A smart supply chain management method based on artificial intelligence

Abstract

The present invention discloses an intelligent supply chain management method based on artificial intelligence, which relates to the technical field of supply chain management methods. From a system module, historical sales data, inventory data, supplier delivery time, and production cycle are extracted and collected through a data acquisition module, and historical sales data are modeled through the data acquisition module. Based on the forecast data and through a data analysis and evaluation module, the impact of different orders on inventory is evaluated, and urgent procurement needs are made based on the maximum out-of-stock quantity and the necessity of inventory adjustment is evaluated. According to the evaluation of inventory adjustment, and through an automated decision-making and execution module, the inventory level is intelligently adjusted, and the inventory status and supply chain performance indicators are monitored in real time. By improving the accuracy of demand forecasting, optimizing inventory management, strengthening supply chain collaboration, enhancing risk management capabilities, and innovating supply chain management models, significant economic and social benefits are brought to enterprises.

Description

Technical Field

[0001] The present invention relates to the technical field of supply chain management methods, and specifically to an intelligent supply chain management method based on artificial intelligence. Background Art

[0002] With the intensification of global market competition and the rapid changes in consumer demand, traditional supply chain management methods can no longer meet the operational needs of modern enterprises. The rapid development of artificial intelligence (AI) technology has brought revolutionary changes to supply chain management. By integrating AI technology, enterprises can achieve intelligent, automated and real-time management of the supply chain, improve the response speed of the supply chain, reduce costs, enhance risk resistance and improve overall competitiveness.

[0003] Traditional supply chain management methods often rely on empirical judgment or simple statistical models for demand forecasting, which is difficult to accurately reflect the complexity of market changes, resulting in frequent inventory backlogs and out-of-stock phenomena, inaccurate demand forecasts, and a lack of effective inventory optimization strategies, resulting in high inventory costs and excessive capital occupation, while affecting the flexibility and responsiveness of the supply chain, which in turn leads to inefficient inventory management. In addition, there is a serious phenomenon of information islands between various links in the supply chain, and there is a lack of effective information sharing and coordination mechanisms, which affects the overall efficiency and reliability of the supply chain. In addition, traditional supply chain management methods lack effective risk identification and response capabilities when facing emergencies and uncertain factors, resulting in an increased risk of supply chain disruptions. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent supply chain management method based on artificial intelligence to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above purpose, the present invention provides the following technical solutions, and the intelligent supply chain management method based on artificial intelligence is as follows: From the system module, the data collection module extracts and collects historical sales data, inventory data, supplier delivery time, and production cycle; Model the historical sales data through the data collection module to generate forecast data for demand forecasts in the future; Based on the forecast data, and through the data analysis and evaluation module, the impact of different orders on inventory is evaluated, urgent purchase needs are made based on the maximum stock-out quantity, and the necessity of inventory adjustment is evaluated. The data analysis and evaluation module includes an order filling optimization evaluation unit, a purchase demand management unit for maintaining inventory levels, and an inventory balance adjustment evaluation unit; Based on the evaluation of inventory adjustments, intelligently adjust inventory levels through automated decision-making and execution modules, specifically trigger replenishment orders, optimize inventory allocation, and monitor inventory status and supply chain performance indicators in real time; Synchronously upload the collected, predicted, evaluated and adjusted data to the database server; The equipment used in the system module includes ERP system, sales system, and inventory management system; The equipment used in the data acquisition module includes ETL tools; The equipment used in the data analysis and evaluation module includes a computing server and a GPU; The equipment used by the automated decision-making and execution module includes an automated warehouse system.

[0006] Optionally, the calculation formula of the order filling optimization evaluation unit is as follows: ; in: DY is the order filling optimization value; d is the total number of orders; DI a is the ath order priority, reflecting DI a Reflects the priority of orders, including orders from VIP customers and orders that are about to exceed the delivery deadline; DL a is the quantity value of the ath order; KD a is the filling value of the current inventory, KD a Reflects the current inventory for the ath order quantity DL a The number of occupied fills; YJ a To predict the value of arrival, YJ a The expected number of future orders is DL. a The number of occupants.

