Method and system for product selection optimization, electronic equipment and computer program product

Abstract

The embodiment of the invention provides a method and a system for product selection optimization, electronic equipment and a computer program product, and belongs to the field of electronic commerce. The method comprises the steps of determining a comprehensive weight of each evaluation index according to a subjective weight and an objective weight of each evaluation index in a plurality of evaluation indexes of a sample commodity; obtaining the score of each evaluation index of the sample commodity in the current time period; based on the comprehensive weight and the score in the current time period, obtaining a comprehensive score of the sample commodity in the current time period; according to the comprehensive score of the sample commodity in the current time period and the historical comprehensive score in the set time period, obtaining a function of the comprehensive score and time; and performing function clustering analysis on the comprehensive scores of the plurality of sample commodities of the plurality of preset commodity attributes and time, and determining attributes of candidate commodities. According to the embodiment of the invention, the market performance attribute of the commodity can be accurately and comprehensively determined, and a powerful decision basis is provided for operation resource adjustment, update and elimination of the commodity.

Description

Technical Field

[0001] The present invention relates to the technical field of e-commerce, and in particular to a method for optimizing product selection, a system for optimizing product selection, an electronic device, a computer-readable storage medium, and a computer program product. Background Art

[0002] In the process of convenience store business operation, the traditional method of product selection is usually based on the subjective judgment of product purchasers and the basic sales data and market share data of the products within the enterprise to evaluate the operating status of the products. However, these methods often ignore the deep connection between the internal and external markets of the products and the comprehensive analysis of the market performance data of the categories to which the products belong. Therefore, the evaluation results of these methods are often not comprehensive and accurate, and it is difficult to provide a strong decision-making basis for the adjustment and update and elimination of the operating resources of the products. Summary of the invention

[0003] The purpose of the embodiments of the present invention is to provide a method for product selection optimization, a system for product selection optimization, an electronic device, a computer-readable storage medium and a computer program product. The method can partially or completely solve the problem that the prior art is difficult to accurately and comprehensively determine the market performance attributes of a product, which in turn makes it difficult to provide a strong decision-making basis for the adjustment of the operational resources and the update and elimination of the product.

[0004] In order to achieve the above-mentioned purpose, an embodiment of the present invention provides a method for product selection optimization, the method comprising: determining the comprehensive weight of each evaluation indicator according to the subjective weight and objective weight of each evaluation indicator among multiple evaluation indicators of the sample product; obtaining the score of each evaluation indicator of the sample product in the current time period; obtaining the comprehensive score of the sample product in the current time period according to the comprehensive weight of each evaluation indicator and the score of each evaluation indicator in the current time period; obtaining the function of the comprehensive score of the sample product and time according to the comprehensive score of the sample product in the current time period and the historical comprehensive score in a set time period before the current time period; and clustering analysis of the function of the comprehensive score of multiple sample products of multiple preset product attributes and time to determine the attributes of the candidate product among the multiple preset product attributes.

[0005] Optionally, the subjective weight and objective weight of each of the multiple evaluation indicators of the sample product are determined by the following steps: using the hierarchical analysis method to determine the subjective weight of each of the multiple evaluation indicators of the sample product; using the improved entropy weight method to determine the objective weight of each of the multiple evaluation indicators of the sample product.

[0006] Optionally, the method of determining the objective weight of each evaluation indicator of the sample product using the improved entropy weight method includes: constructing an initial data matrix of the sample product; performing forward processing and normalization processing on the multiple evaluation indicators of the sample product; determining the entropy value of each evaluation indicator; and determining the objective weight of each evaluation indicator according to the entropy value of each evaluation indicator; wherein the entropy value H of each evaluation indicator is j Determined by the following formula:

[0007]

[0008] Among them, m represents the total number of sample products, n represents the total number of evaluation indicators, and x ij It is the result after processing the jth evaluation index of the i-th sample product.

[0009] Optionally, determining the objective weight of each evaluation indicator according to the entropy value of each evaluation indicator comprises: determining the objective weight ω of each evaluation indicator by the following formula: j :

[0010]

[0011] in,

[0012]

[0013] 0≤ω j ≤1,

[0014] ω j is the objective weight of the jth evaluation index, H k represents the entropy value of the kth evaluation index, is the mean of all entropy values ​​that are not 1.

[0015] Optionally, before executing the step of determining the comprehensive weight of each evaluation indicator, the method further includes: constructing a multi-level evaluation indicator system based on internal business data and external business data of sample products and market performance data of the categories to which the sample products belong.

[0016] Optionally, the multi-level evaluation index system includes three levels, the first level includes multiple evaluation indicators of the sample product, the second level is multiple secondary capability domains divided according to multiple categories by the multiple evaluation indicators in the first level, and the third level is multiple first-level capability domains divided according to multiple categories by the multiple secondary capability domains of the second level; according to the comprehensive weight of each evaluation indicator and the score of each evaluation indicator in the current time period, obtaining the comprehensive score of the sample product in the current time period includes: adding the comprehensive weights of the multiple evaluation indicators under each secondary capability domain to obtain the comprehensive weight of each secondary capability domain; adding the comprehensive weights of the multiple secondary capability domains under each first-level capability domain to obtain the comprehensive weight of each first-level capability domain; weighted summing the scores of the multiple evaluation indicators under each secondary capability domain in the current time period to obtain the score of each secondary capability domain; weighted summing the scores of the multiple secondary capability domains under each first-level capability domain to obtain the score of each first-level capability domain; and according to the comprehensive weights and scores of the multiple first-level capability domains, obtaining the comprehensive score of the sample product in the current time period.

