Data analysis and optimization control method in enterprise supply chain management

Abstract

The invention relates to the technical field of data processing, and particularly discloses a data analysis and optimization control method in enterprise supply chain management. The method mainly comprises the steps of obtaining historical data of a product Ci; and judging whether the historical data of the product Ci covers a time range exceeding one year, and if so, calculating the profit growth rate Gi of the product Ci based on the profit value Pij of the product Ci in the historical time period t1j. According to the method, the product with the maximum profit increase potential is identified through deep analysis and prediction of historical data, finally, under the condition that manpower and financial resources are limited, it is ensured that the turnover rate of the high-profit-value product is kept in the optimal state, the holding cost is reduced, the capital turnover rate is increased, the market response speed is increased, the competitiveness is enhanced, and the market competitiveness is improved. And finally, the overall profitability is improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a data analysis and optimization control method in enterprise supply chain management. Background Art

[0002] Through close cooperation with suppliers, enterprises can achieve timely replenishment, improve supply chain flexibility, establish reasonable inventory levels, and monitor inventory turnover and sales in real time. While meeting customer needs, they can also reduce capital occupation, improve customer satisfaction, and enhance market competitiveness.

[0003] High turnover has significant advantages in corporate supply chain management, such as improving liquidity, reducing holding costs, responding quickly to market changes, improving customer satisfaction, and optimizing operational efficiency. However, with a wide variety of products and sales affected by seasonality, if human and financial resources are insufficient, it may not be possible to ensure high turnover for all products. This situation can easily lead to uneven resource allocation, resulting in out-of-stock of some high-demand products, affecting sales and customer experience. At the same time, some products may face inventory backlogs, increasing holding costs, and ultimately having a negative impact on overall profitability. Summary of the invention

[0004] This application provides a data analysis and optimization control method in enterprise supply chain management, which solves the problem in the prior art that, when there are many product types and sales are affected by seasonality, if human and financial resources are insufficient, it may not be possible to ensure a high turnover rate for all products. It ensures that the turnover rate of high-profit value products remains at the optimal state, reduces holding costs, and improves capital turnover, which helps to improve market response speed, enhance competitiveness, and ultimately achieve an improvement in overall profitability.

[0005] In the first aspect, the present application provides a data analysis and optimization control method in enterprise supply chain management, which comprises the following steps: obtaining product C i The historical data includes: Product C i In the historical time period t1 j The profit value P ij 、Product C i In the historical time period t1 j Daily sales data D ijk , where C i represents the i-th product, t1 j represents the jth time period in history, P ij represents the profit value of the i-th product in the j-th historical time period, D ijk represents the sales volume of the i-th product on the k-th day of the j-th historical time period; iDoes the historical data cover a period of more than one year? If so, based on product C i In the historical time period t1 j The profit value P ij Computing Products C i The profit growth rate G i , where G i represents the profit growth rate of the i-th product; obtain the predicted product, which is a product that meets at least one of the following a and b: a. In the historical time period t1 j The profit value P ij b. The n1 products with the highest profit value; b. The n2 products with the highest profit growth rate and the highest profit value; Use LSTM model to process and predict the product in the historical time period t1 j Daily sales data D ijk , get the predicted product C1 i1 In the future time period t p Daily sales data D i1pq , where C1 i1 represents the i1th prediction product, t p represents the pth time period in the future, D i1pq represents the sales volume of the i1th predicted product on the qth day of the pth future time period; based on the daily sales data D i1pq Calculate C1 for all predicted products i1 In the future time period t p The estimated profit value P; Get the predicted product C with the largest estimated profit value P max , calculate the predicted product C max In the future time period t p The best inventory BI within; based on the best inventory BI, calculate the forecast product C max The turnover rate TO.

[0006] Furthermore, the profit growth rate G i for: Among them, P i(j-n) Represents the profit value of the ith product in the (jn)th historical time period.