[0007] Optionally, the calculation formula of the purchase demand management unit under the maintenance of inventory level is as follows: ; AQ = (L*BC)*JH; in: CX is the purchase demand value; AQ is the additional inventory value; L is the critical value, which reflects the critical value of the standard normal distribution of the service level; BC is the standard deviation of demand forecast; JH is the delivery time, which refers to the total number of days from the time the order is placed to the actual delivery of the product to the customer; Max a (DL a -KD a) indicates that among all orders, the demand exceeds the maximum value of the current inventory.

[0008] Optionally, the calculation formula of the inventory balance adjustment evaluation unit is as follows: ; YK=(DK+CG); in: PT is the inventory balance adjustment value; YK is the expected inventory value; DK is the current inventory value, CG is the purchase quantity; Used to account for the relative sizes of order quantities and inventory quantities in inventory balancing adjustments.

[0009] Optionally, the inventory balance adjustment value PT output by the inventory balance adjustment evaluation unit is evaluated as follows: S1. Setting an expected range of inventory balance adjustment value PT through the system module to determine the inventory status; S2. If the inventory balance adjustment value PT is close to or exceeds the preset expected range, it is considered that there is excess inventory; S3. If the inventory balance adjustment value PT is far below the preset expected range and is insufficient to meet the demand in the future, the inventory is considered insufficient; S4. If the inventory balance adjustment value PT is within the expected range, the inventory is considered to be in a balanced state.

[0010] Optionally, the automated decision-making and execution module is adjusted based on the analysis of the inventory balance adjustment evaluation unit as follows: S1. First, observe whether the expected inventory value YK meets the demand in the future. If the expected inventory value YK does not meet the demand, further analyze which orders contribute the most, that is, which orders have the largest DL a -KD a The value of is the largest; S2. It should be noted that, among the analyzed orders, the orders whose demand exceeds the current inventory and whose delivery deadline is tight will have their priority increased; S3. If the inventory balance adjustment value PT shows insufficient inventory, and increasing the order priority is not enough to solve the problem, the purchase plan needs to be adjusted, specifically by increasing the purchase volume CG. At the same time, it is necessary to negotiate with the supplier to shorten the delivery time or find an alternative supplier; S4. The adjusted purchasing plan will affect the future expected inventory value YK, and further affect the subsequent order filling strategy, that is, the evaluation result of the subsequent order filling optimization evaluation unit.

[0011] Optionally, the additional inventory value AQ is specifically the inventory amount maintained in addition to the normal inventory, which is used to deal with uncertain factors, including demand fluctuations and supply delays, which may cause out-of-stock risks, and the additional inventory value AQ does not directly participate in meeting daily demand.

[0012] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention uses AI technology to conduct in-depth mining and analysis of multi-dimensional information on historical sales data, market trends, and consumer behavior to generate more accurate and timely demand forecasts. By continuously optimizing the forecasting model, the forecasting accuracy is improved, the risk of inventory backlogs and out-of-stocks is reduced, and customer satisfaction and loyalty are improved.

[0013] 2. The present invention dynamically adjusts inventory levels according to the actual situation of the supply chain and demand forecast results to ensure that inventory costs are reduced while meeting customer needs, and uses AI technology to achieve automated management and optimization of inventory. For example, the intelligent warehousing system guides robots to sort, transport and store goods, thereby improving operational efficiency and accuracy.

[0014] 3. The present invention establishes a unified digital platform to realize information sharing and collaborative operation among various links of the supply chain. Through the technical means of the Internet of Things and big data, it can obtain data information of various links of the supply chain in real time, strengthen the integration and allocation of internal and external resources, so as to promote collaborative cooperation among enterprises, suppliers, distributors, and logistics companies, thereby improving the response speed and flexibility of the supply chain.

[0015] 4. The present invention uses AI technology to monitor the operation of each link of the supply chain in real time, promptly detect abnormal situations and take corresponding countermeasures to ensure the stable operation of the supply chain. Through the technical means of smart contracts and blockchain, the transactions in the supply chain can be made transparent, traceable and tamper-proof, thereby reducing the risk of fraud and improving trust.