[0017] Optionally, obtaining the score of each evaluation indicator of the sample product in the current time period includes: obtaining the value of each evaluation indicator of multiple evaluation indicators of the sample product in the current time period; and normalizing the value of each evaluation indicator in the current time period to obtain the score of each evaluation indicator in the current time period.

[0018] Optionally, the multiple preset product attributes include seasonal products, strategic products, star products, products to be kept on the shelves and eliminated products, and the comprehensive scores of the sample products of the multiple preset product attributes are clustered and analyzed as a function of time. Determining the attributes of the candidate products among the multiple preset product attributes includes: obtaining the distance between the comprehensive scores of the sample products of the multiple preset product attributes and the function of time and the distance between their derivative functions; weighted summing the distances between the functions and the distances between their derivative functions to obtain a weighted distance function of the sample products of the multiple preset product attributes; selecting five sample products that are seasonal products, strategic products, star products, products to be kept on the shelves and eliminated products as the initial cluster centers of the weighted distance function, and calculating the distance between the candidate product and the initial cluster center of the five sample products; and determining the attribute corresponding to the sample product that is closest to the initial cluster center of the five sample products as the attribute of the candidate product.

[0019] On the other hand, the present invention also provides a system for product selection optimization, the system comprising: a first determination module, used to determine the comprehensive weight of each evaluation indicator of the sample product according to the subjective weight and objective weight of each evaluation indicator of the multiple evaluation indicators of the sample product; a first acquisition module, used to obtain the score of each evaluation indicator of the sample product in the current time period; a second acquisition module, used to obtain the comprehensive score of the sample product in the current time period according to the comprehensive weight of each evaluation indicator and the score of each evaluation indicator in the current time period; a third acquisition module, used to obtain the function of the comprehensive score of the sample product and time according to the comprehensive score of the sample product in the current time period and the historical comprehensive score in a set time period before the current time period; and a second determination module, which performs cluster analysis on the function of the comprehensive score of multiple sample products of multiple preset product attributes and time to determine the attributes of the candidate product among the multiple preset product attributes.

[0020] On the other hand, the present invention also provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method for product selection optimization as described.

[0021] On the other hand, the present invention also provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the method for product selection optimization as described above.

[0022] On the other hand, the present invention also provides a computer program product, including a computer program, which implements the method for product selection optimization as described when executed by a processor.

[0023] Through the above technical scheme, the present invention determines the comprehensive weight of each evaluation index through subjective weight and objective weight, and obtains the comprehensive score of the sample product in the current time period based on the comprehensive weight of each evaluation index and the score of each evaluation index in the current time period. From the comprehensive evaluation of subjective weight and objective weight, valuable market information can be extracted comprehensively and accurately, which improves the accuracy and comprehensiveness of determining the market performance attributes of the product; at the same time, based on the functional relationship between the comprehensive score and time in the current time period and the historical time period, the operating conditions and development potential of the product are fully explored, which provides a favorable basis for the adjustment of the operating resources of the product and the update and elimination mechanism.

[0024] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0026] Figure 1 is a structural schematic diagram of a method for product selection optimization provided by a first embodiment of the present invention;

[0027] Figure 2 is a schematic diagram of the structure of a three-level evaluation index system provided in an embodiment of the present invention;

[0028] Figure 3 This is the main framework diagram of the data evaluation model for product selection optimization provided by the first embodiment of the present invention;

[0029] Figure 4 It is a schematic diagram of the structure of a system for product selection optimization provided by the second embodiment of the present invention. DETAILED DESCRIPTION

[0030] The specific implementation of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.

[0031] like Figure 1 It is a structural schematic diagram of a method for product selection optimization provided by the first embodiment of the present invention, and the method includes the following steps S10-S14.

[0032] S10. Determine a comprehensive weight of each evaluation indicator according to the subjective weight and the objective weight of each evaluation indicator among the multiple evaluation indicators of the sample product.

[0033] Among them, before executing the step of determining the comprehensive weight of each evaluation indicator, the method also includes: constructing a multi-level evaluation indicator system based on the internal business data and external business data of the sample products and the market performance data of the categories to which the sample products belong.

[0034] Exemplarily, the sample goods in the embodiment of the present invention are, for example, mineral water in the packaged beverage category, and can be specifically detailed to a certain brand of mineral water. The sample goods in the embodiment of the present invention can also be different goods of various types, such as mineral water in the packaged beverage category, liquor, beer, etc. in the alcohol category. Regardless of whether the number of sample goods is one or more, the method of determining the objective weight and subjective weight of each evaluation indicator in the sample business is the same. Among them, the evaluation indicators can specifically be the year-on-year compound growth rate of revenue, the month-on-month compound growth rate of revenue, the year-on-year compound growth rate of gross profit, the month-on-month compound growth rate of gross profit, the average daily revenue of a single store, the average daily gross profit of a single store, etc., which represent the efficiency of commodity operation. The basket rate, the number of days without sales, the joint rate, etc. represent the efficiency of commodity operation. The evaluation index refers to an indicator used to characterize the sales status of commodities, which can specifically include operating efficiency, sales profitability, inventory control, commodity competitiveness and customer service.

