derived from Table 3 above. Specifically, the factor analysis results show that the 8 original variables were extracted into 8 components, explaining a total variance of 100%. The component matrix shows significant differences in the loading distribution of variables across different components: capacity utilization (0.864) and mechanization level of fully mechanized mining (0.447) have high loadings on component 1, reflecting the dimension of production efficiency; R&D expenditure intensity (0.874) dominates component 2, reflecting the characteristics of innovation investment; the proportion of employees with technician qualifications (1.218) has a prominent loading on component 3, representing the quality of human resources; return on total assets (1.103) makes a significant contribution to component 4, representing asset profitability; land reclamation rate (1.717) and return on net assets (1.577) play a dominant role in components 5 and 6, respectively, reflecting the characteristics of environmental governance and capital returns; it is worth noting that components 7 and 8 have abnormally high loading values ​​(such as 3.016 and -7.707), which may be due to extreme values ​​in the data or overfitting of the model, and their statistical significance needs to be tested in conjunction with eigenvalues. The absolute values ​​of the component loadings are generally greater than the critical value of 0.3, indicating an interpretable correlation between the variables and components. However, the high cross-variable loadings of components 7-8 suggest that the stability of the factor structure needs further verification.

[0082] The component formulas for each factor are as follows: ; ; ; ; ; ; ; .

[0083] By substituting the target indicator data corresponding to each business item into the component formulas corresponding to each principal factor, the score value corresponding to each principal factor for each business item can be obtained.

[0084] Then, the weights corresponding to each principal factor can be used as follows (as shown in the formula). The scores are weighted and calculated to obtain the comprehensive score F for each business line. ; In some embodiments, the weight corresponding to each principal factor can be the ratio of its variance contribution rate in the variance explanation table to the sum of the variance contribution rate weights corresponding to each principal factor.

[0085] After obtaining the comprehensive score for each business line, the comprehensive scores can be sorted in descending order to obtain the ranking of comprehensive scores as shown in Table 5 below: Table 5

[0086] Based on the comprehensive score for each of the above business lines, the management capabilities of each business line can be evaluated.

[0087] The following combination Figure 5 The above management capability assessment process will be explained in detail with specific examples. This process includes: Step S501: Obtain the original datasets corresponding to the multiple business lines to be evaluated; each original dataset includes: indicator data corresponding to the corresponding business line and multiple evaluation indicators respectively; Step S502: Standardize the indicator data corresponding to each evaluation indicator based on the Z-score algorithm; Step S503: Input the indicator data of multiple business lines corresponding to each evaluation indicator into the K-nearest neighbor model, and filter out the target indicator data corresponding to the evaluation indicator based on the K-nearest neighbor model. Step S504: Obtain the target indicator data corresponding to each business line, construct the target data, and determine the target evaluation indicators corresponding to the business line based on this. Step S505: Perform KMO detection and Bartlett test on the target datasets corresponding to each business line to obtain the first test result and the second test result. Step S506: Determine whether the first detection result and the second detection result meet the preset conditions; if yes, proceed to step S507; otherwise, proceed to step S509. Step S507: Perform factor analysis on the target datasets corresponding to each business line to obtain multiple principal factors, the weights of each principal factor, and the loading coefficients of each principal factor with each target evaluation index. Step S508: Based on the target indicator data corresponding to the business line, the loading coefficient and weight of each main factor, determine the comprehensive score value corresponding to the business line. Step S509: End the process.

[0088] Based on the same inventive concept, this application also provides a management capability assessment device, see [link to relevant documentation]. Figure 6 The device includes: The data acquisition module 601 is used to acquire the original datasets corresponding to multiple business lines to be evaluated in response to the management capability assessment instruction; each original dataset includes: the indicator data corresponding to the corresponding business line and multiple evaluation indicator items respectively; The data filtering module 602 is used to filter out the target indicator data corresponding to the above evaluation indicator from multiple indicator data based on the K-nearest neighbor model for each evaluation indicator, taking the indicator data of each business line corresponding to the above evaluation indicator as input. The indicator determination module 603 is used to obtain the target dataset corresponding to each business line, and determine the evaluation indicators corresponding to each target indicator data in the target dataset as the target evaluation indicators corresponding to the business line; the target dataset includes: multiple target indicator data corresponding to the corresponding business line; The scoring determination module 604 is used to perform factor analysis on the target datasets corresponding to each business line, and determine the comprehensive score value corresponding to each business line based on the factor analysis results.

[0089] In one possible implementation, the parameters of the K-nearest neighbor model include reference values ​​for each evaluation index and the value of K; the data filtering module 602 is specifically used for: Calculate the distance between the data of each of the above indicators and the reference values ​​of the above evaluation indicators respectively; Based on the distances corresponding to the above multiple indicator data, select K indicator data as target indicator data corresponding to the above evaluation indicators.

