Fast moving consumer goods market state evaluation method and system based on clustering algorithm

Abstract

The invention discloses a fast moving consumer goods market state evaluation method and system based on a clustering algorithm, and relates to the technical field of data analysis. In order to overcome the defect that traditional market state evaluation cannot comprehensively reflect the current situation of the market, an adopted scheme comprises the following steps: aiming at representative fast selling products in a specified area, acquiring four types of indexes of the fast selling products in a preset period, i.e., a sufficient reservation rate, a sufficient reservation surface, a retail price index and a social stock-sales ratio; performing normalization and interval preprocessing on the acquired four types of indexes; on the basis of the four types of preprocessed indexes, calculating the weight of the indexes by using a CRITIC algorithm, and constructing a weight vector; the weight vectors are clustered based on a clustering algorithm, and the fast moving consumer goods market state is divided. The method is used for visually reflecting the current market situation.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to a method and system for evaluating the market status of fast-moving consumer goods based on a clustering algorithm. Background Art

[0002] With the rapid development of the fast-moving consumer goods (FMCG) industry, real-time analysis and accurate evaluation of retail terminal data have become crucial for supply chain management and market decision-making. Traditional market status assessments rely primarily on manual experience or single indicators, which are unable to fully reflect market supply and demand relationships, price fluctuations, and sales activity.

[0003] While some research has used clustering algorithms to analyze the FMCG market, most haven't incorporated multi-dimensional key market indicators and lack standardized pre-processing procedures, resulting in unstable clustering results and difficulty in practical application. Therefore, there is an urgent need for a FMCG market status assessment method with a high degree of automation, rich evaluation dimensions, and reliable clustering results. Summary of the Invention

[0004] In response to the needs and deficiencies of current technological development, the present invention provides a method and system for evaluating the market status of fast-moving consumer goods based on a clustering algorithm.

[0005] In a first aspect, the present invention provides a method for evaluating the market status of fast-moving consumer goods based on a clustering algorithm. The technical solution adopted to solve the above technical problems is as follows: A method for evaluating the market status of fast-moving consumer goods (FMCG) based on a clustering algorithm comprises the following steps: S1. For representative fast-moving consumer goods in a designated area, collect four indicators: order-to-full rate, order-to-full surface, retail price index, and social inventory-to-sales ratio within a preset period; S2. Normalize and pre-process the four types of indicators collected; S3. Based on the preprocessed four indicators, the CRITIC algorithm is used to calculate their weights and construct a weight vector; S4. Cluster the weight vectors based on the K-means clustering algorithm to divide the fast-moving consumer goods market status.

[0006] Optionally, step S2 specifically includes: S2.1. For the four types of indicators collected, use Min-Max to normalize the data and map the indicator value range to the [0, 1] interval. The formula is as follows: , Where, Indicates the value of a certain type of indicator after data normalization; Indicates the original data value of this type of indicator; Indicates the minimum value of this type of indicator among all samples; Indicates the maximum value of this type of indicator in all samples; S2.2. For the normalized four index values, discretize them into five level intervals of "extremely high, high, moderate, low, and extremely low" to form an interval matrix , ,in, They correspond to one of the five level intervals of "extremely high, high, moderate, low, and extremely low" where the four types of indicator values are located.

[0007] Further optionally, step S3 specifically includes: S3.1. Based on the four types of indicators after preprocessing, calculate the standard deviation of each type of indicator , the formula is as follows: , Where, represents the standard deviation of the j-th indicator under the i-th sample; represents the normalized value of the j-th indicator under the i-th sample; n represents the number of samples; Represents the average value of the normalized value of the j-th indicator under all samples; S3.2. Use Pearson's correlation coefficient to calculate the correlation coefficient between any two indicators , and calculate the conflict degree between any two types of indicators , the formula is as follows: , Where, It represents the sum of the conflict degrees between the j-th indicator and all indicators; m represents the total type of indicators, with a value of 4; represents the correlation coefficient between the j-th indicator and the k-th indicator; S3.3, based on standard deviation and conflict Calculate the weight of each indicator , get the weight vector , the formula is as follows: , , Where, represents the standard deviation of the j-th category indicator; It represents the sum of the conflict degrees between the j-th category index and all category indexes; represents the comprehensive value of "contrast-conflict degree" of the j-th type of indicators; m represents the total type of indicators, with a value of 4; It represents the sum of the comprehensive values of “contrast-conflict degree” of the j-th indicator; The weight of each indicator Normalize the data to get the final weight .

