Production adjustment method and device based on product score, equipment and medium

Abstract

The invention relates to the technical field of big data analysis, and discloses a production adjustment method and device based on product scoring, equipment and a medium, and the method comprises the steps: constructing a plurality of three-level index matrixes according to sample data and an index system of a target historical period, and determining effective three-level indexes and three-level weights thereof through employing a principal component analysis algorithm; and determining a basic index score of each effective third-level index of the target product through normalization processing, obtaining a comprehensive weight of each effective third-level index according to the first-level weight of each first-level index, the second-level weight of each second-level index and the third-level weight of each effective third-level index, and determining a comprehensive product score of the target product. According to the method, the comprehensive score of the target product is determined through the comprehensive weight and the index score of each effective three-level index, so that the competitiveness of the target product is comprehensively known, and the effective adjustment of the production plan is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of big data analysis, and in particular to a production adjustment method, device, equipment and medium based on product scoring. Background Art

[0002] As market competition becomes increasingly fierce, the competitiveness assessment of cigarette products has become a core part of corporate strategy formulation. However, traditional assessment methods are usually manual experience judgment and analysis methods, which use charts to empirically derive product competitiveness scores. This method is also limited by the shortcomings of single dimension, few indicators, and insufficient quantification. It is difficult to fully and accurately reflect the comprehensive competitiveness level of products, affecting the effectiveness of corporate decision-making, and thus making it difficult to adjust product production in a timely and effective manner. Summary of the invention

[0003] In view of this, the present invention provides a production adjustment method, device, computer equipment and storage medium based on product scoring to solve the current problem that it is difficult to fully and accurately reflect the comprehensive competitiveness level of products, resulting in the inability to adjust product production in a timely and effective manner.

[0004] In a first aspect, the present invention provides a production adjustment method based on product scoring, the method comprising:

[0005] According to the sample data of the target historical period and the preset indicator system, a three-level indicator matrix corresponding to each second-level indicator is constructed, wherein the sample data includes: quantitative information of each product under the third-level indicator; the preset indicator system includes: multiple first-level indicators, multiple second-level indicators and multiple third-level indicators, as well as the subordinate relationship between each indicator;

[0006] The principal component analysis algorithm is used to process each three-level indicator matrix to determine the effective three-level indicators in each three-level indicator matrix and the three-level weights of each effective three-level indicator;

[0007] Normalizing the sample data corresponding to the effective third-level indicators in each third-level indicator matrix to obtain the basic indicator score of each effective third-level indicator of the target product;

[0008] Determine the comprehensive weight of each valid third-level indicator based on the first-level weight of each first-level indicator, the second-level weight of each second-level indicator and the third-level weight of each valid third-level indicator;

[0009] Based on the comprehensive product score of the target product determined by the basic indicator scores and comprehensive weights of each effective third-level indicator of the target product, the production plan of the target product is adjusted.

[0010] The method provided in this aspect determines the three-level indicator matrix corresponding to each secondary indicator in the preset indicator system through sample data of the target historical period, and performs principal component analysis to determine the effective three-level indicators and corresponding weights in each three-level indicator matrix. At the same time, the basic indicator score of each effective three-level indicator is determined through normalization processing, so as to obtain the comprehensive product score of the target product by combining the weights corresponding to each first-level indicator and each second-level indicator in the indicator system and the weights corresponding to each effective three-level indicator. Then, the production plan of the target product is adjusted to ensure that the product score can accurately represent the competitiveness of the product, so as to achieve the effectiveness of the adjustment of the production plan.

[0011] In an optional implementation, the first-level weight of each first-level indicator is determined as follows:

[0012] Performing big data analysis on industry research data through a pre-trained first data analysis model to determine the correlation between the first-level indicators and construct a judgment matrix;

[0013] A hierarchical analysis is performed on the judgment matrix to determine the primary weight of each primary indicator.

[0014] In this implementation mode, the correlation between each indicator is determined through a data analysis model, thereby constructing a judgment matrix and performing a hierarchical analysis to scientifically and accurately set the weights of the first-level indicators and ensure the accuracy of subsequent comprehensive product score calculations.

[0015] In an optional implementation, the secondary weights of the secondary indicators are determined as follows:

[0016] The sales terminal data in the database is analyzed by a pre-trained second data analysis model to determine the membership of each secondary indicator under each primary indicator to the preset evaluation set and to construct a fuzzy relationship matrix corresponding to each primary indicator;

[0017] According to the fuzzy relationship matrix corresponding to each primary indicator, the secondary weights of each secondary indicator under each primary indicator are determined.

[0018] In this implementation mode, the degree of membership of the secondary indicators under each primary indicator to the preset evaluation set is determined through a data analysis model, thereby constructing a fuzzy relationship matrix, and then determining the secondary weights of each secondary indicator under each primary indicator, so as to scientifically and accurately set the secondary weights and ensure the accuracy of subsequent comprehensive product score calculations.

[0019] In an optional implementation, performing hierarchical analysis on the judgment matrix to determine the primary weight of each primary indicator includes:

[0020] Determine the maximum eigenvalue of the judgment matrix, determine the eigenvector corresponding to the maximum eigenvalue as the target eigenvector, and determine the weights to be confirmed corresponding to each primary indicator based on the target eigenvector;

[0021] Determine the consistency test result corresponding to the maximum eigenvalue and the target eigenvector, and judge whether the test result meets the preset threshold requirement;

[0022] If it meets the requirements, the weight to be confirmed is determined as the first-level weight;

[0023] If not, the maximum eigenvalue is updated, and the eigenvector corresponding to the maximum eigenvalue is determined as the target eigenvector, and based on the target eigenvector, the weights to be confirmed corresponding to each primary indicator are determined.

[0024] This implementation method performs a consistency check after preliminarily determining the first-level weights of each first-level indicator, and reconstructs the judgment matrix when the check results are unqualified, so as to ensure that the final first-level weights meet the preset requirements and ensure the scientific nature of the first-level weight setting.

[0025] In an optional implementation, the first-level indicators include: product brand, product performance and product sales environment;

[0026] The affiliation relationship between the first-level indicators and the second-level indicators is:

[0027] The secondary indicators under the product brand include: basic brand information and brand sales;

[0028] The secondary indicators under the product performance include: regional performance, competition duration;

[0029] The secondary indicators under the product sales environment include: sales terminal coverage and comprehensive situation of sales terminals.

[0030] This implementation method ensures the comprehensiveness of the preset indicator system by setting first-level indicators in the three aspects of brand, expressiveness and sales environment, and determining the second-level indicators under each first-level indicator from these three perspectives, so that the final comprehensive product score can fully reflect the comprehensive competitiveness of the product.

[0031] In an optional implementation, the use of a principal component analysis algorithm to process each three-level indicator matrix to determine the effective three-level indicators in each three-level indicator matrix and the three-level weights of each effective three-level indicator includes:

[0032] Determine the correlation coefficient matrix of each three-level indicator matrix, and calculate the variance contribution rate of each three-level indicator in each three-level indicator matrix according to the correlation coefficient matrix of each three-level indicator matrix;

[0033] The effective third-level indicators and the third-level weights of each effective third-level indicator are determined according to the variance contribution rate of each third-level indicator.

[0034] This implementation method calculates the variance contribution rate of each third-level indicator to screen out effective third-level indicators and determine their corresponding third-level weights, thereby ensuring the accuracy of third-level indicator screening and weight determination.