[0016] 5. The intelligent supply chain management method based on AI technology of the present invention breaks the limitations of traditional supply chain management, realizes the intelligent, automated and real-time management of the supply chain, and through continuous optimization and innovation, provides enterprises with more efficient, flexible and reliable supply chain management solutions, helping enterprises to improve their competitiveness and achieve sustainable development. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flowchart of the method of intelligent supply chain management method based on artificial intelligence; Figure 2 It is a structural schematic diagram of the data analysis and evaluation module of the present invention; Figure 3 This is a schematic diagram of the structure of each module of the intelligent supply chain management method based on artificial intelligence; Figure 4 Schematic diagram of the evaluation process of the inventory balance adjustment value of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] Regarding this intelligent supply chain management method based on artificial intelligence, it is different from traditional supply chain management methods. Traditional supply chain management methods often rely on experience judgment or simple statistical models for demand forecasting, which is difficult to accurately reflect the complexity of market changes, and at the same time affects the flexibility and response speed of the supply chain, which in turn leads to inefficient inventory management. In addition, there is a serious phenomenon of information islands between various links in the supply chain, and there is a lack of effective information sharing and coordination mechanisms, which affects the overall efficiency and reliability of the supply chain. In addition, traditional supply chain management methods lack effective risk identification and response capabilities when facing emergencies and uncertain factors, resulting in an increased risk of supply chain disruptions. The intelligent supply chain management method based on artificial intelligence in this algorithm unit effectively solves the shortcomings of traditional supply chain management methods by improving the accuracy of demand forecasting, optimizing inventory management, strengthening supply chain collaboration, enhancing risk management capabilities and innovating supply chain management models, and has brought significant economic and social benefits to enterprises.

[0020] For example, see Figures 1 to 4 , this implementation provides an intelligent supply chain management method based on artificial intelligence, and the intelligent supply chain management method based on artificial intelligence is as follows: From the system module, the data collection module extracts and collects historical sales data, inventory data, supplier delivery time, and production cycle; Model the historical sales data through the data collection module to generate forecast data for demand forecasts in the future; Based on the forecast data, the impact of different orders on inventory is evaluated through the data analysis and evaluation module, urgent purchase needs are made based on the maximum stock-out quantity, and the necessity of inventory adjustment is evaluated. The data analysis and evaluation module includes an order filling optimization evaluation unit, a purchase demand management unit to maintain inventory levels, and an inventory balance adjustment evaluation unit; Based on the evaluation of inventory adjustments, intelligently adjust inventory levels through automated decision-making and execution modules, specifically trigger replenishment orders, optimize inventory allocation, and monitor inventory status and supply chain performance indicators in real time; Synchronously upload the collected, predicted, evaluated and adjusted data to the database server; The equipment used in the system module includes ERP system, sales system, and inventory management system; The equipment used in the data collection module includes ETL tools; The equipment used in the data analysis and evaluation module includes computing servers and GPUs; The equipment used in the automated decision-making and execution module includes an automated warehouse system.

[0021] Database servers are used to store and manage supply chain data, computing servers are used to perform complex computing tasks such as demand forecasting and inventory optimization. If deep learning and high-performance computing technologies are used, GPU acceleration is required. ETL tools are used for data extraction, conversion and loading. Automated warehouse systems include automated storage equipment and handling robots to achieve automated inventory management.

[0022] In this embodiment, the system cooperates with each other through three algorithm units, and each of them undertakes important computing tasks in the intelligent supply chain management method based on artificial intelligence, and jointly promotes the optimization and upgrading of the supply chain through the cyclic influence between each other. Combining the three operation results of DY, CX and PT, it also brings significant economic and social benefits to the enterprise. DY is the order filling optimization value. Based on the priority, current inventory, expected arrival quantity and order demand of each order item, the optimal order filling sequence and quantity are determined through weighted calculation, which helps to maximize customer satisfaction and supply chain efficiency under limited resources. CX is the procurement demand value. Based on the order filling optimization value DY, combined with the additional inventory and the current The previous inventory level and the maximum shortage are used to calculate the purchasing demand in the future, which helps to ensure the stability and continuity of the supply chain. PT is the inventory balance adjustment value, which aims to calculate and adjust the inventory level by comprehensively considering multiple factors such as order quantity, current inventory, purchasing demand, expected inventory level and additional inventory, so as to achieve the best balance between inventory cost and meeting customer needs. The calculation result of PT can also affect the calculation of DY and CX. They are interrelated and influence each other, and jointly promote the optimization and upgrading of the supply chain, so that the three algorithms of this system have a high correlation and entanglement, so that the overall algorithm system can perform automatic feedback and optimization according to the actual situation to be closer to reality.