[0008] Further optionally, step S4 specifically includes: S4.1. Integrate the preprocessed four indicators and weight vectors to construct a five-dimensional feature space containing indicator features and weight information, and set the number of clusters K = 5; S4.2. Use the Euclidean distance to measure the distance between different samples in the five-dimensional feature space and perform clustering using the K-means clustering algorithm: Initially, five cluster centers are randomly selected, the distance between each sample and each center is calculated and assigned to the closest cluster. The mean of each cluster is then recalculated as the new center, and the algorithm is repeated until the cluster centers converge. S4.3. Based on the clustering results and the different level intervals of the four indicators, classify the FMCG market status.

[0009] Preferably, step S4.3 is performed to divide the FMCG market status into five categories: "hot, tight, flat, loose, and soft" based on the clustering results and the different level intervals of the four indicators, where: When the FMCG market status is "hot", FMCG is extremely scarce, and the four indicators are ranked as follows: extremely high order-fill rate, extremely high order-fill surface, extremely high retail price index, and extremely low social stock-to-sales ratio; When the FMCG market status is "tight", the supply of FMCG is insufficient, and the level range of the four indicators is as follows: high order-filling rate, high order-filling area, high retail price index, and low social stock-to-sales ratio; When the FMCG market status is "flat", the supply and demand of FMCG is balanced, and the level ranges of the four indicators are: the order-fill rate, order-fill surface, retail price index and social inventory-to-sales ratio are all moderate; When the FMCG market is in the "loose" category, there is an oversupply of FMCG products, and the four indicators have the following rating ranges: low order-fill rate, low order-fill surface, low retail price index, and high social inventory-to-sales ratio; When the FMCG market status is "soft", there is a serious oversupply of FMCG, and the grade ranges of the four indicators are: extremely low order fulfillment rate, extremely low order fulfillment area, extremely low retail price index, and extremely high social inventory-to-sales ratio.

[0010] In a second aspect, the present invention provides a fast-moving consumer goods market status evaluation system based on a clustering algorithm. The technical solution adopted to solve the above technical problems is as follows: A fast-moving consumer goods market status evaluation system based on clustering algorithm, comprising: The data collection module is used to collect four indicators of representative fast-moving consumer goods in a specified area within a preset period, including order-full rate, order-full surface, retail price index, and social inventory-to-sales ratio; The preprocessing module is used to normalize and preprocess the four types of indicators collected; The weight calculation module is used to calculate the weights of the four types of indicators after preprocessing using the CRITIC algorithm and construct the weight vector; The K-means clustering module is used to cluster weight vectors based on the K-means clustering algorithm and divide the fast-moving consumer goods market status.

[0011] Optionally, the preprocessing module involved specifically includes a normalization unit and a binning unit; The normalization unit uses Min-Max to normalize the data for the four types of indicators collected, mapping the indicator value range to the [0, 1] interval. The formula is as follows: , Where, Indicates the value of a certain type of indicator after data normalization; Indicates the original data value of this type of indicator; Indicates the minimum value of this type of indicator among all samples; Indicates the maximum value of this type of indicator in all samples; For the four normalized index values, the interval units are discretized into five level intervals of "very high, high, moderate, low, and very low" to form an interval matrix , ,in, They correspond to one of the five level intervals of "extremely high, high, moderate, low, and extremely low" where the four types of indicator values are located.