[0035] In an optional implementation, the method further includes:

[0036] After determining the correlation coefficient matrix of each three-level indicator matrix, perform a suitability test on each correlation coefficient matrix and judge whether the test result is qualified;

[0037] If qualified, the variance contribution rate of each third-level indicator in each qualified third-level indicator matrix is ​​determined according to the correlation coefficient matrix corresponding to each qualified third-level indicator matrix.

[0038] This implementation method performs a qualification test on the three-level indicator matrix. When the test result is qualified, the contribution rate of each subordinate three-level indicator is determined according to the corresponding three-level indicator matrix, so as to ensure that the three-level indicator matrix for principal component analysis is suitable for principal component analysis and ensure the validity of each valid three-level indicator and its weight determined by the principal component analysis.

[0039] In a second aspect, the present invention provides a production adjustment device based on product scoring, the device comprising:

[0040] A three-level indicator matrix construction module is used to construct a three-level indicator matrix corresponding to each secondary indicator according to sample data of the target historical period and a preset indicator system, wherein the sample data includes: quantitative information of each product under the three-level indicator; the preset indicator system includes: multiple first-level indicators, multiple second-level indicators and multiple third-level indicators, as well as the subordinate relationship between each indicator;

[0041] A three-level indicator weight determination module is used to process each three-level indicator matrix using a principal component analysis algorithm to determine the effective three-level indicators in each three-level indicator matrix and the three-level weights of each effective three-level indicator;

[0042] A basic indicator score determination module is used to normalize the sample data corresponding to the effective three-level indicators in each three-level indicator matrix to obtain the basic indicator score of each effective three-level indicator of the target product;

[0043] An indicator comprehensive weight determination module is used to determine the comprehensive weight of each valid third-level indicator according to the first-level weight of each first-level indicator, the second-level weight of each second-level indicator and the third-level weight of each valid third-level indicator;

[0044] The product production plan adjustment module is used to adjust the production plan of the target product based on the comprehensive product score of the target product determined by the basic indicator scores of each effective third-level indicator of the target product and the comprehensive weight.

[0045] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the production adjustment method based on product scoring of the above-mentioned first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0046] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the production adjustment method based on product scoring according to the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1 is a flow chart of a production adjustment method based on product scoring according to an embodiment of the present invention;

[0049] Figure 2 is a flow chart of another production adjustment method based on product scoring according to an embodiment of the present invention;

[0050] Figure 3 is an example diagram of a cigarette competitiveness index system according to an embodiment of the present invention;

[0051] Figure 4 This is a schematic diagram of content splitting of an indicator system according to an embodiment of the present invention;

[0052] Figure 5 This is an example diagram of a three-level indicator weight determination process according to an embodiment of the present invention;

[0053] Figure 6 This is an example diagram of a first-level weight allocation process according to an embodiment of the present invention.

[0054] Figure 7 is a structural block diagram of a production adjustment device based on product scoring according to an embodiment of the present invention;

[0055] Figure 8 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0057] As market competition becomes increasingly fierce, the competitiveness assessment of cigarette products has become a core part of corporate strategy formulation. However, traditional assessment methods are usually manual experience judgment and analysis methods, which use charts to empirically derive product competitiveness scores. This method is also limited by the shortcomings of single dimension, few indicators, and insufficient quantification. It is difficult to fully and accurately reflect the comprehensive competitiveness level of products, affecting the effectiveness of corporate decision-making, and thus making it difficult to adjust product production in a timely and effective manner.

[0058] To this end, an embodiment of the present invention provides a production adjustment method based on product scoring, which determines the three-level indicator matrix corresponding to each secondary indicator in a preset indicator system through sample data of the target historical period, and performs principal component analysis to determine the effective three-level indicators and corresponding weights in each three-level indicator matrix. At the same time, the basic indicator score of each effective three-level indicator is determined through normalization processing, so as to obtain the comprehensive product score of the target product by combining the weights corresponding to each first-level indicator in the indicator system, the weights corresponding to each second-level indicator, and the weights of each effective three-level indicator. Then, the production plan of the target product is adjusted to ensure that the product score can accurately represent the competitiveness of the product, so as to achieve the effectiveness of the production plan adjustment.

[0059] According to an embodiment of the present invention, an embodiment of a production adjustment method based on product scoring is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0060] In this embodiment, a production adjustment method based on product scoring is provided, which can be used for adjusting the production plan of the above-mentioned cigarette products. Figure 1 is a flow chart of a production adjustment method based on product scoring according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0061] Step S101, based on the sample data of the target historical period and the preset indicator system, construct a three-level indicator matrix corresponding to each second-level indicator. The sample data includes: quantitative information of each product under the three-level indicator; the preset indicator system includes: multiple first-level indicators, multiple second-level indicators and multiple third-level indicators, as well as the affiliation between each indicator.

[0062] The market competitiveness of a product is often jointly determined by the competitiveness in multiple directions, and the competitiveness in each direction will be affected by the corresponding indicators in multiple dimensions. Therefore, through in-depth analysis of a large amount of market data, business operation data, and feedback data from past project practices, the internal correlation and logical relationship between various business indicators can be sorted out. At the same time, combined with a detailed study of many successful cases and mature practice models in the industry, the subdivided subordinate indicators can be further explored and determined, thereby obtaining the final indicator system, that is, the preset indicator system, which includes multiple first-level indicators to characterize the competitiveness of products in different directions. For each first-level indicator, based on the affiliation, multiple second-level indicators under it can be determined, and multiple third-level indicators under each second-level indicator can be determined. The second-level indicators are used to characterize the indicator information of different dimensions under the corresponding direction of a certain first-level indicator, and the third-level indicators are used to characterize the indicator information of different dimensions under a certain second-level indicator.

[0063] For example, for cigarette products, a first-level indicator used to characterize product competitiveness may be the product brand, and its subordinate second-level indicators may be basic brand information and brand sales information. For the basic brand information, its subordinate third-level indicators may be specific indicators such as concentration, innovation capability, and brand coverage.

[0064] For the sample data of the target historical period, it can be the closest statistical period, such as the characterization data of multiple products for each three-level indicator in the past quarter, where the multiple products include: target products, that is, the products for which the comprehensive product score is ultimately calculated, and other products, which can be other brands under the cigarette factory and brands under other rival cigarette factories. For cigarette manufacturers, their internal databases will count the multi-dimensional data corresponding to their own various brand products and various brand products of competitors, that is, the data corresponding to the three-level indicators mentioned above.

[0065] A third-level indicator matrix corresponding to each secondary indicator is constructed through these data. For each third-level indicator matrix, the column corresponds to the data information of a certain third-level indicator corresponding to multiple products, and the row corresponds to the data information of each third-level indicator under a certain secondary indicator corresponding to a certain product.

[0066] For example, for each three-level indicator matrix, each row corresponds to the data information of multiple three-level indicators corresponding to a product in the target historical period. For example, a second-level indicator A has three-level indicators under it: A-1, A-2, A-3. Then the row vector in the three-level indicator matrix corresponding to the second-level indicator A is the data corresponding to the three indicators A-1, A-2, A-3 of a certain product, and so on.

[0067] Step S102, using the principal component analysis algorithm to process each three-level indicator matrix, to determine the effective three-level indicators in each three-level indicator matrix and the three-level weights of each effective three-level indicator.