[0035] Before determining the comprehensive weight of each evaluation index, the embodiment of the present invention first constructs a multi-level evaluation index system based on the internal business data and external business data of the sample commodity and the market performance data of the category to which the sample commodity belongs. Among them, the internal business data may specifically be sales data, inventory data, etc. The external business data may specifically be market share data, customer feedback, competitor sales data, competitiveness data, etc. The market performance data of the category to which the commodity belongs, such as category sales, market share, etc., is used to evaluate the position and influence of the commodity in the market. Based on the above-mentioned collected data, a complete multi-level evaluation system is constructed using existing means. For specific details, please refer to the subsequent steps.

[0036] The embodiment of the present invention combines the operating data of the internal and external markets of the commodity and the market performance data of the category to which the commodity belongs, comprehensively evaluates the operating status and development potential of the sample commodity from multiple aspects, and provides a powerful decision-making basis for the adjustment, update and elimination of the commodity's operating resources.

[0037] Furthermore, through the following steps S101-S102, the subjective weight and the objective weight of each evaluation index of the multiple evaluation indexes of the sample product are determined.

[0038] S101. Determine the subjective weight of each evaluation index among multiple evaluation indexes of sample products by using the analytic hierarchy process.

[0039] For example, the subjective weight is determined by using the analytic hierarchy process, which is a decision analysis method that combines qualitative and quantitative methods to solve complex problems with multiple objectives. This method combines quantitative analysis with qualitative analysis, uses the decision maker's experience to judge the relative importance of indicators, and reasonably gives the weight of each criterion of each decision plan. It is more effectively applied to those topics that are difficult to solve with quantitative methods.

[0040] The steps of the hierarchical analysis method are as follows: First, establish a hierarchical model. Divide the decision-making goals, factors to be considered (decision-making criteria) and decision-making objects into the highest level, middle level and lowest level according to their mutual relationship, and draw a hierarchical diagram. In the hierarchical analysis model, the goal level represents the purpose, that is, the evaluation of the commodity management level; the criterion level is the intermediate link involved in achieving the predetermined goal, that is, the various evaluation indicators for evaluating the commodity management level. The solution level represents the specific solution to the problem, that is, the sample commodity representing the evaluation of the management level.

[0041] Secondly, construct a judgment matrix. Usually, 1 to 9 points are selected as the score scale. By comparing the evaluation indicators of the same layer in the criterion layer, the scoring matrix is ​​obtained, that is, the judgment matrix D = (d ij The meaning of each element in the matrix (i.e., evaluation index) is shown in Table 1.

[0042]

[0043]

[0044] Table 1

[0045] According to Table 1, the important relationship between the evaluation indicators is determined, and the scoring matrix is ​​obtained, that is, the judgment matrix D = (d ij ). In the process of determining the importance of indicators between two indicators, it can be determined by experts in a subjective form.

[0046] Secondly, the consistency check of the judgment matrix. This step takes into account that since the evaluation process involves multiple indicators and multiple stages of judgment, experts are prone to inconsistent judgment results when judging the importance of indicators. Therefore, before sorting the judgment matrix, a consistency check is required. In the hierarchical analysis method, for the judgment matrix D = (d ij ) is based on the matrix theory in linear algebra. The consistency of the judgment matrix is ​​related to the degree of change of its characteristic roots. The consistency index of the judgment matrix is ​​the negative average of the remaining characteristic roots except its largest characteristic root, that is, Among them, λ max is the maximum eigenvalue of the judgment matrix, and n is the matrix order.

[0047] The larger the CI value, the greater the deviation from complete consistency, and vice versa. The order of the matrix will affect the consistent error of acceptable judgment. Therefore, the average random consistency index RI value is introduced, and the RI value can be obtained by looking up the table.

[0048] In actual operation, the random consistency ratio CR value is usually used to observe the consistency of the judgment matrix, that is, When CR is less than the set value (for example, 0.1), the judgment matrix is ​​considered to have satisfactory consistency. Otherwise, the judgment matrix should be adjusted (the importance of each evaluation index or the corresponding score should be adjusted) to make it have satisfactory consistency.

[0049] Finally, the subjective weight of each evaluation index is calculated. The weight of the judgment index is realized by calculating the maximum characteristic root and its eigenvector of the judgment matrix. max The eigenvector W of is normalized and recorded as W'. The element W of W' i That is the weight of the corresponding evaluation index i.

[0050] In order to ensure the accuracy and reliability of the evaluation results, this model uses a subjective and objective fusion method to determine the weight of each indicator. The subjective method uses the analytic hierarchy process to determine the importance of each indicator through the experience and judgment of experts; the objective method uses the improved entropy weight method to calculate the weight based on the objective characteristics of the data. By combining subjective and objective methods, the model can determine the weight of each indicator more scientifically and reasonably.

[0051] S102: Determine the objective weight of each evaluation index of the sample product by using an improved entropy weight method.

[0052] In the determination of objective weights, the improved entropy weight method is used to calculate objective weights. In information theory, entropy is a measure of uncertainty. The greater the uncertainty, the greater the entropy and the greater the amount of information contained; the smaller the uncertainty, the smaller the entropy and the smaller the amount of information contained. The entropy weight method is used to calculate the weights of each evaluation index, that is, the effective information of the evaluation index is used for calculation. The greater the effective information, the greater the weight. According to the characteristics of entropy, the randomness and disorder of an event can be judged by calculating the entropy value, and the degree of discreteness of an indicator can also be judged by the entropy value. The greater the discreteness of the indicator, the greater the influence of the indicator on the comprehensive evaluation (the greater the weight). For example, if the sample data has the same value under a certain indicator, the influence of the indicator on the overall evaluation is 0, and the weight is 0. The entropy weight method is an objective weighting method because it only depends on the discreteness of the data itself.