[0012] Further optionally, the weight calculation module involved specifically performs the following operations: (1) Based on the four types of indicators after preprocessing, calculate the standard deviation of each type of indicator , the formula is as follows: , Where, represents the standard deviation of the j-th indicator under the i-th sample; represents the normalized value of the j-th indicator under the i-th sample; n represents the number of samples; Represents the average value of the normalized value of the j-th indicator under all samples; (2) Use Pearson correlation coefficient to calculate the correlation coefficient between any two types of indicators , and calculate the conflict degree between any two types of indicators , the formula is as follows: , Where, It represents the sum of the conflict degrees between the j-th indicator and all indicators; m represents the total type of indicators, with a value of 4; represents the correlation coefficient between the j-th indicator and the k-th indicator; (3) Based on standard deviation and conflict Calculate the weight of each indicator , get the weight vector , the formula is as follows: , , Where, represents the standard deviation of the j-th category indicator; It represents the sum of the conflict degrees between the j-th category index and all category indexes; represents the comprehensive value of "contrast-conflict degree" of the j-th type of indicators; m represents the total type of indicators, with a value of 4; It represents the sum of the comprehensive values of “contrast-conflict degree” of the j-th indicator; The weight of each indicator Normalize the data to get the final weight .

[0013] Optionally, the K-means clustering module may perform the following operations: (i) Integrate the preprocessed four indicators and the weight vector to construct a five-dimensional feature space containing indicator features and weight information, and set the number of clusters K = 5; (ii) The distance between different samples in the five-dimensional feature space is measured using the Euclidean distance, and clustering is performed using the K-means clustering algorithm: initially, five cluster centers are randomly selected, the distance between each sample and each center is calculated and assigned to the closest cluster, and the mean of each cluster is recalculated as the new center. This process is repeated until the cluster centers converge. (iii) Based on the clustering results and the different level intervals of the four indicators, the market status of fast-moving consumer goods is divided.

[0014] Preferably, step (iii) is performed to divide the FMCG market status into five categories: "hot, tight, flat, loose, and soft" based on the clustering results and the different level intervals of the four indicators, where: When the FMCG market status is "hot", FMCG is extremely scarce, and the four indicators are ranked as follows: extremely high order-fill rate, extremely high order-fill surface, extremely high retail price index, and extremely low social stock-to-sales ratio; When the FMCG market status is "tight", the supply of FMCG is insufficient, and the level range of the four indicators is as follows: high order-filling rate, high order-filling area, high retail price index, and low social stock-to-sales ratio; When the FMCG market status is "flat", the supply and demand of FMCG is balanced, and the level ranges of the four indicators are: the order-fill rate, order-fill surface, retail price index and social inventory-to-sales ratio are all moderate; When the FMCG market is in the "loose" category, there is an oversupply of FMCG products, and the four indicators have the following rating ranges: low order-fill rate, low order-fill surface, low retail price index, and high social inventory-to-sales ratio; When the FMCG market status is "soft", there is a serious oversupply of FMCG, and the grade ranges of the four indicators are: extremely low order fulfillment rate, extremely low order fulfillment area, extremely low retail price index, and extremely high social inventory-to-sales ratio.

[0015] Compared with the prior art, the present invention provides a method and system for evaluating the market status of fast-moving consumer goods based on a clustering algorithm. The system has the following beneficial effects: The present invention can more comprehensively depict the market status by integrating multi-dimensional indicators; automatically adjust the importance of indicators by calculating weights through the CRITIC algorithm to improve the objectivity of evaluation; and construct the five states of "tight, flat, loose and soft" through the K-means clustering algorithm to intuitively reflect the current market situation. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Attachment Figure 1 is a flow chart of a method according to embodiment 1 of the present invention; Attachment Figure 2 This is a module connection block diagram of the second embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the technical solution, the technical problems solved and the technical effects of the present invention more clear, the technical solution of the present invention is clearly and completely described below in conjunction with specific embodiments.