[0007] Furthermore, the historical data also includes: predicted product C1 i1 The sales price R of each predicted product C1 i1 Cost c; calculate the forecast product C1 i1 In the future time period t p Total sales volume within D t , D t for Where m represents the number of days in the future time period; the forecast product C1 i1In the future time period t p The estimated profit value P is: P = (Rc) · D t .

[0008] Further, the predicted product C is calculated max In the future time period t p Best in class inventory BI includes: Calculate forecasts for product C max In the future time period t p Safety stock SS, safety stock SS is: SS = Where Z represents the service level factor, σ d represents the standard deviation of daily demand, L represents the replenishment lead time in days; based on the forecast product C max Daily sales data D maxpq Calculate the prediction product C max In the future time period t p Demand within D t , where D maxpq Represents the predicted product C max Demand on the qth day of the pth future time period; obtain the forecast product C max In the future time period t p The fixed cost S for each order and C for each forecast product max In the future time period t p The holding cost H in the inventory; calculate the economic order quantity EOQ, EOQ is: Calculate the prediction product C max The optimal inventory BI is: BI=EOQ+SS.

[0009] Furthermore, D t for: Among them, D max,q Represents the predicted product C max The sales volume on the qth day in the future time period;

[0010] Furthermore, the turnover rate TO is:

[0011] Furthermore, if product C i If the historical data does not cover a period of more than one year, then the product C i Compared with similar products C z The similarity S of the same product is the highest similarity S. z To replace product C i .

[0012] Furthermore, the similarity S is: Where A=[a1,a2,a3,…,a n],B=[b1,b2,b3,…,b n ], a n and b n Represents product C i and product C z The characteristic vector of

[0013] Furthermore, the feature vector includes at least one of the following: sales volume, profit, category, brand, and specification.

[0014] Furthermore, the time period is 7 days or more.

[0015] In the second aspect, the present application provides a data analysis and optimization control system in enterprise supply chain management, which adopts the data analysis and optimization control method in enterprise supply chain management described in the first aspect, and includes: a data acquisition module, a growth rate calculation module, a predicted product acquisition module, a predicted product determination module, a model prediction module, an estimated profit value calculation module and an optimal inventory and turnover rate calculation module.

[0016] Data acquisition module, which is used to obtain product C i The historical data includes: Product C i In the historical time period t1 j The profit value P ij 、Product C i In the historical time period t1 j Daily sales data D ijk , where C i represents the i-th product, t1 j represents the jth time period in history, P ij represents the profit value of the i-th product in the j-th historical time period, D ijk Represents the sales volume of the ith product on the kth day of the jth historical time period.

[0017] Growth rate calculation module, which is used to determine product C i Does the historical data cover a period of more than one year? If so, based on product C i In the historical time period t1 j The profit value P ij Computing Products C i The profit growth rate C i , where G i Represents the profit growth rate of the ith product.

[0018] The forecast product determination module is used to obtain the forecast product, which is a product that meets at least one of the following a and b: a. In the historical time period t1 j The profit value P ij b. the n1 products with the highest profit value; b. the n2 products with the highest profit growth rate whose profit value is greater than the preset value.

[0019] The model prediction module is used to use the LSTM model to process the prediction product in the historical time period t1 j Daily sales data D ijk , get the predicted product C1 i1 In the future time period t p Daily sales data D i1pq , where C1 i1 represents the i1th prediction product, t p represents the pth time period in the future, D i1pq Represents the sales volume of the i1th predicted product on the qth day of the pth future time period.

[0020] The estimated profit value calculation module is used to calculate the estimated profit value based on the daily sales data D i1pq Calculate C1 for all predicted products i1 In the future time period t p The estimated profit value P.

[0021] The optimal inventory and turnover rate calculation module is used to calculate the forecast product C max In the future time period t p Optimal inventory BI within; Calculate forecast product C based on optimal inventory BI max The turnover rate TO.

[0022] In a third aspect, the present application provides a data analysis and optimization control device in enterprise supply chain management, which includes: a memory and a processor.