[0053] Furthermore, the step of determining the objective weight of each evaluation index of the sample product by using the improved entropy weight method includes steps S1021-S1024.

[0054] S1021. Construct an initial data matrix of sample products.

[0055] For example, there are m sample products and n evaluation indicators, then the initial data matrix is ​​Y = [y ij ] m×n , where yij is the value of the j-th evaluation index of the ith sample product.

[0056] S1022. Perform positive processing and normalization processing on multiple evaluation indicators of the sample products.

[0057] For example, evaluation indicators can also be divided into positive indicators and negative indicators according to their nature. Negative indicators refer to data indicators that measure the bad or deteriorated process or result. Product defect rate, number of customer complaints, etc. are all negative indicators. These indicators can reveal problems in products or services, thereby prompting companies to improve product or service quality. Positive indicators refer to indicators that can reflect the progress or effectiveness of a process or activity in the expected direction. Sales growth rate, customer satisfaction, etc. are typical positive indicators, which can directly reflect the market performance and customer satisfaction of the company.

[0058] For the evaluation indicators that are positive indicators, normalization is performed. For the evaluation indicators that are negative indicators, positive processing is performed first and then normalization. The positive processing is performed as follows:

[0059] The data matrix after forward transformation is normalized to eliminate the dimensional differences between the indicators: Finally, we get the data matrix X after forward and normalization = [x ij ] m×n , where x ij It is the result after processing the jth evaluation index of the i-th sample product.

[0060] S1023. Determine the entropy value of each evaluation indicator.

[0061] Among them, the entropy value H of each evaluation index is j Determined by the following formula:

[0062]

[0063] Among them, m represents the total number of sample products, n represents the total number of evaluation indicators, and x ij It is the result after processing the jth evaluation index of the i-th sample product.

[0064] For example, the traditional entropy value e j The calculation formula is:

[0065]

[0066] in,

[0067] In order to improve the differentiation between indicators, this patent uses an exponential entropy calculation formula.

[0068] When β>1, the improved entropy value is more sensitive than the traditional one; when β<1, the improved entropy value is more robust than the traditional one. When β=1, it is the traditional entropy value. In order to improve the discrimination between indicators, the embodiment of the present invention uses β=2, that is:

[0069] S1024: Determine the objective weight of each evaluation indicator according to the entropy value of each evaluation indicator.

[0070] Further, determining the objective weight of each evaluation indicator according to the entropy value of each evaluation indicator includes:

[0071] The objective weight ω of each evaluation index is determined by the following formula: j :

[0072]

[0073] in,

[0074]

[0075] 0≤ω j ≤1,

[0076]

[0077] ω j represents the objective weight of the jth evaluation index, H k represents the entropy value of the kth evaluation index, is the mean of all entropy values ​​that are not 1.

[0078] Exemplarily, the traditional entropy weight calculation formula is: When calculating H j →1, its slight change will cause the entropy weight to change exponentially, which will cause a certain error in the weight calculation result, especially when the entropy value calculation formula after using the exponential function form is improved. j →1, it is more sensitive. In order to reduce the error, the improved entropy weight method is used:

[0079]

[0080] in, 0≤ω j ≤1,

[0081] In the above steps, the subjective weight W of the evaluation index i is determined iWhen the subjective weight W of any rating index j can be obtained j Finally, the subjective weight W of each evaluation index is j With objective weight ω j After weighted summation, the comprehensive weight Z of each evaluation index is obtained. j :Z j =αW j +βω j .

[0082] S11. Obtain the score of each evaluation indicator of the sample product in the current time period.

[0083] Based on the determined weights, this module is responsible for calculating the scores of each product on each indicator and summarizing them to obtain the comprehensive scores of the products. These scores can reflect the performance of the products in different aspects and provide a basis for subsequent clustering analysis and operational optimization.

[0084] Furthermore, obtaining the score of each evaluation indicator of the sample product in the current time period includes the following steps S111-S112.

[0085] S111. Obtain the value of each evaluation index of the sample product in the current time period.

[0086] For example, the current time period may be data of the sample product in the past month. The value of each evaluation index of the sample product in the past month is obtained by existing technical means.

[0087] S112: Normalize the value of each evaluation indicator in the current time period to obtain the score of each evaluation indicator in the current time period.

[0088] For example, when calculating the indicator score, the indicator value is first normalized, specifically using the following formula: sv ij is the normalized value of the evaluation index j of the i-th sample product, v ij is the original value of the average index j of the ith commodity (i.e. the value before normalization), min j {v ij} is the minimum and maximum value of evaluation index j among all sample products. j {v ij} is the maximum value of evaluation index j among all sample products.

[0089] Secondly, the sigmoid function is used to convert the value of each evaluation indicator into a score. The sigmoid function is Any real number can be mapped to a value between 0 and 1.

[0090] The score for positive indicators is score ij =100×σ(sv ij ); the score for negative indicators is score ij =100×σ(-sv ij ); the score for the neutral indicator is score ij =100×σ(|sv ij -avg j {v ij}|), where avg j {v ij} is the mean of index j in all commodities. Among them, the neutral index is an indicator used in technical analysis to describe the intermediate state of the market, indicating that the market has no obvious directional characteristics. Moving average is one of the commonly used neutral indicators. Investors can use the moving average to judge the potential changes in market trends, but the moving average itself does not directly reflect the upward or downward trend of the market.