[0018] Example 1: Refer to the attached Figure 1 This embodiment proposes a method for evaluating the market status of fast-moving consumer goods based on a clustering algorithm, which includes the following steps: S1. For representative fast-moving consumer goods in a designated area, collect four indicators: order-fill rate, order-fill surface, retail price index and social inventory-to-sales ratio within a preset period.

[0019] S2. Normalize and pre-process the four types of indicators collected, including: S2.1. For the four types of indicators collected, use Min-Max to normalize the data and map the indicator value range to the [0, 1] interval. The formula is as follows: , Where, Indicates the value of a certain type of indicator after data normalization; Indicates the original data value of this type of indicator; Indicates the minimum value of this type of indicator among all samples; Indicates the maximum value of this type of indicator in all samples; S2.2. For the normalized four index values, discretize them into five level intervals of "extremely high, high, moderate, low, and extremely low" to form an interval matrix , ,in, They correspond to one of the five level intervals of "extremely high, high, moderate, low, and extremely low" where the four types of indicator values are located.

[0020] When dividing the grade intervals, the interval [0-0.2] is "very low", the interval (0.2-0.4] is "low", the interval (0.4-0.6] is "moderate", the interval (0.6-0.8] is "high", and the interval (0.8-1] is "very high".

[0021] S3. Based on the four pre-processed indicators, the CRITIC algorithm is used to calculate their weights and construct a weight vector, which includes: S3.1. Based on the four types of indicators after preprocessing, calculate the standard deviation of each type of indicator , the formula is as follows: , Where, represents the standard deviation of the j-th indicator under the i-th sample; represents the normalized value of the j-th indicator under the i-th sample; n represents the number of samples; Represents the average value of the normalized value of the j-th indicator under all samples; S3.2. Use Pearson's correlation coefficient to calculate the correlation coefficient between any two indicators , and calculate the conflict degree between any two types of indicators , the formula is as follows: , Where, It represents the sum of the conflict degrees between the j-th indicator and all indicators; m represents the total type of indicators, with a value of 4; represents the correlation coefficient between the j-th indicator and the k-th indicator; S3.3, based on standard deviation and conflict Calculate the weight of each indicator , get the weight vector , the formula is as follows: , , Where, represents the standard deviation of the j-th category indicator; It represents the sum of the conflict degrees between the j-th category index and all category indexes; represents the comprehensive value of "contrast-conflict degree" of the j-th type of indicators; m represents the total type of indicators, with a value of 4; It represents the sum of the comprehensive values of “contrast-conflict degree” of the j-th indicator; The weight of each indicator Normalize the data to get the final weight .

[0022] It should be added that CRITIC stands for CriteriaImportanceThroughIntercriteriaCorrelation, which means in Chinese a weight determination method based on the correlation between indicators. It is an objective weighting method that calculates weights by analyzing the comparative strength of indicators (the degree of data fluctuation) and the conflict between indicators (the size of the correlation) to avoid interference from subjective factors.

[0023] S4. Cluster the weight vectors based on the K-means clustering algorithm to classify the FMCG market status, including: S4.1. Integrate the preprocessed four indicators and weight vectors to construct a five-dimensional feature space containing indicator features and weight information, and set the number of clusters K = 5; S4.2. Use the Euclidean distance to measure the distance between different samples in the five-dimensional feature space and perform clustering using the K-means clustering algorithm: Initially, five cluster centers are randomly selected, the distance between each sample and each center is calculated and assigned to the closest cluster. The mean of each cluster is then recalculated as the new center, and the algorithm is repeated until the cluster centers converge. S4.3. Based on the clustering results and the different level ranges of the four indicators, the FMCG market status is divided into five categories: "hot, tight, flat, loose, and soft", among which: When the FMCG market is in the "hot" category, the four indicators are ranked as follows: extremely high order-fill rate, extremely high order-fill surface, extremely high retail price index, and extremely low social inventory-to-sales ratio; When the FMCG market is in the "tight" category, the rating ranges of the four indicators are: high order-filling rate, high order-filling surface, moderate retail price index, and moderate social inventory-to-sales ratio; When the FMCG market status is "flat", the level ranges of the four indicators are as follows: the order fulfillment rate, order fulfillment surface, retail price index and social inventory-to-sales ratio are all moderate; When the FMCG market status is "loose", the level ranges of the four indicators are: low order-fill rate, low order-fill surface, low retail price index, and high social stock-to-sales ratio; When the fast-moving consumer goods market status is "soft", the grade ranges of the four indicators are: extremely low order fulfillment rate, extremely low order fulfillment area, extremely low retail price index, and extremely high social inventory-to-sales ratio.