[0023] The memory is used to store computer programs; the processor is used to implement the steps of the data analysis and optimization control method in enterprise supply chain management as described in the first aspect when executing the computer program.

[0024] The technical solution provided by this application has at least the following technical effects or advantages:

[0025] Through in-depth analysis and prediction of historical data, the solution can identify products with the greatest potential for profit growth and calculate the optimal inventory BI and turnover rate TO based on these predicted data. This effectively solves the problem that when there are many product types and sales are affected by seasonality, it may not be possible to ensure high turnover rates for all products due to insufficient human and financial resources. Through precise resource allocation, the turnover rate of high-profit value products is ensured to be maintained at the optimal state, reducing holding costs and improving capital turnover rate, which helps to improve market response speed, enhance competitiveness, and ultimately achieve an improvement in overall profitability. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a flow chart of a data analysis and optimization control method in enterprise supply chain management in Embodiment 1 of the present application;

[0027] Figure 2 For this application Figure 1 Schematic diagram of the process of calculating the optimal inventory. DETAILED DESCRIPTION

[0028] To solve the problems raised in the background technology, this application, through in-depth analysis and prediction of historical data, can identify products with the greatest potential for profit growth, and calculate the optimal inventory BI and turnover rate TO based on these predicted data. Ultimately, under the condition of limited human and financial resources, by optimizing inventory levels, it is possible to reduce holding costs, avoid waste of resources, and achieve a more even inventory distribution.

[0029] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0030] Embodiment 1:

[0031] like Figure 1-Figure 2 As shown, this embodiment provides a data analysis and optimization control method in enterprise supply chain management, which includes the following steps:

[0032] S100. Get product C i The historical data includes: Product C i In the historical time period t1 j The profit value P ij 、Product C i In the historical time period t1 j Daily sales data D ijk , where C i represents the i-th product, t1 j represents the jth time period in history, P ij represents the profit value of the i-th product in the j-th historical time period, D ijkRepresents the sales volume of the ith product on the kth day of the jth historical time period.

[0033] In S100, historical data of products are collected. These data include profit values ​​and daily sales data of different products in different historical time periods. Through these detailed historical data, we can gain in-depth insights into the past performance of products and provide solid data support for subsequent market forecasts, strategic planning and decision-making.

[0034] In S100, the time period is 7 days or longer. A longer time period can provide richer historical data, which helps to more accurately capture the sales volume and profit trends of products, thereby improving the stability and reliability of the forecasting model. A time period of 7 days or longer matches the company's sales settlement cycle, which facilitates the combination of forecast results with actual business operations and achieves more refined inventory management and sales strategy adjustments. In addition, a longer period can also reduce misjudgments caused by short-term fluctuations.

[0035] S200. Determine product C i Does the historical data cover a period of more than one year? If so, based on product C i In the historical time period t1 j The profit value P ij Computing Products C i The profit growth rate G i , where G i Represents the profit growth rate of the ith product.

[0036] In S200, when the historical data of a product covers more than one year, the enterprise can obtain rich sales and profit information, which provides a solid foundation for data analysis. Sufficient historical data can not only help the enterprise identify seasonal changes and market trends, but also reveal the performance of products in different time periods. This in-depth analysis enables the enterprise to more accurately predict future sales, thereby optimizing inventory management and resource allocation.

[0037] In S200, by calculating the profit growth rate, companies can accurately evaluate changes in the profitability of their products. The profit growth rate calculation enables companies to clearly identify products with significant growth potential, which show a large room for growth. On this basis, companies can predict that in the future, these products with high growth potential are likely to become their most profitable product lines. Given their broad profit prospects, these products may become high-turnover products in the company's operations to maximize their market value and economic benefits.

[0038] S300. Obtain a predicted product, where the predicted product is a product that satisfies at least one of the following a and b:

[0039] a. In historical time period t1 j The profit value P ij Highest n1 products.

[0040] b. The profit value of the product is greater than the preset value, and the n2 products with the highest profit growth rate.