[0091] After determining the comprehensive weight of each evaluation indicator and the score in the current time period, the final score of each evaluation indicator is: Z j ×score ij .

[0092] S12. Obtain a comprehensive score of the sample product in the current time period according to the comprehensive weight of each evaluation indicator and the score of each evaluation indicator in the current time period.

[0093] Furthermore, the multi-level evaluation index system includes three levels, the first level includes multiple evaluation indicators of the sample products, the second level is multiple secondary capability domains divided according to multiple categories by the multiple evaluation indicators in the first level, and the third level is multiple first-level capability domains divided according to multiple categories by the multiple secondary capability domains of the second level; the comprehensive score of the sample products in the current time period is obtained according to the comprehensive weight of each evaluation indicator and the score of each evaluation indicator in the current time period, including the following steps S121-S125.

[0094] For example, the constructed multi-level evaluation index system includes three levels: Figure 2 It is a structural diagram of the three-level evaluation index system provided in the embodiment of the present invention. The first level includes multiple evaluation indicators of the sample products, such as inventory turnover days, out-of-stock rate, average sales price index, revenue PSD ratio, gross profit PSD ratio, major category ranking, medium category ranking, etc., member transaction ratio, member income ratio, repurchase rate, etc.

[0095] The second level is a plurality of secondary capability domains divided according to a plurality of categories by the plurality of evaluation indicators in the first level. For example, the inventory turnover days, out-of-stock days rate and GMROI (gross profit return rate of goods) are divided into the secondary capability domain according to the inventory turnover efficiency and inventory return ability, that is, the inventory turnover days and out-of-stock days rate are further divided into the inventory turnover efficiency under the secondary capability domain; GMROI is further divided into the inventory return ability under the secondary capability domain. The secondary capability domain is further divided into a plurality of first-level capability domains according to a plurality of categories, such as further dividing the inventory turnover efficiency and the inventory return ability into the first-level capability domain according to the inventory control category. Similarly, the sales performance and ranking are divided into the first-level capability domain according to the competitiveness of the goods. The division of other evaluation indicators is similar and will not be repeated here one by one.

[0096] S121. Add the comprehensive weights of multiple evaluation indicators under each secondary capability domain to obtain the comprehensive weight of each secondary capability domain.

[0097] For example, the comprehensive weight Z of each evaluation index obtained in step S1024 is j Finally, the comprehensive weights of multiple evaluation indicators under the secondary capability domain are added together. For example, the comprehensive weights of inventory turnover days and out-of-stock days are added together to get the comprehensive weight of inventory turnover efficiency. Similarly, the comprehensive weights of member transaction proportion, member income proportion and repurchase rate are added together to get the comprehensive weight of user stickiness.

[0098] S122. Add the comprehensive weights of multiple second-level capability domains under each first-level capability domain to obtain the comprehensive weight of each first-level capability domain.

[0099] For example, the comprehensive weight of inventory turnover efficiency and the comprehensive weight of inventory return capability are added together to obtain the comprehensive weight of inventory control, and the weight of customer service is the comprehensive weight of user stickiness. The comprehensive weight of sales performance and the comprehensive weight of ranking are added together to obtain the comprehensive weight of product competitiveness.

[0100] S123. Perform a weighted sum of the scores of multiple evaluation indicators under each secondary capability domain in the current time period to obtain a score for each secondary capability domain.

[0101] For example, in step S11, the score of each evaluation indicator in the current time period is determined. ijIn the case of , the scores of multiple evaluation indicators under the secondary capability domain in the current time period are weighted and summed. For example, if the scores of storage turnover days and out-of-stock product rate are known, the scores of inventory turnover efficiency are obtained by weighting them according to the set weights such as 0.5 each. Similarly, the score of GMROI is obtained by weighting it with 1 to obtain the score of inventory return capability. Similarly, the scores of each secondary capability domain can be obtained, such as the scores of sales performance, ranking, user stickiness, etc., which will not be listed here one by one.

[0102] S124. Perform a weighted sum of the scores of multiple second-level capability domains under each first-level capability domain to obtain a score for each first-level capability domain.

[0103] Exemplarily, the scores of each first-level capability domain are obtained by weighted summing the scores of each second-level capability domain obtained in step S123. For example, the scores of inventory turnover efficiency and inventory return capability are weighted (for example, the weights are both set to 0.5) to obtain the score of inventory control.

[0104] S125. Obtain a comprehensive score for the sample product in the current time period based on the comprehensive weights and scores of the multiple first-level capability domains.

[0105] Exemplarily, the comprehensive weights of various first-level capability domains are determined according to the above steps S121-S122, and the scores of various first-level capability domains are determined according to steps S123-S124, and the weighted sum of each first-level capability domain is obtained to obtain the comprehensive score of the sample product in the current time period. Assuming that a certain brand of mineral water is determined according to the above steps, the comprehensive weight of sales profit in its first-level capability domain is 30%, and the score is 88 points; the comprehensive weight of operating efficiency is 20%, and the score is 81 points; the comprehensive weight of inventory control is 20%, and the score is 54 points; the comprehensive weight of product competitiveness is 20%, and the score is 87 points; the comprehensive weight of customer service is 10%, and the score is 74 points. Finally, the weighted sum is calculated to obtain a comprehensive score of 78.2 for the sample product in the current time period, which can be rounded up to 79 points.