[0024] Example 2: Refer to the attached Figure 2 This embodiment proposes a fast-moving consumer goods market status evaluation system based on a clustering algorithm, which includes: The data collection module is used to collect four indicators of representative fast-moving consumer goods in a specified area within a preset period, including order-full rate, order-full surface, retail price index, and social inventory-to-sales ratio; The preprocessing module is used to normalize and preprocess the four types of indicators collected; The weight calculation module is used to calculate the weights of the four types of indicators after preprocessing using the CRITIC algorithm and construct the weight vector; The K-means clustering module is used to cluster weight vectors based on the K-means clustering algorithm and divide the fast-moving consumer goods market status.

[0025] It should be added that CRITIC stands for CriteriaImportanceThroughIntercriteriaCorrelation, which means in Chinese a weight determination method based on the correlation between indicators. It is an objective weighting method that calculates weights by analyzing the comparative strength of indicators (the degree of data fluctuation) and the conflict between indicators (the size of the correlation) to avoid interference from subjective factors.

[0026] In this embodiment, the preprocessing module specifically includes a normalization unit and a binning unit; The normalization unit uses Min-Max to normalize the data for the four types of indicators collected, mapping the indicator value range to the [0, 1] interval. The formula is as follows: , Where, Indicates the value of a certain type of indicator after data normalization; Indicates the original data value of this type of indicator; Indicates the minimum value of this type of indicator among all samples; Indicates the maximum value of this type of indicator in all samples; For the four normalized index values, the interval units are discretized into five level intervals of "very high, high, moderate, low, and very low" to form an interval matrix , ,in, They correspond to one of the five level intervals of "extremely high, high, moderate, low, and extremely low" where the four types of indicator values are located.

[0027] When dividing the grade intervals, the interval [0-0.2] is "very low", the interval (0.2-0.4] is "low", the interval (0.4-0.6] is "moderate", the interval (0.6-0.8] is "high", and the interval (0.8-1] is "very high".

[0028] In this embodiment, the weight calculation module performs the following operations: (1) Based on the four types of indicators after preprocessing, calculate the standard deviation of each type of indicator , the formula is as follows: , Where, represents the standard deviation of the j-th indicator under the i-th sample; represents the normalized value of the j-th indicator under the i-th sample; n represents the number of samples; Represents the average value of the normalized value of the j-th indicator under all samples; (2) Use Pearson correlation coefficient to calculate the correlation coefficient between any two types of indicators , and calculate the conflict between any two types of indicators , the formula is as follows: , Where, It represents the sum of the conflicts between the j-th indicator and all indicators of the same category; m represents the total type of indicators, and its value is 4; represents the correlation coefficient between the j-th indicator and the k-th indicator; (3) Based on standard deviation and conflict Calculate the weight of each indicator , get the weight vector , the formula is as follows: , , Where, represents the standard deviation of the j-th category indicator; It represents the sum of the conflicts between the j-th category index and all category indexes; represents the comprehensive value of "contrast-conflict degree" of the j-th type of indicators; m represents the total type of indicators, with a value of 4; It represents the sum of the comprehensive values of “contrast-conflict degree” of the j-th indicator; The weight of each indicator Normalize the data to get the final weight .