[0041] In S300, the predicted product is determined based on a or b, and the commodities that may generate the highest profit in the future are pre-identified and screened out so that a more accurate prediction analysis can be performed later using the prediction model.

[0042] First, we examine the profit performance of each product over its historical time period, especially those products that have shown high profit peaks in historical records and have the potential for sustained growth. These products have proven their profitability in past market competition, and their growth trends also indicate that they may continue to maintain high profit levels in the future. This criterion aims to narrow the forecast range and only take into consideration those options that are most likely to become high-profit products in the future.

[0043] In addition to paying attention to historical profit peaks, the present invention also sets a clear preset profit value threshold, and on this basis further screens out those commodities whose profit growth rate is higher than the industry average and reaches or exceeds the preset profit standard. This criterion ensures that the selected commodities are not only profitable in the current market environment, but also have strong growth momentum, indicating that there is greater room for profit improvement in the future.

[0044] By setting specific screening criteria, the products with the greatest potential to generate the greatest profits can be quickly identified, and more focus can be placed on those options that have been strictly screened and are most likely to become high-profit commodities in the future. Compared with traditional comprehensive forecasting methods, the present invention focuses more on those options that have been strictly screened and are most likely to become high-profit commodities in the future, which can effectively reduce the amount of calculation, thereby improving forecasting efficiency and accuracy. This not only saves valuable time and resources for enterprises, but also enables enterprises to concentrate on in-depth analysis and optimization of high-potential commodities, so as to occupy a favorable position in future market competition and achieve sustainable profit growth.

[0045] S400. Use LSTM model to process the predicted product in the historical time period t1 j Daily sales data D ijk , get the predicted product C1 i1 In the future time period t p Daily sales data D i1pq , where C1 i1 represents the i1th prediction product, t p represents the pth time period in the future, D i1pqRepresents the sales volume of the i1th predicted product on the qth day of the pth future time period.

[0046] In S400, the LSTM model can capture the long-term dependencies in the data. The LSTM model is used to analyze the historical time period t1. j The daily sales data within t1 is processed to predict the future daily sales. During training, the historical daily sales data t1 j Preprocessing is performed. The LSTM model includes an input layer, an LSTM layer, and an output layer. The LSTM layer can capture time dependencies. The training data set is used for model training. The model parameters are updated through the back propagation algorithm and the optimizer. The loss function is used to evaluate the gap between the prediction and the actual value. After the training is completed, the model performance is evaluated through the validation set to ensure its generalization ability. Finally, the trained model is used to predict future sales. The feature data can be input to output the predicted product C1 i1 In the future time period t p Daily sales data D i1pq .

[0047] S500. Based on daily sales data D i1pq Calculate C1 for all predicted products i1 In the future time period t p The estimated profit value P.

[0048] In S500, by calculating all predicted products C1 i1 In the future time period t p The estimated profit value P can be used to select the product with the largest profit value.

[0049] S600. Obtain the predicted product C with the maximum estimated profit value P max , calculate the predicted product C max In the future time period t p Optimal inventory BI within; Calculate forecast product C based on optimal inventory BI max The turnover rate TO.

[0050] In S600, in the actual operating environment with limited human resources, the optimal inventory level BI is accurately calculated and maintained for key products with the largest estimated profit value P, to ensure that the turnover rate TO of these high-profit value products remains at the optimal state, aiming to minimize holding costs and improve operational efficiency through refined inventory management, thereby maximizing profits under resource-constrained conditions. By maintaining a high turnover rate TO, we ensure that these high-profit value products can be quickly converted from inventory to sales, reducing inventory backlogs and capital occupation, thereby reducing holding costs, which not only accelerates the return of funds, but also provides enterprises with more financial flexibility to cope with market fluctuations and seize new business opportunities.

[0051] In S200, the profit growth rate G i for: Among them, P i(j-n) It represents the profit value of the ith product in the (jn)th historical time period, thus being able to intuitively reflect the change in profitability of the product in a specific time period and facilitate identification of growth trends.