[0090] In one possible implementation, the aforementioned K-nearest neighbor model is obtained by iteratively training a base model based on the original datasets and comprehensive score values ​​corresponding to multiple business lines in multiple capability management assessment operations prior to the current capability management assessment operation; wherein, the following operations are performed during one iteration of training: Obtain the original datasets and comprehensive scores for multiple business lines in the current capability management assessment operation, and filter the target business line from the multiple business lines based on the comprehensive scores; Determine the reference outputs for each target business line; The original datasets corresponding to each target business line are input into the K-nearest neighbor model obtained after the previous training iteration, with the reference output corresponding to each target business line as the objective, and the K-nearest neighbor model obtained after the previous training iteration is trained.

[0091] In one possible implementation, the data filtering module 602 described above is specifically used for: For each evaluation indicator, calculate the average value and standard deviation of the indicator data corresponding to the above evaluation indicators for each business line; For each indicator data corresponding to each business line, the above indicator data is standardized based on the average value and standard deviation of the evaluation indicators corresponding to the above indicator data. Based on the K-nearest neighbor model, target indicator data corresponding to the above evaluation indicators are selected from multiple standardized indicator data.

[0092] In one possible implementation, the scoring determination module 604 is specifically used for: A sampling suitability test is performed on the target datasets corresponding to each business line to obtain the first detection result; Bartlett's test of sphericity was performed on the target datasets corresponding to each business line to obtain the second detection result; The first and second test results are determined to meet the preset conditions.

[0093] In one possible implementation, the scoring determination module 604 is specifically used for: Factor analysis was performed on the target datasets corresponding to each business line to obtain multiple principal factors and the loading coefficients of each principal factor and each target evaluation index. For each business line, the comprehensive score value is determined based on the target indicator data corresponding to the business line and the loading coefficient of each main factor.

[0094] In one possible implementation, the scoring determination module 604 is specifically used for: Based on the target indicator data corresponding to the above business lines and the loading coefficients of each principal factor, the score value corresponding to each principal factor is determined. The comprehensive score for the above business lines is determined by weighted calculation based on the weight and score value of each main factor. The aforementioned weights are determined based on the variance contribution rate of the corresponding principal factors during factor analysis.

[0095] Based on the same inventive concept, this application also provides an electronic device that can realize the function of the aforementioned management capability assessment device. (Refer to...) Figure 7 The electronic device includes: At least one processor 701 and a memory 702 connected to at least one processor 701. In this embodiment, the specific connection medium between the processor 701 and the memory 702 is not limited. Figure 7 The example shown is the connection between processor 701 and memory 702 via bus 700. Bus 700 is... Figure 7The connections between other components are indicated by thick lines and are for illustrative purposes only, not as limiting information. The 700 bus can be divided into address bus, data bus, control bus, etc., for ease of representation. Figure 7 The term is represented by a single thick line, but this does not imply that there is only one bus or one type of bus. Alternatively, the processor 701 can also be called a controller; there is no restriction on the name.

[0096] In this embodiment, memory 702 stores instructions executable by at least one processor 701. By executing the instructions stored in memory 702, at least one processor 701 can perform the management capability assessment method described above. Processor 701 can implement... Figure 6 The functions of each module in the device shown.

[0097] The processor 701 is the control center of the device. It can connect to various parts of the control device through various interfaces and lines. By running or executing instructions stored in memory 702 and calling data stored in memory 702, the processor can perform various functions and process data, thereby monitoring the device as a whole.

[0098] In one possible design, processor 701 may include one or more processing units. Processor 701 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may also not be integrated into processor 701. In some embodiments, processor 701 and memory 702 may be implemented on the same chip; in some embodiments, they may also be implemented on separate chips.

[0099] The processor 701 can be a general-purpose processor, such as a central processing unit (CPU), digital signal processor, application-specific integrated circuit, field-programmable gate array or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the management capability assessment method disclosed in the embodiments of this application can be directly manifested as execution by a hardware processor, or execution by a combination of hardware and software modules within the processor.

[0100] Memory 702, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. Memory 702 may include at least one type of storage medium, such as flash memory, hard disk, multimedia card, card-type memory, random access memory (RAM), static random access memory (SRAM), programmable read-only memory (PROM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), magnetic storage, magnetic disk, optical disk, etc. Memory 702 can be any other medium capable of carrying or storing desired program code in the form of instructions or data structures that can be accessed by a computer, but is not limited thereto. In the embodiments of this application, memory 702 can also be a circuit or any other device capable of implementing storage functions for storing program instructions and / or data.

[0101] By designing and programming the processor 701, the code corresponding to the application anomaly detection method described in the foregoing embodiments can be embedded into the chip, thereby enabling the chip to execute the code during runtime. Figure 1 The steps of the management capability assessment method in the illustrated embodiment are as follows. How to design and program the processor 701 is a technique well-known to those skilled in the art and will not be described further here.

[0102] This application also provides a computer-readable storage medium storing computer-executable instructions required to execute the processor, including a program required to execute the processor.

[0103] In some possible implementations, various aspects of the management capability assessment method provided in this application can also be implemented in the form of a program product, which includes program code. When the program product is run on an electronic device, the program code is used to cause the electronic device to perform the steps in the management capability assessment method according to the various exemplary embodiments of this application described above.