[0029] In this embodiment, the K-means clustering module specifically performs the following operations: (i) Integrate the preprocessed four indicators and the weight vector to construct a five-dimensional feature space containing indicator features and weight information, and set the number of clusters K = 5; (ii) The distance between different samples in the five-dimensional feature space is measured using the Euclidean distance, and clustering is performed using the K-means clustering algorithm: initially, five cluster centers are randomly selected, the distance between each sample and each center is calculated and assigned to the closest cluster, and the mean of each cluster is recalculated as the new center. This process is repeated until the cluster centers converge. (iii) Based on the clustering results and the different level ranges of the four indicators, the FMCG market status is divided into five categories: "hot, tight, flat, loose, and soft", among which: When the FMCG market status is "hot", FMCG is extremely scarce, and the four indicators are ranked as follows: extremely high order-fill rate, extremely high order-fill surface, extremely high retail price index, and extremely low social stock-to-sales ratio; When the FMCG market status is "tight", the supply of FMCG is insufficient, and the level range of the four indicators is as follows: high order-filling rate, high order-filling area, high retail price index, and low social stock-to-sales ratio; When the FMCG market status is "flat", the supply and demand of FMCG is balanced, and the level ranges of the four indicators are: the order-fill rate, order-fill surface, retail price index and social inventory-to-sales ratio are all moderate; When the FMCG market is in the "loose" category, there is an oversupply of FMCG products, and the four indicators have the following rating ranges: low order-fill rate, low order-fill surface, low retail price index, and high social inventory-to-sales ratio; When the FMCG market status is "soft", there is a serious oversupply of FMCG, and the grade ranges of the four indicators are: extremely low order fulfillment rate, extremely low order fulfillment area, extremely low retail price index, and extremely high social inventory-to-sales ratio.

[0030] In summary, the method and system for evaluating the market status of fast-moving consumer goods based on a clustering algorithm of the present invention can more comprehensively characterize the market status by integrating multi-dimensional indicators; automatically adjust the importance of indicators by calculating weights through the CRITIC algorithm to improve the objectivity of evaluation; and construct the five states of "tight, flat, loose and soft" through the K-means clustering algorithm to intuitively reflect the current market situation.

[0031] The above specific examples are used to illustrate the principles and implementation methods of the present invention in detail. These examples are only used to help understand the core technical content of the present invention. Based on the above specific embodiments of the present invention, any improvements and modifications made by those skilled in the art without departing from the principles of the present invention should fall within the scope of patent protection of the present invention.

Claims

1. A method for evaluating the market status of fast-moving consumer goods based on clustering algorithm, characterized by: The steps include: S1. For representative fast-moving consumer goods in a designated area, collect four indicators: order-to-full rate, order-to-full surface, retail price index, and social inventory-to-sales ratio within a preset period; S2. Normalize and pre-process the four types of indicators collected; S3. Based on the preprocessed four indicators, the CRITIC algorithm is used to calculate their weights and construct a weight vector; S4. Cluster the weight vectors based on the K-means clustering algorithm to divide the fast-moving consumer goods market status.

2. The method for evaluating the market status of fast-moving consumer goods based on clustering algorithm according to claim 1, characterized in that: The step S2 specifically includes: S2.

1. For the four types of indicators collected, use Min-Max to normalize the data and map the indicator value range to the [0, 1] interval. The formula is as follows: ， Where, Indicates the value of a certain type of indicator after data normalization; Indicates the original data value of this type of indicator; Indicates the minimum value of this type of indicator among all samples; Indicates the maximum value of this type of indicator in all samples; S2.

2. For the normalized four-category index values, discretize them into five level intervals: "very high, high, moderate, low, and very low" to form an interval matrix , ,in, They correspond to one of the five level intervals of "extremely high, high, moderate, low, and extremely low" where the four types of indicator values are located.

3. The method for evaluating the market status of fast-moving consumer goods based on clustering algorithm according to claim 2, characterized in that: The step S3 specifically includes: S3.

1. Based on the four types of indicators after preprocessing, calculate the standard deviation of each type of indicator , the formula is as follows: ， Where, represents the standard deviation of the j-th indicator under the i-th sample; represents the normalized value of the j-th indicator under the i-th sample; n represents the number of samples; Represents the average value of the normalized value of the j-th indicator under all samples; S3.