[0052] In S500, the historical data also includes: predicted product C1 i1 The sales price R of each predicted product C1 i1 The cost c; the calculation of the estimated profit value P includes the following steps:

[0053] S510. Calculate and predict product C1 i1 In the future time period t p Total sales volume in D t , D t for: Here, m represents the number of days in the future time period.

[0054] S520. Prediction product C1 i1 In the future time period t p The estimated profit value P is: P = (Rc) · D t .

[0055] Use historical data to accurately assess the profitability of the forecasted product. The historical data covers the forecasted product C1 i1 The sales price R of each predicted product C1 i1 The cost c is calculated by summarizing the future time periods t p The sales data within the period is calculated using the formula P = (Rc)·D t The calculation can directly reflect the relationship between sales price and cost, making it easier to evaluate profitability.

[0056] In S600, the predicted product C is calculated. max In the future time period t p The best inventory BI includes:

[0057] S610. Calculate and predict product C max In the future time period t p The safety stock SS is: Where Z represents the service level factor, σ d represents the standard deviation of daily demand, and L represents the replenishment lead time in days.

[0058] In S610, safety stock SS is the additional inventory set to cope with demand fluctuations and supply uncertainties, Z represents the extent to which the enterprise hopes to meet customer demand, and σ d It reflects the volatility of daily sales. L represents the time required from placing an order to receiving the goods. By calculating the safety stock, sufficient inventory can be maintained during the replenishment period, thereby improving customer satisfaction and reducing sales losses caused by out-of-stock.

[0059] S620. Based on the prediction product C max Daily sales data D maxpq Calculate the prediction product C max In the future time period t p Demand within D t , where D maxpq Represents the predicted product C max The demand on the qth day of the pth future time period.

[0060] In S620, D t for: Among them, D max,q Represents the predicted product C max The sales volume on the qth day in the future time period.

[0061] S630. Obtain predicted product C max In the future time period t p The fixed cost S for each order and C for each forecast product max In the future time period t p The holding cost H in the inventory; calculate the economic order quantity EOQ, EOQ is: Calculate the prediction product C max The optimal inventory BI is: BI=EOQ+SS.

[0062] In S630, the fixed cost S of each order of the predicted product in the future time period can cover the order processing fee, transportation fee and other fixed expenses directly related to the order. The holding cost H of each predicted product in the future time period involves the storage fee, insurance premium, depreciation loss and all expenses related to inventory maintenance.

[0063] In S630, Economic Order Quantity (EOQ) is a classic inventory management model. Its core purpose is to help enterprises determine an optimal order quantity in order to find a balance between ordering cost and holding cost. Through the EOQ model, the number of products that should be purchased for each order can be calculated, so as to minimize costs while ensuring supply continuity. This model fully considers the trade-off between ordering cost and holding cost, allowing enterprises to manage their inventory resources more effectively while meeting market demand.

[0064] After determining the economic order quantity EOQ, the best inventory BI (Best Inventory) of the predicted product can be further calculated. The best inventory refers to the inventory that the company should hold in the future to meet market demand and minimize costs. It is equal to the sum of the economic order quantity EOQ and the safety stock SS.

[0065] In S600, the turnover rate TO is:

[0066] The uncertainty of market demand and seasonal changes in product sales are taken into account by setting a dynamic safety stock, so that the turnover rate can be predicted in advance and adapt to the sales trend of the highest-profit commodities, and the inventory turnover strategy can be determined more accurately, which not only avoids the cost increase caused by inventory backlogs, but also ensures sufficient supply of inventory during critical periods. This turnover rate calculation method not only improves the efficiency of inventory management, but also provides strong support for enterprises to seize market opportunities and optimize resource allocation, ensuring that enterprises can flexibly respond to seasonal fluctuations in product sales and maximize profits.

[0067] In S200, if product C i If the historical data does not cover a period of more than one year, then the product C i Compared with similar products C z The similarity S of the same product is the highest similarity S. z To replace product C i .