[0106] S13. Obtain a function of the comprehensive score of the sample product and time based on the comprehensive score of the sample product in the current time period and the historical comprehensive score in a set time period before the current time period.

[0107] Exemplarily, the above steps calculate the comprehensive score of the sample product in the current time period. Similarly, the historical comprehensive score in the set time period before the current time period can be calculated. For example, the same method can be used to obtain the historical comprehensive score of the product in the past 23 months, plus the comprehensive score of the current month, to obtain the comprehensive score of the product in the past 24 months.

[0108] With respect to only considering the absolute value level of the current score, the embodiment of the present invention also considers the trend and structural characteristics of historical data, and performs function fitting on the comprehensive score of each commodity in the past 24 months and time to obtain the functional relationship between the comprehensive score and time.

[0109] The collected discrete comprehensive scores and time data of each period are reconstructed by fitting the function curve to obtain a continuous smooth curve. Commonly used processing methods include basis function smoothing, linear smoothing, rough penalty method, and local weighted smoothing. The basis function smoothing method can flexibly smooth discrete data by selecting basis functions that match the properties of the function to be estimated and controlling the number of basis functions. Compared with the linear smoothing method, the basis function smoothing method can achieve better smoothing effect.

[0110] In addition, the estimation of the basis function coefficients can also achieve similar effects to the local weighted smoothing method and the rough penalty smoothing method, and when the fitted function formula is used for further analysis, the function formula after basis function smoothing has certain computational advantages. Therefore, the present invention prefers the basis function smoothing method. When selecting the basis function, it is usually hoped to select a basis to obtain a better fit to the original data through a smaller number of K basis functions, not only to describe the data well in terms of data characteristics, but also to make the calculation as convenient as possible. When fitting the function curve, the comprehensive score data of each product in the past 24 months is regarded as a function of time, and the orthogonal polynomial basis function is used to fit the score data of each product. The fitting method uses the weighted least squares method to eliminate the heteroscedasticity phenomenon: Among them φ1(t), φ2(t),…,φ k (t) are mutually orthogonal polynomial functions.

[0111] Orthogonal polynomials are polynomials with special properties. When you integrate the product of any two such polynomials, if the two polynomials are the same, the integral result is a constant. If the two polynomials are different, the integral result is 0, that is:

[0112]

[0113] The orthogonal property helps reduce errors in the fitting process because it allows the coefficients of each orthogonal polynomial to be determined independently without being affected by the other terms. This can reduce errors due to interactions between terms.

[0114] S14. Perform cluster analysis on the comprehensive scores of multiple sample products of multiple preset product attributes and the function of time to determine the attributes of the candidate product among the multiple preset product attributes.

[0115] Among them, the multiple preset product attributes include seasonal products, strategic products, star products, keep-on-shelf and eliminated products, and the comprehensive scores of the sample products of the multiple preset product attributes are clustered and analyzed as a function of time. Determining the attributes of the candidate products among the multiple preset product attributes includes the following steps S141-S144.

[0116] S141. Obtain the distance between the function of the comprehensive score of sample products with multiple preset product attributes and time, and the distance between their derivative functions.

[0117] Exemplarily, the method in the above steps can obtain the function of the comprehensive score of the sample product and time. Similarly, multiple preset product attributes can be obtained. For example, firstly select the function of the comprehensive score of five representative products, namely seasonal products, strategic products, star products, products that are kept on the shelves and eliminated products, and calculate the distance between the functions of the sample products: the distance between the functions is: Corresponding to obtain the derivative function distance:

[0118] S142. Perform a weighted summation of the distances between the functions and the distances between their derivative functions to obtain a weighted distance function of sample commodities with multiple preset commodity attributes.

[0119] Exemplarily, the weighted distance function is: D(x i (t),x j (t)) = ω × d0 (x i (t),x j (t))+(1-ω)·d1(x i (t),x j (t)),ω∈[0,1].

[0120] S143, selecting five sample commodities, namely seasonal commodities, strategic commodities, star commodities, commodities to be kept on shelves and obsolete commodities, as initial cluster centers of the weighted distance function, and calculating the distance between the candidate commodity and the initial cluster centers of the five sample commodities.

[0121] Exemplarily, five representative commodities (representatively seasonal commodities, strategic commodities, star commodities, commodities to be kept on shelves and obsolete commodities) are first selected as initial cluster centers, and then the distance between each candidate commodity and the five initial cluster centers is calculated.

[0122] S144: Determine the attribute corresponding to the sample product that is closest to the initial cluster center of the five sample products as the attribute of the candidate product.

[0123] Exemplarily, when there are multiple candidate commodities (for example, A, B, C, D, E, F, G, etc.), assuming that candidate commodity A is closest to the sample commodity with the attribute of seasonal commodity, then it can be determined that the attribute of candidate commodity A is a seasonal commodity. Similarly, multiple candidate commodities are calculated in the same way, and each commodity is included in the initial cluster center that is closest to it, thus obtaining five types of commodities. After a clustering is completed, multiple iterations are performed until the result is stable to obtain the final clustering result. At the same time, combined with the properties of the five categories of commodities, the candidate commodities are divided into five categories: seasonal commodities, strategic commodities, star commodities, commodities to be kept on the shelves, and commodities to be eliminated. Figure 3 It is the main framework diagram of the data evaluation model suitable for product selection optimization provided by the first embodiment of the present invention.