[0068] For each three-level indicator matrix, the principal component analysis method can be used to determine the effective three-level indicators and the corresponding three-level weights in the three-level indicator matrix. When performing principal component analysis, the information concentration principle of the data is used, and the variance explanation rate is used to calculate the weights to obtain the coefficients and weights of each indicator. For each set of three-level indicators under each secondary indicator, the principal component analysis method is used once to determine the indicator weights based on the correlation between the data.

[0069] Specifically, when principal component analysis is performed on a group of third-level indicators under a second-level indicator, the original data of the third-level indicators is processed using the maximum-minimum normalization method. Let the original data matrix be X, where each row represents a sample and each column represents a third-level indicator. For the jth column of data x ij (i represents the sample number), the normalized data y ij The calculation formula is:

[0070]

[0071] Get the normalized data matrix Y.

[0072] Next, determine the correlation coefficient matrix R of the data matrix Y, where the element r ij It represents the correlation coefficient between the data in the i-th column and the j-th column in Y. When determining the correlation coefficient matrix, the historical data can be mined through the data mining model to determine the correlation coefficients between different indicators, or the correlation analysis can be directly performed on the above data matrix Y to determine the correlation coefficients between different indicators and obtain the correlation coefficient matrix R.

[0073] Then calculate the eigenvalue λ of the correlation coefficient matrix i and the eigenvector The subscript i indicates the eigenvalue and eigenvector corresponding to a certain level 3 indicator in the level 3 indicator matrix. Calculate the variance contribution rate of each principal component: Finally, draw the cumulative variance contribution rate graph, add up the variance contribution rates from high to low, select the first n principal components with cumulative variance contribution rates exceeding 95%, that is, the first n third-level indicators, and use these first n third-level indicators as valid third-level indicators. At the same time, normalize the variance contribution rates of these first n third-level indicators as the third-level weights of each valid third-level indicator.

[0074] By analogy, the above steps are performed on the third-level indicator matrix corresponding to each second-level indicator to obtain the effective third-level indicators and corresponding third-level weights under each second-level indicator.

[0075] Step S103, normalizing the sample data corresponding to the effective three-level indicators in each three-level indicator matrix to obtain the basic indicator score of each effective three-level indicator of the target product.

[0076] For each three-level indicator matrix, the elements corresponding to each row are the data information of multiple three-level indicators corresponding to a certain product. For example, a certain second-level indicator is basic brand information, and its subordinate third-level indicators are: concentration (the number of gears covered by the brand), innovation ability (the number of products under the brand), brand coverage (the number of cities covered by the brand) and other third-level indicators. The elements of each row in the three-level indicator matrix are the data information of a certain product for each indicator. The order of magnitude and units of this information are different. Therefore, for an effective third-level indicator, the data information corresponding to multiple products needs to be normalized.

[0077] At the same time, the three-level indicator matrix is ​​initially screened in combination with the effective three-level indicators determined above, and the values ​​corresponding to the ineffective three-level indicators are eliminated. The three-level indicator matrix after the elimination is normalized. For the specific normalization steps, reference can be made to the relevant content in the above step S102, except that in step S102, all the three-level indicators under a certain second-level indicator are normalized to obtain scores of multiple products for each effective three-level indicator, from which the scores of the target products for each effective three-level indicator, i.e., the basic indicator scores, are selected.

[0078] For example, for a certain secondary indicator, the subordinate tertiary indicators are: concentration (number of gears covered by the brand), innovation ability (number of products under the brand), brand coverage (number of cities covered by the brand), time on the market, and time in the market; the total number of products is 6, including 1 target product and 5 other products, so the tertiary indicator matrix is ​​a 6-row 5-column matrix; assuming that the effective tertiary indicators are concentration, innovation ability, and brand coverage, the matrix after screening the effective tertiary indicators is a 6-row 3-column matrix, assuming that the elements of the column vector corresponding to the first column are: {20000, 15000, 16000, 10000, 30000, 24000}, indicating the concentration corresponding to product 1, that is, the number of gears covered by the product is 10,000, and so on. After normalization, the scores of each product for concentration are: {5, 2.5, 3, 0, 10, 7} (out of 10). Product 1 is the target product, and its basic indicator score for the effective indicator of concentration is 5 points. Similarly, the above steps are processed for the third-level indicator matrix corresponding to each second-level indicator to obtain the basic indicator score of each effective third-level indicator.

[0079] Step S104, determining the comprehensive weight of each valid third-level indicator according to the first-level weight of each first-level indicator, the second-level weight of each second-level indicator and the third-level weight of each valid third-level indicator.

[0080] For each first-level indicator and the second-level indicators under each first-level indicator, the corresponding weights can be determined in advance through data mining algorithms, that is, the first-level weight of each first-level indicator and the second-level weight of its subordinate second-level indicators. At the same time, the third-level weights of the effective third-level indicators under each second-level indicator are combined to calculate the comprehensive weight of each effective third-level indicator.

[0081] For example, assuming that the first-level indicators are: A, B, C, the corresponding first-level weights are: 0.2, 0.3, 0.5; the second-level indicators under the first-level indicator A are: A-1, A-2, A-3, A-4, and the corresponding second-level weights are: 0.1, 0.2, 0.3, 0.4; the second-level indicator A-1, the subordinate valid third-level indicators are: A-1-a, A-1-b, A-1-c, and the corresponding third-level weights are: 0.3, 0.3, 0.4. For the valid third-level indicator A-1-a, its corresponding comprehensive third-level weight is: 0.2*0.1*0.3.

[0082] Step S105, adjusting the production plan of the target product based on the comprehensive product score of the target product determined by the basic indicator scores of each valid third-level indicator of the target product and the comprehensive weight.

[0083] After determining the comprehensive three-level weights of each effective three-level indicator, combined with the corresponding basic indicator scores, the comprehensive product scores of each effective three-level indicator can be obtained. After obtaining the comprehensive product score of the target product, cigarette manufacturers can understand their product competitiveness in the entire cigarette market based on the comprehensive product score of their own products, and thus adjust subsequent production plans. If the product competitiveness represented by the comprehensive product score is strong, the production scale can be expanded; if the product competitiveness represented is weak, the production scale can be appropriately reduced, and at the same time, the reasons for the competitiveness disclosure can be further analyzed in depth to improve product competitiveness.

[0084] Furthermore, based on the above method, a competitiveness analysis can be conducted on the brands of multiple cigarette manufacturers on the market based on the market data collected by cigarette manufacturers, so as to understand the comprehensive product scores corresponding to the product competitiveness of each manufacturer in the current market, and generate corresponding statistical reports and perform corresponding visualization processing in combination with the grades corresponding to each product, and generate intuitive visualization analysis reports to understand the advantages and disadvantages of products in the current market, so as to adjust the production plans of the corresponding products of each brand globally.

[0085] The production adjustment method based on product scoring provided in this embodiment determines the three-level indicator matrix corresponding to each secondary indicator in the preset indicator system through sample data of the target historical period, and performs principal component analysis to determine the effective three-level indicators and corresponding weights in each three-level indicator matrix. At the same time, the basic indicator score of each effective three-level indicator is determined through normalization processing, so as to obtain the comprehensive product score of the target product by combining the comprehensive weights of each effective three-level indicator determined by the weights corresponding to each first-level indicator and each second-level indicator in the indicator system and the weights corresponding to each effective three-level indicator, and then adjust the production plan of the target product, thereby ensuring that the product score can accurately represent the competitiveness of the product to achieve the effectiveness of the production plan adjustment.