[0068] By comparing products C i Compared with similar products C z The similarity S can select the product C that is closest to the product. z , when product C i When historical data is insufficient, use similar products C z Rich historical data can indirectly predict product C iBy understanding the future performance of the company, we can accurately identify those products that are likely to become the most profitable in the future, manage these high-potential products in a timely and efficient manner, implement high-turnover strategies to meet market demand, reduce inventory costs, improve sales efficiency and customer satisfaction, and ultimately drive a substantial increase in corporate profitability.

[0069] Among them, the similarity S is: Where A=[a1,a2,a3,…,a n ],B=[b1,b2,b3,…,b n ], a n and b n Represents product C i and product C z The characteristic vector of A·B=a1b1+a2b2+a3b3+…+a n b n , The feature vector includes at least one of the following: sales volume, profit, category, brand, and specification.

[0070] Use cosine similarity to evaluate product C i With C z It combines multi-dimensional feature vectors, such as sales volume, profit, category, brand and specification, which together constitute a comprehensive portrait of product market performance, improve the accuracy of similarity assessment between products, and enhance the scientific nature of decision-making in data-scarce situations.

[0071] Embodiment 2:

[0072] This embodiment provides a data analysis and optimization control system in enterprise supply chain management, which includes: a data acquisition module, a growth rate calculation module, a predicted product acquisition module, a predicted product determination module, a model prediction module, an estimated profit value calculation module and an optimal inventory and turnover rate calculation module.

[0073] Data acquisition module, which is used to obtain product C i The historical data includes: Product C i In the historical time period t1 j The profit value P ij 、Product C i In the historical time period t1 j Daily sales data D ijk , where C i represents the i-th product, t1 j represents the jth time period in history, P ij represents the profit value of the i-th product in the j-th historical time period, D ijkRepresents the sales volume of the ith product on the kth day of the jth historical time period.

[0074] Growth rate calculation module, which is used to determine product C i Does the historical data cover a period of more than one year? If so, based on product C i In the historical time period t1 j The profit value P ij Computing Products C i The profit growth rate G i , where G i Represents the profit growth rate of the ith product.

[0075] The forecast product determination module is used to obtain the forecast product, which is a product that meets at least one of the following a and b: a. In the historical time period t1 j The profit value P ij b. the n1 products with the highest profit value; b. the n2 products with the highest profit growth rate whose profit value is greater than the preset value.

[0076] The model prediction module is used to use the LSTM model to process the prediction product in the historical time period t1 j Daily sales data D ijk , get the predicted product C1 i1 In the future time period t p Daily sales data D i1pq , where C1 i1 represents the i1th prediction product, t p represents the pth time period in the future, D i1pq Represents the sales volume of the i1th predicted product on the qth day of the pth future time period.

[0077] The estimated profit value calculation module is used to calculate the estimated profit value based on the daily sales data D i1pq Calculate C1 for all predicted products i1 In the future time period t p The estimated profit value P.

[0078] The optimal inventory and turnover rate calculation module is used to calculate the forecast product C max In the future time period t p Optimal inventory BI within; Calculate forecast product C based on optimal inventory BI max The turnover rate TO.

[0079] This embodiment has all the advantages of the data analysis and optimization control method in enterprise supply chain management in Embodiment 1, and can automatically implement the steps of the data analysis and optimization control method in enterprise supply chain management.

[0080] Embodiment three:

[0081] This embodiment provides a data analysis and optimization control device in enterprise supply chain management, which includes: a memory and a processor.

[0082] The memory is used to store computer programs; the processor is used to implement the steps of the data analysis and optimization control method in enterprise supply chain management in embodiment 1 when executing the computer program.