[0124] Based on the comprehensive rating data of the goods, this module uses the clustering method of functional data to divide the goods into different types. By capturing the dynamic characteristics of the rating data changing over time, this module can achieve more accurate product classification, which helps companies understand the characteristics and needs of different types of goods. Combining the clustering results and the evaluation index system, this module conducts a comprehensive evaluation of each category and brand. By comparing the performance of different categories and brands on various indicators, this module can identify categories and brands with excellent performance, and provide decision support for the company's resource allocation and brand building.

[0125] Based on the clustering and evaluation results, this module is responsible for formulating targeted operational optimization strategies. For different types of goods, this module proposes corresponding adjustment suggestions, such as adjusting pricing strategies, optimizing inventory management, and improving customer service quality, to help companies better meet market demand and improve operational efficiency and economic benefits.

[0126] In summary, this comprehensive multi-indicator data evaluation model achieves a comprehensive, scientific and accurate evaluation and optimization of commodity operations by integrating the advantages of internal and external market data, subjective and objective fusion weights, functional data clustering methods, category evaluation, brand evaluation, etc.

[0127] This innovative achievement mainly achieves the following effects: First, a comprehensive assessment of the current business status and development potential of the commodity is made. By deeply analyzing the business data of a large number of internal and external markets of commodities and the market performance data of the categories to which the commodities belong, and combining a complete three-level evaluation index system model, this innovative achievement can comprehensively and systematically evaluate the current business status and development potential of the commodity. This comprehensive assessment enables enterprises to more accurately understand the performance of commodities in the market and customer needs, and provides solid data support for the formulation of more targeted operating strategies. Then, a basis for decision-making is provided. The evaluation method and model involved in the embodiment of the present invention provide a basis for decisions such as the adjustment of commodity operating resources and the updating and elimination of commodities. Based on the comprehensive evaluation results, the sales company can adjust the commodity operation strategy in a targeted manner, optimize resource allocation, and improve sales profitability and market competitiveness.

[0128] Secondly, this innovative achievement determines the indicator weights through a subjective and objective fusion method, making the evaluation results more scientific and reasonable. The subjective method uses the analytic hierarchy process, which fully considers the experience and judgment of experts; the objective method uses the improved entropy weight method to determine the weights based on the objective characteristics of the data. This subjective and objective fusion method not only improves the objectivity of the evaluation, but also retains the subjective judgment of experts, making the weight determination more comprehensive and reliable.

[0129] In addition, this innovative achievement has achieved accurate classification of product types through a clustering method based on functional data. By analyzing the rating data of products in the past 24 months, this method can capture the dynamic characteristics of product ratings over time, thereby accurately classifying products into different types. This accurate product classification helps sales companies better understand the characteristics and needs of different types of products, and provides strong support for the formulation of personalized operation strategies.

[0130] In addition, the patent also combines the basic conclusions of category evaluation and brand evaluation models. This means that when evaluating the current business status and development potential of a product, the influencing factors of the product category and brand will also be considered, further improving the accuracy and operability of the evaluation.

[0131] Finally, it supports operational decision-making. The patent realizes a closed loop from data to decision-making, helping sales company managers make scientific decisions based on objective data. Through comprehensive evaluation of commodities, sales companies can better grasp market dynamics, adjust business strategies in a timely manner, and enhance competitiveness and sustainable development capabilities.

[0132] like Figure 4It is a structural diagram of a system for product selection optimization provided by the second embodiment of the present invention. The system 40 for product selection optimization includes: a first determination module 401, which is used to determine the comprehensive weight of each evaluation indicator according to the subjective weight and objective weight of each evaluation indicator in multiple evaluation indicators of the sample product; a first acquisition module 402, which is used to obtain the score of each evaluation indicator of the sample product in the current time period; a second acquisition module 403, which is used to obtain the comprehensive score of the sample product in the current time period according to the comprehensive weight of each evaluation indicator and the score of each evaluation indicator in the current time period; a third acquisition module 404, which is used to obtain the function of the comprehensive score of the sample product and time according to the comprehensive score of the sample product in the current time period and the historical comprehensive score in the set time period before the current time period; and a second determination module 405, which performs cluster analysis on the function of the comprehensive score of multiple sample products of multiple preset product attributes and time to determine the attributes of the candidate product among the multiple preset product attributes.

[0133] The principles and effects of the system for product selection optimization provided by the second embodiment of the present invention are the same as those of the first embodiment described above, and will not be described in detail herein.

[0134] The third embodiment of the present invention also provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method for product selection optimization as described.

[0135] The principles and effects of the electronic device provided by the third embodiment of the present invention are the same as those of the first embodiment described above, and will not be described in detail herein.

[0136] The fourth embodiment of the present invention also provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the method for product selection optimization as described above.

[0137] The principles and effects of the computer-readable storage medium provided by the fourth embodiment of the present invention are the same as those of the first embodiment described above, and will not be described in detail herein.

[0138] The fifth embodiment of the present invention also provides a computer program product, including a computer program, which implements the method for product selection optimization as described when executed by a processor.

[0139] The principles and effects of the computer program product provided by the fifth embodiment of the present invention are the same as those of the first embodiment described above, and will not be described in detail herein.

[0140] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0141] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0142] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0143] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0144] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0145] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0146] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0147] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, sample product or device that includes a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements inherent to such process, method, sample product or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, sample product or device that includes the elements.