[0086] According to an embodiment of the present invention, another embodiment of a production adjustment method based on product scoring is provided, which can be used for adjusting the production plan of the above-mentioned cigarette products. Figure 2 is a flow chart of another production adjustment method based on product scoring according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0087] Step S201, based on the sample data of the target historical period and the preset indicator system, construct a three-level indicator matrix corresponding to each second-level indicator. The sample data includes: quantitative information of each product under the three-level indicator; the preset indicator system includes: multiple first-level indicators, multiple second-level indicators and multiple third-level indicators, as well as the affiliation between each indicator.

[0088] Specifically, the first-level indicators include: product brand, product performance and product sales environment;

[0089] The relationship between the primary and secondary indicators is:

[0090] Secondary indicators under product brands include: basic brand information, brand sales;

[0091] Secondary indicators under product performance include: regional performance, competition duration;

[0092] The secondary indicators under the product sales environment include: sales terminal coverage and comprehensive situation of sales terminals.

[0093] It can be understood that cigarette manufacturers can use professional data collection and analysis tools to widely collect data related to tobacco product production, sales, market feedback and other aspects, use data analysis models to integrate and process these data, refer to the existing mature evaluation models of similar products and related theoretical research results, and formulate an indicator framework for the tobacco product competitiveness indicator system, such as Figure 3 , which is an example diagram of a cigarette competitiveness index system according to an embodiment of the present invention.

[0094] refer to Figure 3 As shown, it can be seen that the first-level indicators are indicators of three dimensions: product brand, product performance, and product sales environment. For these three first-level indicators, their respective second-level indicators are basic brand situation, brand sales situation; regional performance, competition duration; sales terminal coverage, sales terminal comprehensive situation.

[0095] For example, the third-level indicators under each second-level indicator can be referred to Figure 4 , is a schematic diagram of the content splitting of an indicator system according to an embodiment of the present invention. The corresponding meanings of each third-level indicator are given, and by statistically analyzing the relevant data in the target historical period, the specific data of multiple products for these third-level indicators are obtained, and then the third-level indicator matrix corresponding to each second-level indicator is constructed.

[0096] Exemplarily, the data information of each product for each third-level indicator, that is, the data sources used to construct the three-level indicator matrix may include: Sales data system: access to the company's internal sales data system to obtain sales data, market share and other sales data in real time. Consumer big data application: use big data analysis tools to mine consumer behavior data and identify market trends and potential opportunities for cigarette products. The key data include but are not limited to: Sales: total sales in each time period, reflecting market coverage. Market share: the proportion of brand products in the market, reflecting brand status. Consumer evaluation: collected through questionnaires, online evaluations, etc., to evaluate brand reputation. Competitive product sales: sales, market share and other data of major competitors, used for comparative analysis.

[0097] Exemplarily, on the basis of the indicator system, the indicator matrix is ​​determined. In order to better measure the relationship between indicators, the study selects the appropriate multi-indicator decision-making method according to the needs and characteristics of different levels, that is, uses the correlation combination method of multi-level matrices to improve the accuracy and effectiveness of multi-indicator decision-making.

[0098] The research is divided into three main aspects (first-level indicators). The three first-level indicators can be divided into six second-level indicators, and the six second-level indicators can be divided into 24 third-level indicators.

[0099] The set of primary indicators is: U = {brand competitiveness u1, product performance competitiveness u2, sales environment competitiveness u3}

[0100] The secondary indicators under each primary indicator constitute corresponding subsets. For example, the secondary indicators under brand competitiveness u1 constitute the set U1 = {u 11 ,u 12 ,…}, the third-level indicators under each second-level indicator constitute a more detailed subset.

[0101] Step S202, using the principal component analysis algorithm to process each three-level indicator matrix, to determine the effective three-level indicators in each three-level indicator matrix and the three-level weights of each effective three-level indicator.

[0102] Specifically, in step S202, it includes:

[0103] Step S202-1, determining the correlation coefficient matrix of each three-level indicator matrix, and calculating the variance contribution rate of each three-level indicator in each three-level indicator matrix according to the correlation coefficient matrix of each three-level indicator matrix.

[0104] The elements in the correlation coefficient matrix are the relative correlations between the indicators. Assume that the correlation coefficient matrix R of the third-level indicator matrix Y corresponding to a second-level indicator has an element r ij Represents the correlation coefficient between the data in the i-th column and the j-th column in Y.

[0105] The correlation coefficient can be determined by analyzing the correlation coefficient between each indicator through a trained data mining model, or by directly calculating the covariance matrix of the three-level indicator matrix as the correlation coefficient matrix. The three-level indicator matrix can also be standardized to eliminate the influence of the dimension and order of magnitude of the variables corresponding to different indicators, and then the correlation coefficient between different indicators can be determined through the conventional correlation coefficient calculation formula.

[0106] The eigenvalues ​​of the correlation coefficient matrix are calculated to obtain the eigenvalues ​​of the columns corresponding to each indicator and the eigenvectors corresponding to each eigenvalue. The eigenvalues ​​of the columns corresponding to each indicator are normalized to obtain the variance contribution rate corresponding to each eigenvalue.

[0107] Furthermore, after determining the correlation coefficient matrix of each three-level indicator matrix, a suitability test is performed on each correlation coefficient matrix, and it is determined whether the test result is qualified;

[0108] If qualified, the variance contribution rate of each third-level indicator in the corresponding three-level indicator matrix is ​​calculated according to the correlation coefficient matrix.

[0109] It can be understood that not all three-level indicator matrices are suitable for determining weights through principal component analysis. Before performing the subsequent principal component analysis, it is necessary to perform a suitability test on the correlation coefficient matrix corresponding to each three-level indicator matrix. If qualified, the variance contribution rate of each three-level indicator in the corresponding three-level indicator matrix is ​​calculated according to the correlation coefficient matrix, that is, the step corresponding to the variance contribution rate in the above step S202-1 is executed. If qualified, it means that the corresponding three-level indicator matrix is ​​not suitable for determining the effective three-level indicators and their corresponding weights through principal component analysis. Each of the three-level indicators can be directly regarded as an effective three-level indicator, and the weight of each effective three-level indicator can be determined according to the mean or experience to set a preset value.

[0110] Specifically, after obtaining the correlation coefficient matrix R, the element r ij Represents the correlation coefficient between the data in the i-th column and the j-th column in Y.

[0111] When conducting a suitability test, the specific contents are as follows:

[0112] Calculate the KMO value and the p-value of Bartlett's test of sphericity.

[0113] The KMO value calculation formula is:

[0114]

[0115] in is the partial correlation coefficient.

[0116] Perform Bartlett's sphericity test and calculate the test statistic X 2 , the calculation formula is:

[0117]

[0118] Where n is the number of samples, p is the number of indicators, and λ i is the eigenvalue of the correlation coefficient matrix R. For example, for an exemplary data under a certain actual situation, the specific test results are shown in Table 1 below:

[0119] Table 1

[0120] project Numeric KMO test value 0.78 Bartlett's test of sphericity 309.77 Bartlett's test of sphericity p-value 0.00

[0121] When KMO>0.6 and p-value<0.05, the data are considered suitable for PCA. The specific numerical value can be set according to the actual situation and is not restricted here.