[0083] 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 in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps of the functions specified in a box or multiple boxes. Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they know the basic creative concept. Therefore, the attached claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

Claims

1. Data analysis and optimization control method in enterprise supply chain management, characterized by: The following steps are involved: Get Product C i The historical data includes: Product C i In the historical time period t1 j The profit value P ij 、Product C i In the historical time period t1 j Daily sales data D ijk , where C i represents the i-th product, t1 j represents the jth time period in history, P ij represents the profit value of the i-th product in the j-th historical time period, D ijk represents the sales volume of the i-th product on the k-th day in the j-th historical time period; Judging Product C i Does the historical data cover a period of more than one year? If so, based on product C i In the historical time period t1 j The profit value P ij Computing Products C i The profit growth rate G i , where G i represents the profit growth rate of the i-th product; Get the predicted product. The predicted product is the product that meets at least one of the following a and b: a. In historical time period t1 j The profit value P ij The highest n1 products; b. The profit value of the product is greater than the preset value, and the n2 products with the highest profit growth rate; Use LSTM model to process and predict products in historical time period t1 j Daily sales data D ijk , get the predicted product C1 i1 In the future time period t p Daily sales data D i1pq , where C1 i1 represents the i1th prediction product, t p represents the pth time period in the future, D i1pq represents the sales volume of the i1th predicted product on the qth day of the pth future time period; Based on the daily sales data D i1pq Calculate C1 for all predicted products i1 In the future time period t p The estimated profit value P; Get the predicted product C with the largest estimated profit value P max , calculate the predicted product C max In the future time period t p The best inventory BI within; based on the best inventory BI, calculate the forecast product C max The turnover rate TO.

2. The data analysis and optimization control method in enterprise supply chain management according to claim 1, characterized in that: Profit growth rate G i for: Among them, P i(j-n) Represents the profit value of the ith product in the (jn)th historical time period.

3. The data analysis and optimization control method in enterprise supply chain management according to claim 1, characterized in that: The historical data also includes: forecast product C1 i1 The sales price R of each predicted product C1 i1 The cost of c; Calculate the prediction product C1 i1 In the future time period t p Total sales volume within D t , D t for: Where m represents the number of days in the future time period; Prediction product C1 i1 In the future time period t p The estimated profit value P is: P = (Rc) · D t .

4. The data analysis and optimization control method in enterprise supply chain management according to claim 1, characterized in that: Calculate the prediction product C max In the future time period t p The best inventory BI includes: Calculate the prediction product C max In the future time period t p The safety stock SS is: Where Z represents the service level factor, σ d represents the standard deviation of daily demand, and L represents the replenishment lead time in days; Based on the prediction product C max Daily sales data D maxpq Calculate the prediction product C max In the future time period t p Demand within D t , where D maxpq Represents the predicted product C max The demand on the qth day of the pth future time period; Get predicted product C max In the future time period t p The fixed cost S for each order and C for each forecast product max In the future time period t p The holding cost H in the inventory; calculate the economic order quantity EOQ, EOQ is: Calculate the prediction product C max The optimal inventory BI is: BI=EOQ+SS.

5. The data analysis and optimization control method in enterprise supply chain management according to claim 4, characterized in that: D t for: Among them, D max,q Represents the predicted product C max The sales volume on the qth day in the future time period.

6. The data analysis and optimization control method in enterprise supply chain management according to claim 4, characterized in that: The turnover rate TO is:

7. The data analysis and optimization control method in enterprise supply chain management according to claim 1, characterized in that: If product C i If the historical data does not cover a period of more than one year, then the product C i Compared with similar products C z The similarity S of the same product is the highest similarity S. z To replace product C i .

8. The data analysis and optimization control method in enterprise supply chain management according to claim 7, characterized in that: The similarity S is: Where A=[a1,a2,a3,…,a n ],B=[b1,b2,b3,…,b n ], a n and b n Represents product C i and product C z The characteristic vector of A·B=a1b1+a2b2+a3b3+…+a n b n , 9. The data analysis and optimization control method in enterprise supply chain management according to claim 8, characterized in that: The feature vector includes at least one of the following: sales volume, profit, category, brand, and specification.

10. The data analysis and optimization control method in enterprise supply chain management according to claim 1, characterized in that: The time period is 7 days or more.

Images