[0023] See also Figures 1 to 4 , the calculation formula of the order filling optimization evaluation unit is as follows: ; in: DY is the order filling optimization value; d is the total number of orders; DI a is the ath order priority, reflecting DI a Reflects the priority of orders, including orders from VIP customers and orders that are about to exceed the delivery deadline; DL a is the quantity value of the ath order; KD a is the filling value of the current inventory, KD a Reflects the current inventory for the ath order quantity DL a The number of occupied fills; YJ a To predict the value of arrival, YJ a The expected number of future orders is DL. a The number of occupants.

[0024] In this embodiment: First, the DI in this algorithm unit a is the ath order priority, which is a subjective value based on smart supply chain management rules, used to indicate the relative importance of different orders. The priority can be determined based on a variety of factors, such as the importance of the customer, the urgency of the order, the value of the order, and the delivery deadline. For example, orders from VIP customers have a higher priority, and orders that are about to exceed the delivery deadline will also be given a higher priority. In other words, in special cases, orders that are about to exceed the delivery deadline will have a higher priority than VIP customers' orders. Different orders need to be optimized at different levels according to their different situations. In addition, d is the total number of orders, that is, the number of orders that need to be processed at present, DL a is the quantity of the ath order, i.e. the quantity of goods requested by the customer, YJ a To predict the value of arrival, this is usually based on the supplier's delivery plan and historical data forecast, and finally evaluate the output of the a-th order priority DI a It is a value that comprehensively evaluates all order priorities and inventory status; This algorithm unit enables high-priority orders to be processed and shipped faster, reducing customer waiting time and thus improving customer satisfaction. By accurately calculating the filling priority of each order item, limited inventory and logistics resources can be allocated more effectively to avoid resource waste.

[0025] Enhanced supply chain flexibility: In the face of rapid changes in market demand, Formula One is able to dynamically adjust order filling strategies to make the supply chain more flexible and agile.

[0026] See also Figures 1 to 4 , the calculation formula of the purchase demand management unit to maintain the inventory level is as follows: ; AQ = (L*BC)*JH; in: CX is the purchase demand value; AQ is the additional inventory value; L is the critical value, which reflects the critical value of the standard normal distribution of the service level; BC is the standard deviation of demand forecast; JH is the delivery time, which refers to the total number of days from the time the order is placed to the actual delivery of the product to the customer; Max a (DL a -KD a ) indicates that among all orders, the demand exceeds the maximum value of the current inventory.

[0027] In this embodiment, first, AQ is the additional inventory value, which is usually based on historical demand data, demand fluctuation rate and supply delay time. It is the inventory amount maintained in addition to the normal inventory to cope with the risk of out-of-stock caused by uncertain factors, including demand fluctuation and supply delay. The safety inventory amount does not directly participate in the daily demand satisfaction, but plays a buffer role in emergency situations; Max a (DL a -KD a ) is used to find the maximum value of the demand that exceeds the current inventory among all orders. Specifically, for each order a, calculate DL a -KD a , which is the demand minus the current inventory, and then find the maximum of these differences, which represents the most urgent shortage situation at the moment, that is, which order needs to be replenished most immediately; This algorithm unit can avoid inventory backlogs and capital occupation caused by excessive purchasing by accurately calculating purchasing needs, thereby reducing inventory costs and achieving inventory cost control. Taking into account safety stock and maximum out-of-stock quantity, the purchasing demand management unit can ensure that there is enough inventory to meet customer needs when demand fluctuates and reduce out-of-stock risks. Reasonable purchasing plans can reduce unnecessary logistics and warehousing costs and improve the overall efficiency of the supply chain.

[0028] See also Figures 1 to 4 , the calculation formula for the inventory balance adjustment valuation unit is as follows: ; YK=(DK+CG); in: PT is the inventory balance adjustment value; YK is the expected inventory value; DK is the current inventory value, CG is the purchase quantity; Used to account for the relative sizes of order quantities and inventory quantities in inventory balancing adjustments.