[0122] Step S202-2, determining the effective three-level indicators and the three-level weights of each effective three-level indicator according to the variance contribution rate of each three-level indicator.

[0123] The variance contribution rates are sorted from large to small and accumulated. The first n indicators whose accumulated values ​​are greater than a preset value, such as 95%, are determined as effective components, that is, effective third-level indicators. The variance contribution rates of these first n indicators are normalized again to obtain the corresponding variance contribution rates of each effective third-level indicator.

[0124] At this time, for these n effective three-level indicators, elements are eliminated in combination with the three-level indicator matrix to obtain the matrix corresponding to the effective three-level indicators. From the above related content, it can be seen that each row in the three-level indicator matrix corresponds to the score of a product under each indicator. The score of each product under each effective indicator is determined by the preset linear combination coefficient. The specific calculation method is as follows:

[0125] For the product corresponding to each row in the matrix, suppose the selected n principal components (effective three-level indicators) are F1, F2, ..., F n , for the kth principal component F k , its linear combination expression is:

[0126]

[0127] where a kj It is a preset linear combination coefficient, which can be set by the staff based on experience or obtained by analyzing historical data through a data analysis model, where p is the number of valid indicators.

[0128] For each sample (product) score on each principal component:

[0129]

[0130] (m is the number of samples), calculate the comprehensive score model coefficient of each indicator:

[0131]

[0132] where b k It is the principal component weight determined according to the variance contribution rate.

[0133] Finally, the coefficients of the comprehensive score model are normalized to obtain the weight vector of the third-level index under the second-level index:

[0134] For example, for each secondary indicator, the weight determination method of each subordinate tertiary indicator can refer to Figure 5 , is an example diagram of a three-level indicator weight determination process according to an embodiment of the present invention.

[0135] First, data preprocessing: Use the maximum-minimum normalization method to process the original data to eliminate the impact of dimension and ensure that the data direction is consistent with the business logic.

[0136] Secondly, suitability test: KMO and Bartlett sphericity test were performed. If KMO>0.6 and Bartlett p<0.05, the data were judged to be suitable for PCA analysis.

[0137] Next, principal component extraction: draw a cumulative variance contribution rate graph, select the first n principal components with a contribution rate exceeding 95% to achieve data dimensionality reduction.

[0138] Finally, weight calculation and normalization: Based on principal component analysis, calculate the coefficients of each principal component and build a comprehensive score model. Then, normalize the coefficients to ensure that the sum of the weights is 1, forming a scientific and reasonable weight distribution plan.

[0139] Step S203, normalize the sample data corresponding to the effective three-level indicators in each three-level indicator matrix to obtain the basic indicator score of each effective three-level indicator of the target product. Figure 1 Step S103 of the illustrated embodiment will not be described in detail here.

[0140] Step S204, determining the comprehensive weight of each valid third-level indicator according to the first-level weight of each first-level indicator, the second-level weight of each second-level indicator and the third-level weight of each valid third-level indicator.

[0141] Specifically, in step S204, the first-level weight of each first-level indicator is determined as follows:

[0142] Step S204-1-1, performing big data analysis on the industry research data through the pre-trained first data analysis model, determining the correlation between each primary indicator and constructing a judgment matrix;

[0143] Step S204-1-2, perform hierarchical analysis on the judgment matrix to determine the primary weight of each primary indicator.

[0144] It can be understood as pre-collecting and integrating the rich data resources gathered by the tobacco industry big data platform. These data cover multi-dimensional information of different brands and various products in the industry in different market areas and time periods (such as product sales volume, sales, market share changes, consumer evaluation tendencies, etc.), and then using special data mining algorithms and analysis algorithms to conduct in-depth mining and correlation analysis on these massive data. At the same time, refer to the indicator importance reference standards released by authoritative organizations in the industry based on big data analysis, and use the AHP hierarchical analysis method and FCE fuzzy comprehensive evaluation method to determine the indicator weights.

[0145] Specifically, the first-level weights of each first-level indicator are determined mainly through the AHP hierarchical analysis method. In the process of AHP hierarchical analysis, the judgment matrix plays a key role in determining the weights. It is used to compare the importance or priority between two factors. Each element in the matrix means the importance or priority score of the corresponding row factor relative to the column factor.

[0146] The specific operation of using AHP to determine the weight of the first-level indicators is as follows. For the three first-level indicators (brand competitiveness A, product performance competitiveness B, sales environment competitiveness C), a third-order judgment matrix φ is constructed. Matrix element a ij The data mining and analysis algorithm model, i.e. the pre-trained first data analysis model, is obtained after a comprehensive evaluation based on the correlation of various indicators in the industry big data, the contribution to the overall market competitiveness, etc., i.e. the industry research data, to reflect the relative importance between row factors and column factors, namely:

[0147]

[0148] Through comprehensive and detailed literature research, in-depth integration of the importance information of each indicator obtained from industry big data analysis and relevant reference standards, the final judgment matrix can be shown as follows:

[0149]

[0150] In this example, the element in the first row and second column is 1 / 2, which represents the factor a. 11 Relative to factor a 12The correlation is 1 / 2, and the meanings of the remaining numbers are similar. By processing the above judgment matrix through the hierarchical analysis method, the first-level weights corresponding to each first-level indicator can be obtained.

[0151] Specifically, a hierarchical analysis is performed on the judgment matrix to determine the first-level weights of each first-level indicator, including:

[0152] Step A: Determine the maximum eigenvalue of the judgment matrix, determine the eigenvector corresponding to the maximum eigenvalue as the target eigenvector, and based on the target eigenvector, determine the weights to be confirmed corresponding to each primary indicator.

[0153] Calculate the eigenvalues ​​and eigenvectors of the matrix, and take the eigenvector corresponding to the maximum eigenvalue as the weight vector, that is, the target eigenvector, symbolized by λ max .

[0154] For the first-level indicator judgment matrix φ: calculate the eigenvalue and eigenvector of φ, and the eigenvalue satisfies: where λ is the eigenvalue, is the eigenvector.

[0155] Extract the maximum eigenvalue λ max The corresponding eigenvector And normalize it to get the first-level indicator weight vector:

[0156]

[0157] Each element represents the weight of brand competitiveness, product performance competitiveness, and sales environment competitiveness. As the weights to be confirmed corresponding to each first-level indicator.

[0158] Step B: Determine the consistency test result corresponding to the maximum eigenvalue and the target eigenvector, and judge whether the test result meets the preset threshold requirement.

[0159] Specifically, according to the results obtained in the above steps, the consistency index (CI) and the consistency ratio (CR) are calculated.

[0160] CI is used to evaluate the internal consistency of the judgment matrix, and the calculation formula is:

[0161] where λ max is the largest eigenvalue and n is the order of the matrix.

[0162] CR evaluates the degree of consistency by comparing the consistency index with the random consistency index (Random Index, RI), and the calculation formula is:

[0163] RI is the random consistency index (Random Index, RI), and its value depends on the order n of the judgment matrix.

[0164] For common orders, the value of RI is predetermined and is obtained by averaging a large number of randomly generated consistency matrices.

[0165] In practical applications, when n=3, RI=0.58 is often taken.