2. Use Pearson's correlation coefficient to calculate the correlation coefficient between any two indicators , and calculate the conflict degree between any two types of indicators , the formula is as follows: ， Where, It represents the sum of the conflict degrees between the j-th indicator and all indicators; m represents the total type of indicators, with a value of 4; represents the correlation coefficient between the j-th indicator and the k-th indicator; S3.3, based on standard deviation and conflict Calculate the weight of each indicator , get the weight vector , the formula is as follows: ， ， Where, represents the standard deviation of the j-th category indicator; It represents the sum of the conflict degrees between the j-th category index and all category indexes; represents the comprehensive value of "contrast-conflict" of the j-th type of indicators; m represents the total type of indicators, with a value of 4; It represents the sum of the comprehensive values of "contrast-conflict degree" of the j-th category indicator; The weight of each indicator Normalize the data to get the final weight .

4. The method for evaluating the market status of fast-moving consumer goods based on a clustering algorithm according to claim 3, characterized in that: The step S4 specifically includes: S4.

1. Integrate the preprocessed four indicators and weight vectors to construct a five-dimensional feature space containing indicator features and weight information, and set the number of clusters K = 5; S4.

2. Use the Euclidean distance to measure the distance between different samples in the five-dimensional feature space and perform clustering using the K-means clustering algorithm: Initially, five cluster centers are randomly selected, the distance between each sample and each center is calculated and assigned to the closest cluster. The mean of each cluster is then recalculated as the new center, and the algorithm is repeated until the cluster centers converge. S4.

3. Based on the clustering results and the different level intervals of the four indicators, classify the FMCG market status.

5. The method for evaluating the market status of fast-moving consumer goods based on clustering algorithm according to claim 4, characterized in that: Execute step S4.

3. Based on the clustering results and the different level ranges of the four indicators, the FMCG market status is divided into five categories: "hot, tight, flat, loose, and soft", where: When the FMCG market is in the "hot" category, FMCG products are extremely scarce, and the four indicators are ranked as follows: extremely high order-fill rate, extremely high order-fill surface, extremely high retail price index, and extremely low social stock-to-sales ratio. When the FMCG market is in the "tight" category, there is insufficient supply of FMCG products. The four indicators are ranked as follows: high order-fill rate, high order-fill surface, high retail price index, and low social stock-to-sales ratio. When the FMCG market is in the "flat" category, the supply and demand of FMCG is balanced, and the four indicators are rated as follows: the order-to-full rate, order-to-full surface, retail price index, and social inventory-to-sales ratio are all moderate. When the FMCG market is in the "loose" category, there is an oversupply of FMCG products, and the four indicators are ranked as follows: low order-fill rate, low order-fill surface, low retail price index, and high social stock-to-sales ratio. When the FMCG market status is "soft", there is a serious oversupply of FMCG, and the grade ranges of the four indicators are: extremely low order fulfillment rate, extremely low order fulfillment area, extremely low retail price index, and extremely high social inventory-to-sales ratio.

6. A fast-moving consumer goods market status evaluation system based on clustering algorithm, characterized by: It includes: The data collection module is used to collect four indicators of representative fast-moving consumer goods in a specified area within a preset period, including order-full rate, order-full surface, retail price index, and social inventory-to-sales ratio; The preprocessing module is used to normalize and preprocess the four types of indicators collected; The weight calculation module is used to calculate the weights of the four types of indicators after preprocessing using the CRITIC algorithm and construct the weight vector; The K-means clustering module is used to cluster weight vectors based on the K-means clustering algorithm and divide the fast-moving consumer goods market status.