[0029] In this embodiment, the algorithm unit first evaluates and manages the order filling optimization evaluation unit and the purchase demand management unit under the inventory level maintenance, YK is the expected inventory value, that is, the inventory level after the current inventory plus the purchase volume, It is the sum of the square of the order quantity of all orders and the square of the current inventory quantity, which is used to consider the relative size of the order quantity and the inventory quantity in the inventory balance adjustment. Finally, the inventory balance adjustment value PT output after the adjustment evaluation can be used as a feedback signal to adjust the a-th order priority DI in the order filling optimization evaluation unit. a Specifically, if the inventory balance adjustment value PT indicates that the expected inventory level after inventory balance adjustment is still insufficient to meet the needs of some high-priority orders, then the order priority DI of these orders can be increased by a To ensure that they are processed with priority. In addition, if the inventory balance adjustment value PT shows excess or shortage of inventory, future purchase plans will be adjusted, thus indirectly affecting the order filling strategy. The effect of feedback is to make the supply chain management system more flexible and responsive. By constantly adjusting order priorities and purchasing plans based on inventory conditions and order demand, it can ensure that key orders are met in a timely manner while avoiding inventory backlogs and out-of-stock situations. This feedback mechanism helps to achieve a balance between minimizing inventory costs and maximizing customer service levels. By adjusting the inventory level in this algorithm unit, it is possible to optimize the inventory structure, reduce unsalable products and backlog inventory, improve inventory turnover, achieve the purpose of optimizing the inventory structure, and reduce unnecessary inventory occupancy, thereby releasing more funds for other investment and operating activities, improving the efficiency of capital utilization, and maintaining sufficient inventory to deal with emergencies while avoiding the risks brought by excessive inventory, thereby enhancing the risk resistance of the supply chain.

[0030] See also Figures 1 to 4 , the evaluation of the inventory balance adjustment value PT output by the inventory balance adjustment evaluation unit is as follows: S1. Setting an expected range of inventory balance adjustment value PT through the system module to determine the inventory status; S2. If the inventory balance adjustment value PT exceeds the preset expected range, it is considered that the inventory is in excess; S3. If the inventory balance adjustment value PT is lower than the preset expected range, and the expected inventory value YK is insufficient to meet the demand in the future, the inventory is considered insufficient; S4. If the inventory balance adjustment value PT is within the expected range, the inventory is considered to be in a balanced state; The automated decision-making and execution module makes the following adjustments based on the analysis of the inventory balance adjustment evaluation unit: S1. First, observe whether the expected inventory value YK meets the demand in the future. If the expected inventory value YK does not meet the demand, further analyze which orders contribute the most, that is, which orders have the largest DL a -KD a The value of is the largest; S2. It should be noted that, among the analyzed orders, the orders whose demand exceeds the current inventory and whose delivery deadline is tight will have their priority increased; S3. If the inventory balance adjustment value PT shows insufficient inventory, and increasing the order priority is not enough to solve the problem, the purchase plan needs to be adjusted, specifically by increasing the purchase volume CG. At the same time, it is necessary to negotiate with the supplier to shorten the delivery time or find an alternative supplier; S4. The adjusted purchase plan will affect the expected inventory value YK in the future, which in turn will affect the subsequent order filling strategy, that is, the evaluation results of the subsequent order filling optimization evaluation unit In this embodiment, the algorithm unit makes a policy change based on the evaluation and analysis of the inventory balance adjustment value PT. The inventory balance adjustment value PT is a value that combines the purchase demand, the expected inventory level and the safety stock. Although the inventory balance adjustment value PT itself does not directly indicate the "amount" of inventory, the inventory status can be judged by comparing the inventory balance adjustment value PT with the expected range. The inventory balance adjustment evaluation unit provides real-time inventory status information to the order filling optimization evaluation unit by adjusting the inventory level. This feedback mechanism enables the order filling optimization evaluation unit to more accurately consider the actual inventory situation when optimizing order filling, thereby formulating a more reasonable order processing strategy to achieve the intelligent effect of the real-time feedback mechanism; As inventory levels change, the inventory balance adjustment evaluation unit can trigger the order filling optimization evaluation unit to re-evaluate and adjust the order filling strategy. This dynamic adjustment capability enables the supply chain to respond to market changes more quickly, improve the flexibility and responsiveness of the supply chain, and ensure the dynamic adjustment capability of artificial intelligence. The continuous monitoring and adjustment of inventory levels by the inventory balance adjustment evaluation unit can ensure that the supply chain can maintain a certain stability in the face of uncertain factors. This stability helps to reduce the risk of supply chain disruptions and improve the reliability and resilience of the supply chain. At the same time, the stable inventory level also provides a more reliable decision-making basis for the order filling optimization evaluation unit, further enhancing the overall performance of the supply chain. In summary, in the specific implementation process, the intelligent supply chain management method based on artificial intelligence has brought significant economic and social benefits to enterprises by improving the accuracy of demand forecasting, optimizing inventory management, strengthening supply chain collaboration, enhancing risk management capabilities and innovating supply chain management models.