[0166] Exemplarily, calculated through the above steps, the consistency index CI and consistency ratio CR of the matrix are 0.004 and 0.007 respectively, and the distance between them and 0 is less than the preset threshold. The matrix passes the consistency test, where the preset threshold can be set according to the actual situation and is not limited here.

[0167] Step C: If it meets the requirements, the weight to be confirmed is determined as the first-level weight.

[0168] That is, if the consistency check is passed, the weights will be confirmed.

[0169] Step D: If not, the maximum eigenvalue is updated, and the eigenvector corresponding to the maximum eigenvalue is determined as the target eigenvector, and based on the target eigenvector, the weights to be confirmed corresponding to each primary indicator are determined.

[0170] Specifically, when updating the maximum eigenvalue, the eigenvalues ​​can be sorted according to size, and the second largest eigenvalue is used as the updated maximum eigenvalue, and so on, and the eigenvector of the current updated maximum eigenvalue is determined as the target eigenvector, and the relevant contents in the above steps are re-executed.

[0171] Specifically, the method for determining the first-level weights corresponding to the above-mentioned first-level indicators can be referred to Figure 6 FIG. 1 is a diagram showing an example of a first-level weight allocation process according to an embodiment of the present invention.

[0172] First, construct a judgment matrix: construct a third-order judgment matrix for the three first-level indicators of brand competitiveness A, product performance competitiveness B, and sales environment competitiveness C. Based on judgment, reflect the relative importance of row and column factors.

[0173] Next, calculate the weights: perform eigenvalue decomposition on the judgment matrix, extract the eigenvector corresponding to the maximum eigenvalue as the weight vector, and determine the weights of each primary indicator based on the eigenvector.

[0174] Finally, consistency test: Use consistency index (CI) and consistency ratio (CR) to evaluate the rationality of the judgment matrix. If CR is less than 0.1, the consistency of the judgment matrix is ​​satisfactory and the weight distribution is reasonable.

[0175] Specifically, in step S204, the secondary weight of each secondary indicator is determined as follows:

[0176] Step S204-2-1, perform big data analysis on the sales terminal data in the database through the pre-trained second data analysis model, determine the membership of each secondary indicator under each primary indicator to the preset evaluation set and construct a fuzzy relationship matrix corresponding to each primary indicator.

[0177] Step S204-2-2, according to the fuzzy relationship matrix corresponding to each first-level indicator, determine the second-level weight of each second-level indicator under each first-level indicator.

[0178] For each secondary indicator, we integrate multi-source data such as market research data and consumer evaluation data, i.e. the sales terminal data mentioned above, and input it into a specially constructed data analysis model, i.e. the pre-trained second data analysis model. This model has been trained and learned based on a large number of verified data samples of the same type in the past, and can accurately identify the performance of each secondary indicator in different dimensions based on data characteristics. Through in-depth mining and analysis of data in each dimension, referring to the quantitative results output by the model, and in accordance with the set rules that are in line with industry logic and actual business conditions, we evaluate the sales terminal coverage and the sales terminal comprehensive scoring indicators at different evaluation levels.

[0179] Through the operation and analysis of the model, the membership degree r of each secondary indicator to each evaluation dimension is determined according to the quantitative value corresponding to each evaluation dimension in the preset evaluation set output by the model. ij , which means the degree of membership of the i-th third-level indicator to the j-th evaluation level.

[0180] The fuzzy relationship matrix is ​​constructed as follows:

[0181] (p is the number of secondary indicators, and the evaluation dimensions in the preset rating set are 5).

[0182] For example, for the sales terminal coverage and comprehensive score of sales terminals under the competitiveness of the sales environment, the evaluation set can be: (customer flow, customer satisfaction, market share), that is, the evaluation dimension is 3, and the corresponding weights of the evaluation set are set as: (0.3, 0.2, 0.5).

[0183] Using the fuzzy comprehensive evaluation method, according to the intrinsic correlation between the data of each dimension and its importance to the overall evaluation, these relevant information reflecting the degree of membership are weighted and comprehensively processed.

[0184] That is, multiply the fuzzy relationship matrix R by the weight corresponding to the evaluation set. Assuming that the fuzzy relationship matrix is ​​p×n, where p is the number of indicators and n is the evaluation dimension in the preset evaluation set, the evaluation set corresponds to an n×1 column vector. Multiply the two to get a p×1 column vector, and perform normalization to get the secondary weight corresponding to each secondary indicator. For example, in this case, through the above data and model-based processing flow, the sales terminal coverage and sales terminal comprehensive score weights calculated are 0.22 and 0.78 respectively.

[0185] Illustratively, in one example, through the above-mentioned method of determining the first-level weight corresponding to the first-level indicator and the second-level weight corresponding to the second-level indicator, the first-level and second-level indicator weight results can be obtained as shown in Table 2.

[0186] Table 2

[0187]

[0188] Step S204-3, after determining the primary weight of each primary indicator and the secondary weight of each secondary indicator, the third-level weight of each valid third-level indicator is combined to calculate the comprehensive weight of each valid third-level indicator.

[0189] For example, the weights of the first-level indicators determined by AHP are vector, secondary indicator weight vector determined by FCE And the three-level indicator weight vector determined by PCA Then the comprehensive weight vector is .

[0190] ( represents the corresponding weight transfer and integration operations).

[0191] Combination Figure 4 In the actual calculation process, some of the third-level indicators in the index system content splitting diagram will be eliminated in combination with the actual situation, that is, only the valid third-level indicators will be retained. For ease of understanding, it is assumed that all the third-level indicators are valid third-level indicators. For example, the relevant weights of the cigarette product competitiveness index system are finally obtained as shown in Table 3.

[0192] Table 3

[0193]

[0194]

[0195] For example, after obtaining the weights corresponding to the above-mentioned indicators, a cumulative multiplication calculation may be performed to obtain the competitiveness index of the target tobacco product.

[0196]

[0197] Among them, X j is the superior indicator, x i Indicates the basic indicator scores corresponding to the lower-level indicators decomposed from the upper-level indicators, ω i Represents the weight of the subordinate indicator.

[0198] For example, for the regional performance value under the product performance competitiveness, the calculation process of substituting the value is as follows:

[0199]

[0200] After calculating the values ​​of the regional performance and competition duration indicators under the product performance competitiveness, the product performance competitiveness value can be calculated accordingly:

[0201]

[0202] After calculating the brand competitiveness, product performance competitiveness, and sales environment competitiveness under comprehensive competitiveness, we can calculate the comprehensive competitiveness, which is the highest level indicator value. The calculation formula with the values ​​​​substituted is:

[0203]

[0204] In another calculation method, the above table can be normalized, and the resulting Table 3-1 is as follows:

[0205] Table 3-1

[0206]

[0207] The weight corresponding to each third-level indicator is the comprehensive weight. The basic indicator score and the comprehensive weight of each valid third-level indicator of the target product are weightedly calculated to obtain the comprehensive product score of the target product.

[0208] Step S205, adjusting the production plan of the target product based on the comprehensive product score of the target product determined by the basic indicator scores of each valid third-level indicator of the target product and the comprehensive weight.