7. The fast-moving consumer goods market status evaluation system based on clustering algorithm according to claim 6, characterized in that: The preprocessing module specifically includes a normalization unit and a binning unit; The normalization unit uses Min-Max to normalize the data for the four types of indicators collected, mapping the indicator value range to the [0, 1] interval. The formula is as follows: ， Where, Indicates the value of a certain type of indicator after data normalization; Indicates the original data value of this type of indicator; Indicates the minimum value of this type of indicator among all samples; Indicates the maximum value of this type of indicator in all samples; For the four normalized index values, the interval units are discretized into five level intervals of "very high, high, moderate, low, and very low" to form an interval matrix , ,in, They correspond to one of the five level intervals of "extremely high, high, moderate, low, and extremely low" where the four types of indicator values are located.

8. The fast-moving consumer goods market status evaluation system based on clustering algorithm according to claim 7, characterized in that: The weight calculation module specifically performs the following operations: (1) Based on the four types of indicators after preprocessing, calculate the standard deviation of each type of indicator , the formula is as follows: ， Where, represents the standard deviation of the j-th indicator under the i-th sample; represents the normalized value of the j-th indicator under the i-th sample; n represents the number of samples; Represents the average value of the normalized value of the j-th indicator under all samples; (2) Use Pearson correlation coefficient to calculate the correlation coefficient between any two types of indicators , and calculate the conflict degree between any two types of indicators , the formula is as follows: ， Where, It represents the sum of the conflict degrees between the j-th indicator and all indicators; m represents the total type of indicators, with a value of 4; represents the correlation coefficient between the j-th indicator and the k-th indicator; (3) Based on standard deviation and conflict Calculate the weight of each indicator , get the weight vector , the formula is as follows: ， ， Where, represents the standard deviation of the j-th category indicator; It represents the sum of the conflict degrees between the j-th category index and all category indexes; represents the comprehensive value of "contrast-conflict" of the j-th type of indicators; m represents the total type of indicators, with a value of 4; It represents the sum of the comprehensive values of "contrast-conflict degree" of the j-th category indicator; The weight of each indicator Normalize the data to get the final weight .

9. The fast-moving consumer goods market status evaluation system based on clustering algorithm according to claim 8, characterized in that: The K-means clustering module specifically performs the following operations: (i) Integrate the preprocessed four indicators and the weight vector to construct a five-dimensional feature space containing indicator features and weight information, and set the number of clusters K = 5; (ii) The Euclidean distance is used to measure the distance between different samples in the five-dimensional feature space, and clustering is performed using the K-means algorithm: initially, five cluster centers are randomly selected, the distance between each sample and each center is calculated and assigned to the closest cluster, and the mean of each cluster is recalculated as the new center. This process is repeated until the cluster centers converge. (iii) Based on the clustering results and the different level intervals of the four indicators, the market status of fast-moving consumer goods is divided.

10. The fast moving consumer goods market status evaluation system based on clustering algorithm according to claim 9, characterized in that: Execute step (iii) and classify the FMCG market into five categories: "hot, tight, flat, loose, and soft" based on the clustering results and the different level ranges of the four indicators. When the FMCG market is in the "hot" category, FMCG products are extremely scarce, and the four indicators are ranked as follows: extremely high order-fill rate, extremely high order-fill surface, extremely high retail price index, and extremely low social stock-to-sales ratio. When the FMCG market is in the "tight" category, there is insufficient supply of FMCG products. The four indicators are ranked as follows: high order-fill rate, high order-fill surface, high retail price index, and low social stock-to-sales ratio. When the FMCG market is in the "flat" category, the supply and demand of FMCG is balanced, and the four indicators are rated as follows: the order-to-full rate, order-to-full surface, retail price index, and social inventory-to-sales ratio are all moderate. When the FMCG market is in the "loose" category, there is an oversupply of FMCG products, and the four indicators are ranked as follows: low order-fill rate, low order-fill surface, low retail price index, and high social stock-to-sales ratio. When the FMCG market status is "soft", there is a serious oversupply of FMCG, and the grade ranges of the four indicators are: extremely low order fulfillment rate, extremely low order fulfillment area, extremely low retail price index, and extremely high social inventory-to-sales ratio.

Images