[0104] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0105] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (devices), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0106] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0107] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0108] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0109] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A management capability assessment method, characterized in that, The method includes: In response to the management capability assessment instructions, obtain the original datasets corresponding to the multiple business lines to be assessed; each original dataset includes: the indicator data corresponding to the corresponding business line and multiple evaluation indicators respectively; For each evaluation indicator, the indicator data corresponding to the evaluation indicator for each business line is used as input, and the target indicator data corresponding to the evaluation indicator is selected from multiple indicator data based on the K-nearest neighbor model. For each business line, a target dataset corresponding to the business line is obtained, and the evaluation indicators corresponding to each target indicator data in the target dataset are determined as the target evaluation indicators corresponding to the business line; the target dataset includes: multiple target indicator data corresponding to the corresponding business line; Factor analysis was performed on the target datasets corresponding to each business line, and the comprehensive score value corresponding to each business line was determined based on the factor analysis results.

2. The method according to claim 1, characterized in that, The parameters corresponding to the K-nearest neighbor model include reference values ​​for each evaluation index and the value of K; the step of filtering target index data corresponding to the evaluation index from multiple index data based on the K-nearest neighbor model includes: Calculate the distance between each of the multiple indicator data and the reference value of the evaluation indicator; Based on the distances corresponding to the multiple indicator data, K indicator data are selected from the multiple indicator data as target indicator data corresponding to the evaluation indicator.

3. The method according to claim 2, characterized in that, The K-nearest neighbor model is obtained by iteratively training a basic model based on the original datasets and comprehensive score values ​​corresponding to multiple business lines in multiple capability management assessment operations prior to the current capability management assessment operation. During one iteration of training, the following operations are performed: In the current capability management assessment operation, obtain the original datasets and comprehensive score values ​​corresponding to multiple business lines respectively, and filter the target business line from the multiple business lines based on the comprehensive score values; Determine the reference outputs for each target business line; The original datasets corresponding to each target business line are input into the K-nearest neighbor model obtained after the previous training cycle, with the reference output corresponding to each target business line as the target, and the K-nearest neighbor model obtained after the previous training cycle is trained.

4. The method according to any one of claims 1 to 3, characterized in that, Before selecting the target indicator data corresponding to the evaluation indicator from multiple indicator data based on the K-nearest neighbor model, the method further includes: For each evaluation indicator, calculate the average value and standard deviation of the indicator data corresponding to each business line and the evaluation indicator; For each indicator data corresponding to each business line, the indicator data is standardized based on the indicator data and the average and standard deviation of the evaluation indicators corresponding to the indicator data; The step of selecting target indicator data corresponding to the evaluation indicator from multiple indicator data based on the K-nearest neighbor model includes: The target indicator data corresponding to the evaluation indicator is selected from multiple standardized indicator data based on the K-nearest neighbor model.

5. The method according to claim 1, characterized in that, Before performing factor analysis on the target datasets corresponding to each business line, the method further includes: A sampling suitability test is performed on the target datasets corresponding to each business line to obtain the first detection result; Bartlett's test of sphericity was performed on the target datasets corresponding to each business line to obtain the second detection result; The first and second detection results are determined to meet the preset conditions.

6. The method according to claim 1 or 5, characterized in that, The step of performing factor analysis on the target datasets corresponding to each business line, and determining the comprehensive score value corresponding to each business line based on the factor analysis results, includes: Factor analysis was performed on the target datasets corresponding to each business line to obtain multiple principal factors and the loading coefficients of each principal factor and each target evaluation index. For each business line, a comprehensive score value is determined based on the target indicator data corresponding to the business line and the loading coefficient of each main factor.

7. The method according to claim 6, characterized in that, The determination of the comprehensive score value corresponding to the business line based on the target indicator data corresponding to the business line and the loading coefficient corresponding to each main factor includes: Based on the target indicator data corresponding to the business line and the loading coefficient of each principal factor, the score value corresponding to each principal factor is determined. The comprehensive score value corresponding to the business line is determined by weighted calculation based on the weight and score value of each main factor. The weights are determined based on the variance contribution rate of the corresponding principal factors during factor analysis.

8. A management capability assessment device, characterized in that, The device includes: The data acquisition module is used to acquire the original datasets corresponding to multiple business lines to be evaluated in response to the management capability assessment instructions; each original dataset includes: the indicator data corresponding to the corresponding business line and multiple evaluation indicator items respectively; The data filtering module is used to filter out the target indicator data corresponding to the evaluation indicator from multiple indicator data based on the K-nearest neighbor model for each evaluation indicator, taking the indicator data corresponding to the evaluation indicator of each business line as input. The indicator determination module is used to obtain the target dataset corresponding to each business line, and determine the evaluation indicators corresponding to each target indicator data in the target dataset as the target evaluation indicators corresponding to the business line; the target dataset includes: multiple target indicator data corresponding to the corresponding business line; The scoring determination module is used to perform factor analysis on the target datasets corresponding to each business line, and determine the comprehensive score value corresponding to each business line based on the factor analysis results.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, when executing a computer program stored in the memory, implements the method of any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method described in any one of claims 1-7.

Images