[0209] Based on the comprehensive weights of the three-level indicators shown in Table 3-1 above, the comprehensive weights corresponding to the three-level indicators are determined, and weighted calculations are performed in combination with specific indicator values ​​to obtain the comprehensive product score of a certain cigarette product, that is, the competitiveness score. By statistically calculating the basic indicator scores corresponding to the three-level indicators of each cigarette product in the market and multiplying them by the corresponding comprehensive weights, the competitiveness scores of each cigarette product in the market can be obtained, thereby assisting cigarette manufacturers in making corresponding decisions. Specifically, the corresponding production plan can be adjusted according to the comprehensive score of the target product.

[0210] The production adjustment method based on product scoring provided in the embodiment of the present invention determines the three-level indicator matrix corresponding to each secondary indicator in the preset indicator system through sample data of the target historical period, and performs principal component analysis to determine the effective three-level indicators and corresponding weights in each three-level indicator matrix. At the same time, the basic indicator score of each effective three-level indicator is determined through normalization processing, so as to obtain the comprehensive product score of the target product by combining the weights corresponding to each first-level indicator and the weights corresponding to each second-level indicator in the indicator system and the weights of each effective three-level indicator. Then, the production plan of the target product is adjusted, which ensures that the product score can accurately represent the competitiveness of the product to achieve the effectiveness of the production plan adjustment.

[0211] The above method implementation example is scientific and easy to implement. It builds an index system with reference to authoritative information from multiple sources to reflect product competitiveness, uses advanced technology to determine weights, uses multi-channel data and literature to mine key factors, and makes fine adjustments based on actual data to ensure that it fits the market, thereby helping companies formulate strategies and optimize product layouts.

[0212] Based on the above method embodiments, the present invention proposes a three-dimensional visualization evaluation system for evaluating the market competitiveness of cigarette products, which intuitively displays the competitiveness of each brand in a specific region and provides accurate data and decision-making basis for corporate market strategy formulation. It includes four core modules: data preprocessing, competitiveness calculation, three-dimensional visualization, and decision support. Each module works closely together to achieve comprehensive evaluation and visualization of competitiveness.

[0213] The data preprocessing module is used to collect market data of various brands of cigarette products in Guangxi, including but not limited to sales volume, sales revenue, sales terminal coverage, etc. The collected data is cleaned and standardized to ensure data quality. The maximum-minimum normalization method is used to normalize the scores of various primary indicators (brand competitiveness, product performance competitiveness, sales environment competitiveness) to the [0,1] range to improve the interpretability of the data.

[0214] The competitiveness calculation module is used to calculate the average score of the first-level indicators of all products under each brand, and obtain the average score of brand competitiveness, product performance competitiveness, and sales environment competitiveness. For each product of each brand, the weighted average of all indicators under the product is used as the brand's three-dimensional competitiveness value to ensure the comprehensiveness and accuracy of the evaluation results.

[0215] The three-dimensional visualization module is used to draw a three-dimensional competitiveness diagram of brands and products based on the data output by the competitiveness calculation module. The coordinate axes in the diagram represent brand competitiveness, product performance competitiveness, and sales environment competitiveness, respectively. The position of the points intuitively displays the performance of each brand and product in three dimensions. The size of the points represents the overall strength of the brand or the final competitiveness score of the product, which enhances the readability and information content of the diagram.

[0216] The decision support module is used to provide decision support for market strategy formulation for enterprises based on the results of 3D visualization. For products with weak sales environment competitiveness, it puts forward suggestions such as optimizing sales terminal layout and improving sales terminal performance. It provides data support and theoretical basis for enterprises to formulate market competition strategies and optimize product layout, helping enterprises to improve their market competitiveness.

[0217] Its technical advantages and characteristics mainly include: Scientificity and systematicity: When constructing the index system, this system collects and analyzes a large number of relevant literature and verification results, integrates multi-source information such as market research, industry monitoring and application case feedback, sorts out, compares and analyzes correlation to extract key elements and laws to ensure that the evaluation is scientific and reasonable. Advancedness and operability: Advanced technologies such as AHP hierarchical analysis method and principal component analysis method are used to determine the indicator weights, and fine-tune them based on actual data to make the evaluation results closer to market reality. At the same time, the system is easy to implement and highly operational. Intuitiveness and visualization: The complex competitiveness evaluation results are intuitively displayed through three-dimensional visualization technology, which is convenient for enterprises to quickly understand market dynamics and competitive situations. Decision support: Provide enterprises with accurate data support and decision-making basis, and help enterprises formulate effective market competition strategies and product layout optimization plans.

[0218] In summary, the three-dimensional visualization evaluation system for the market competitiveness of cigarette products proposed in the present invention has significant technical advantages and practical value, and provides strong support for enterprises to formulate market competition strategies and optimize product layout.

[0219] In this embodiment, a production adjustment device based on product scoring is also provided, which is used to implement the above-mentioned embodiments and preferred embodiments, and will not be repeated hereafter. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0220] This embodiment provides a production adjustment device based on product scoring, such as Figure 7 As shown, including:

[0221] The three-level indicator matrix construction module 401 is used to construct a three-level indicator matrix corresponding to each second-level indicator according to the sample data of the target historical period and the preset indicator system. The sample data includes: quantitative information of each product under the three-level indicator; the preset indicator system includes: multiple first-level indicators, multiple second-level indicators and multiple third-level indicators, as well as the subordinate relationship between each indicator.

[0222] The three-level indicator weight determination module 402 is used to process each three-level indicator matrix using the principal component analysis algorithm to determine the effective three-level indicators in each three-level indicator matrix and the three-level weights of each effective three-level indicator.

[0223] The basic indicator score determination module 403 is used to normalize the sample data corresponding to the effective three-level indicators in each three-level indicator matrix to obtain the basic indicator score of each effective three-level indicator of the target product.

[0224] The indicator comprehensive weight determination module 404 is used to determine the comprehensive weight of each valid third-level indicator according to the first-level weight of each first-level indicator, the second-level weight of each second-level indicator and the third-level weight of each valid third-level indicator.

[0225] The product production plan adjustment module 405 is used to adjust the production plan of the target product based on the comprehensive product score of the target product determined by the basic indicator scores of each effective third-level indicator of the target product and the comprehensive weight.

[0226] In some optional implementations, the first-level weight of each first-level indicator is determined as follows:

[0227] Conduct big data analysis on industry research data through the pre-trained first data analysis model to determine the correlation between each first-level indicator and construct a judgment matrix;

[0228] Conduct hierarchical analysis on the judgment matrix to determine the first-level weight of each first-level indicator.

[0229] In some optional implementations, the secondary weight of each secondary indicator is determined as follows:

[0230] The sales terminal data in the database is analyzed by a pre-trained second data analysis model to determine the membership of each secondary indicator under each primary indicator to the preset evaluation set and to construct a fuzzy relationship matrix corresponding to each primary indicator;

[0231] According to the fuzzy relationship matrix corresponding to each first-level indicator, the second-level weights of each second-level indicator under each first-level indicator are determined.

[0232] In some optional implementations, when performing hierarchical analysis on the judgment matrix to determine the primary weight of each primary indicator, it includes:

[0233] Determine the maximum eigenvalue of the judgment matrix, determine the eigenvector corresponding to the maximum eigenvalue as the target eigenvector, and determine the weights to be confirmed corresponding to each primary indicator based on the target eigenvector;

[0234] Determine the consistency test result corresponding to the maximum eigenvalue and the target eigenvector, and judge whether the test result meets the preset threshold requirement;

[0235] If it meets the requirements, the weight to be confirmed will be determined as the first-level weight;

[0236] If not, the maximum eigenvalue is updated, and the eigenvector corresponding to the maximum eigenvalue is determined as the target eigenvector, and based on the target eigenvector, the steps of determining the weights to be confirmed corresponding to each primary indicator are performed.