[0148] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for product selection optimization, characterized in that: The method comprises: Determining a comprehensive weight of each evaluation indicator among multiple evaluation indicators of the sample product according to the subjective weight and the objective weight of each evaluation indicator; Get the score of each evaluation indicator of the sample product in the current time period; Obtaining a comprehensive score of the sample product in the current time period according to the comprehensive weight of each evaluation indicator and the score of each evaluation indicator in the current time period; According to the comprehensive score of the sample product in the current time period and the historical comprehensive score in a set time period before the current time period, a function of the comprehensive score of the sample product and time is obtained; and, A cluster analysis is performed on the comprehensive scores of multiple sample products of multiple preset product attributes and the function of time to determine the attributes of the candidate product among the multiple preset product attributes.

2. The method according to claim 1, characterized in that The subjective weight and objective weight of each evaluation index of the sample product are determined by the following steps: The analytic hierarchy process is used to determine the subjective weight of each evaluation index among multiple evaluation indexes of sample products; The objective weight of each evaluation index of the sample product is determined by using an improved entropy weight method.

3. The method according to claim 2, characterized in that The method of using the improved entropy weight method to determine the objective weight of each evaluation index of the sample product includes: Construct the initial data matrix of sample products; Performing positive processing and normalization processing on multiple evaluation indicators of the sample product; Determining the entropy value of each evaluation indicator; and Determining the objective weight of each evaluation indicator according to the entropy value of each evaluation indicator; Among them, the entropy value H of each evaluation index is j Determined by the following formula: Among them, m represents the total number of sample products, n represents the total number of evaluation indicators, and x ij It is the result after processing the jth evaluation index of the i-th sample product.

4. The method according to claim 3, characterized in that Determining the objective weight of each evaluation indicator according to the entropy value of each evaluation indicator includes: The objective weight ω of each evaluation index is determined by the following formula: j : in, 0≤ω j ≤1, ω j represents the objective weight of the jth evaluation index, H k represents the entropy value of the kth evaluation index, is the mean of all entropy values ​​that are not 1.

5. The method according to claim 1, characterized in that Before executing the step of determining the comprehensive weight of each evaluation index, the method further includes: A multi-level evaluation index system is constructed based on the internal and external operating data of the sample commodities and the market performance data of the categories to which the sample commodities belong.

6. The method according to claim 5, characterized in that The multi-level evaluation index system includes three levels, the first level includes multiple evaluation indicators of the sample product, the second level is multiple secondary capability domains divided according to multiple categories by the multiple evaluation indicators in the first level, and the third level is multiple primary capability domains divided according to multiple categories by the multiple secondary capability domains in the second level; The step of obtaining the comprehensive score of the sample product in the current time period according to the comprehensive weight of each evaluation indicator and the score of each evaluation indicator in the current time period includes: Add the comprehensive weights of multiple evaluation indicators under each secondary capability domain to obtain the comprehensive weight of each secondary capability domain; Add the comprehensive weights of multiple second-level capability domains under each first-level capability domain to obtain the comprehensive weight of each first-level capability domain; The scores of multiple evaluation indicators under each secondary capability domain in the current time period are weighted and summed to obtain the score of each secondary capability domain; The scores of the multiple second-level competency domains under each first-level competency domain are weighted and summed to obtain the score of each first-level competency domain; and, According to the comprehensive weights and scores of the multiple first-level capability domains, a comprehensive score of the sample product in the current time period is obtained.

7. The method according to claim 1, characterized in that The score of each evaluation index of the sample product in the current time period is obtained as follows: Obtaining the value of each evaluation index of the sample product in the current time period; and, The values ​​of each evaluation indicator in the current time period are normalized to obtain the score of each evaluation indicator in the current time period.

8. The method according to claim 1, characterized in that The multiple preset product attributes include seasonal products, strategic products, star products, keep-on-shelf products, and eliminated products. Performing cluster analysis on the comprehensive scores of the sample products of the multiple preset product attributes and the function of time to determine the attributes of the candidate products in the multiple preset product attributes includes: Obtaining the distance between the function of the comprehensive score of sample products with multiple preset product attributes and time and the distance between their derivative functions; The distances between the functions and the distances between their derivative functions are weighted and summed to obtain weighted distance functions of sample products with multiple preset product attributes; Select five sample products, which are seasonal products, strategic products, star products, products to be kept on the shelves, and products to be eliminated, as the initial cluster centers of the weighted distance function, and calculate the distances between the candidate products and the initial cluster centers of the five sample products; and The attribute corresponding to the sample product that is closest to the initial cluster center of the five sample products is determined as the attribute of the candidate product.

9. A system for product selection optimization, characterized in that: The system comprises: A first determination module, configured to determine a comprehensive weight of each evaluation indicator among a plurality of evaluation indicators of the sample product according to a subjective weight and an objective weight of each evaluation indicator; The first acquisition module is used to obtain the score of each evaluation indicator of the sample product in the current time period; A second acquisition module, configured to obtain a comprehensive score of the sample product in the current time period according to the comprehensive weight of each evaluation indicator and the score of each evaluation indicator in the current time period; A third acquisition module is used to obtain a function of the comprehensive score of the sample product and time according to the comprehensive score of the sample product in the current time period and the historical comprehensive score in a set time period before the current time period; and The second determination module performs cluster analysis on the comprehensive scores of multiple sample products of multiple preset product attributes and the function of time to determine the attributes of the candidate product among the multiple preset product attributes.

10. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method for product selection optimization as described in any one of claims 1-8.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method for product selection optimization as described in any one of claims 1 to 8.

12. A computer program product, characterized in that It comprises a computer program which, when executed by a processor, implements the method for product selection optimization as described in any one of claims 1 to 8.

Images