[0237] In some optional implementations, the first-level indicators include: product brand, product performance, and product sales environment;

[0238] The relationship between the primary and secondary indicators is:

[0239] Secondary indicators under product brands include: basic brand information, brand sales;

[0240] Secondary indicators under product performance include: regional performance, competition duration;

[0241] The secondary indicators under the product sales environment include: sales terminal coverage and comprehensive situation of sales terminals.

[0242] In some optional implementations, the three-level indicator weight determination module, when processing each three-level indicator matrix using the principal component analysis algorithm to determine the effective three-level indicators in each three-level indicator matrix and the three-level weights of each effective three-level indicator, includes:

[0243] Determine the correlation coefficient matrix of each three-level indicator matrix, and calculate the variance contribution rate of each three-level indicator in each three-level indicator matrix according to the correlation coefficient matrix of each three-level indicator matrix;

[0244] The effective third-level indicators and the third-level weights of each effective third-level indicator are determined according to the variance contribution rate of each third-level indicator.

[0245] In some optional implementations, the three-level indicator weight determination module, after determining the correlation coefficient matrix of each three-level indicator matrix, performs a suitability test on each correlation coefficient matrix and determines whether the test result is qualified;

[0246] If qualified, the variance contribution rate of each third-level indicator in the corresponding three-level indicator matrix is ​​calculated according to the correlation coefficient matrix.

[0247] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0248] The production regulation device based on product scoring in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0249] The embodiment of the present invention also provides a computer device having the above Figure 7 A production regulator based on product rating is shown.

[0250] See also Figure 8 , Figure 8 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Figure 8 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 8 A processor 10 is taken as an example.

[0251] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0252] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0253] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0254] The memory 20 may include a volatile memory. The computer device also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Figure 8 The example of the connection via bus is taken. The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device may be a touch screen.

[0255] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0256] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A production adjustment method based on product scoring, characterized in that: The method comprises: According to the sample data of the target historical period and the preset indicator system, a three-level indicator matrix corresponding to each second-level indicator is constructed, wherein the sample data includes: quantitative information of each product under the third-level indicator; the preset indicator system includes: multiple first-level indicators, multiple second-level indicators and multiple third-level indicators, as well as the subordinate relationship between each indicator; The principal component analysis algorithm is used to process each three-level indicator matrix to determine the effective three-level indicators in each three-level indicator matrix and the three-level weights of each effective three-level indicator; Normalizing the sample data corresponding to the effective third-level indicators in each third-level indicator matrix to obtain the basic indicator score of each effective third-level indicator of the target product; Determine the comprehensive weight of each valid third-level indicator based on the first-level weight of each first-level indicator, the second-level weight of each second-level indicator and the third-level weight of each valid third-level indicator; Based on the comprehensive product score of the target product determined by the basic indicator scores and comprehensive weights of each effective third-level indicator of the target product, the production plan of the target product is adjusted.

2. The method according to claim 1, characterized in that The first-level weight of each first-level indicator is determined as follows: Performing big data analysis on industry research data through a pre-trained first data analysis model to determine the correlation between the first-level indicators and construct a judgment matrix; A hierarchical analysis is performed on the judgment matrix to determine the primary weight of each primary indicator.

3. The method according to claim 1, characterized in that The secondary weights of the secondary indicators are determined as follows: The sales terminal data in the database is analyzed by a pre-trained second data analysis model to determine the membership of each secondary indicator under each primary indicator to the preset evaluation set and to construct a fuzzy relationship matrix corresponding to each primary indicator; According to the fuzzy relationship matrix corresponding to each primary indicator, the secondary weights of each secondary indicator under each primary indicator are determined.

4. The method according to claim 2, characterized in that: The step of performing hierarchical analysis on the judgment matrix to determine the primary weight of each primary indicator includes: Determine the maximum eigenvalue of the judgment matrix, determine the eigenvector corresponding to the maximum eigenvalue as the target eigenvector, and determine the weights to be confirmed corresponding to each primary indicator based on the target eigenvector; Determine the consistency test result corresponding to the maximum eigenvalue and the target eigenvector, and judge whether the test result meets the preset threshold requirement; If it meets the requirements, the weight to be confirmed is determined as the first-level weight; If not, the maximum eigenvalue is updated, and the eigenvector corresponding to the maximum eigenvalue is determined as the target eigenvector, and based on the target eigenvector, the weights to be confirmed corresponding to each primary indicator are determined.

5. The method according to claim 1, characterized in that The first-level indicators include: product brand, product performance and product sales environment; The affiliation relationship between the first-level indicators and the second-level indicators is: The secondary indicators under the product brand include: basic brand information and brand sales; The secondary indicators under the product performance include: regional performance, competition duration; The secondary indicators under the product sales environment include: sales terminal coverage and comprehensive situation of sales terminals.

6. The method according to claim 1, characterized in that The method of processing each three-level indicator matrix using the principal component analysis algorithm to determine the effective three-level indicators in each three-level indicator matrix and the three-level weights of each effective three-level indicator includes: Determine the correlation coefficient matrix of each three-level indicator matrix, and calculate the variance contribution rate of each three-level indicator in each three-level indicator matrix according to the correlation coefficient matrix of each three-level indicator matrix; The effective third-level indicators and the third-level weights of each effective third-level indicator are determined according to the variance contribution rate of each third-level indicator.

7. The method according to claim 5, characterized in that The method further comprises: After determining the correlation coefficient matrix of each three-level indicator matrix, perform a suitability test on each correlation coefficient matrix and judge whether the test result is qualified; If qualified, the variance contribution rate of each third-level indicator in the corresponding three-level indicator matrix is ​​calculated according to the correlation coefficient matrix.

8. A production adjustment device based on product scoring, characterized in that: The device comprises: A three-level indicator matrix construction module is used to construct a three-level indicator matrix corresponding to each secondary indicator according to sample data of the target historical period and a preset indicator system, wherein the sample data includes: quantitative information of each product under the three-level indicator; the preset indicator system includes: multiple first-level indicators, multiple second-level indicators and multiple third-level indicators, as well as the subordinate relationship between each indicator; A three-level indicator weight determination module is used to process each three-level indicator matrix using a principal component analysis algorithm to determine the effective three-level indicators in each three-level indicator matrix and the three-level weights of each effective three-level indicator; A basic indicator score determination module is used to normalize the sample data corresponding to the effective three-level indicators in each three-level indicator matrix to obtain the basic indicator score of each effective three-level indicator of the target product; An indicator comprehensive weight determination module is used to determine the comprehensive weight of each valid third-level indicator according to the first-level weight of each first-level indicator, the second-level weight of each second-level indicator and the third-level weight of each valid third-level indicator; The product production plan adjustment module is used to adjust the production plan of the target product based on the comprehensive product score of the target product determined by the basic indicator scores of each effective third-level indicator of the target product and the comprehensive weight.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the production adjustment method based on product scoring as described in any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the production adjustment method based on product scoring according to any one of claims 1 to 7.

Images