[0031] For example 2, please refer to Figures 1 to 4 The additional inventory value AQ refers to the additional inventory maintained in addition to the normal inventory, which is used to deal with uncertain factors, including demand fluctuations and supply delays, which may cause out-of-stock risks, and the additional inventory value AQ does not directly participate in meeting daily demand.

[0032] In this embodiment, first, a demand forecasting model is determined, that is, a forecasting model suitable for the business needs of smart supply chain management is selected and developed, the model is trained using historical sales data, and its forecasting accuracy is verified, and various factors affecting demand fluctuations, including seasonal changes, market trends, promotional activities, and supply chain disruption risks, are analyzed, and statistical methods, namely standard deviation conversion and confidence intervals, are used to quantify these uncertainty factors, evaluate the uncertainty factors, and calculate the additional inventory value AQ based on the demand forecast and uncertainty evaluation results. This usually involves setting a service level and a confidence level to determine the additional inventory value AQ required to meet demand within a given time. The commonly used formula is to purchase demand management orders while maintaining inventory levels. It has been stated in the source: AQ=(L*BC)*JH. Subsequently, the safety stock is adjusted according to the actual situation of the supply chain. For example, if the supplier delivery time is unstable, it is considered necessary to increase the safety stock to cope with potential delays. The trade-off between inventory holding cost and out-of-stock cost is considered to optimize the total cost. The calculated safety stock is applied to the actual inventory management system. The inventory level and supply chain performance indicators are monitored in real time to evaluate the effectiveness of the safety stock and make adjustments as needed. Over time, new data is collected and the demand forecasting model and the calculation of the additional inventory value AQ are re-evaluated. The continuous learning ability of the machine learning model is used to continuously optimize the forecasting and inventory management strategies.

[0033] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent supply chain management method based on artificial intelligence, characterized in that: The smart supply chain management methods based on artificial intelligence are as follows: From the system module, use the data collection module to extract and collect historical sales data, inventory data, supplier delivery time, and production cycle; Use the data collection module to model historical sales data and generate forecast data for demand forecasts over a period of time in the future; Based on forecast data and using the data analysis and evaluation module, evaluate the impact of different orders on inventory, make urgent purchases based on the maximum out-of-stock quantity, and evaluate the necessity of inventory adjustments; The data analysis and evaluation module includes an order filling optimization evaluation unit, a purchase demand management unit for maintaining inventory levels, and an inventory balance adjustment evaluation unit; Based on the evaluation of inventory adjustments, and using automated decision-making and execution modules, intelligently adjust inventory levels, specifically trigger replenishment orders, optimize inventory allocation, and monitor inventory status and supply chain performance indicators in real time; The collected, predicted, evaluated and adjusted data are uploaded synchronously to the database server.

2. The intelligent supply chain management method based on artificial intelligence according to claim 1, characterized in that: The equipment used in the system module includes ERP system, sales system, and inventory management system; The equipment used in the data acquisition module includes ETL tools; The equipment used in the data analysis and evaluation module includes computing servers and GPUs; The equipment used by the automated decision-making and execution module includes an automated warehouse system.

3. The intelligent supply chain management method based on artificial intelligence according to claim 2 is characterized by: The calculation formula of the order filling optimization evaluation unit is as follows: ； in: DY is the order filling optimization value; d is the total number of orders; DI a is the ath order priority, reflecting DI a Reflect the priority of orders, including orders from VIP customers and orders that are about to exceed the delivery deadline; DL a is the quantity value of the ath order; KD a is the filling value of the current inventory, KD a Reflects the current inventory for the ath order quantity DL a The number of occupied fills; YJ a To predict the value of arrival, YJ a The expected number of future orders is the value of the ath order quantity DL a The number of occupants.

4. The intelligent supply chain management method based on artificial intelligence according to claim 3 is characterized by: The calculation formula of the purchase demand management unit under the maintenance of inventory level is as follows: ； AQ = (L*BC)*JH; in: CX is the purchase demand value; AQ is the additional inventory value; L is the critical value, which reflects the critical value of the standard normal distribution of the service level; BC is the standard deviation of demand forecast; JH is the delivery time, which refers to the total number of days from the time the order is placed to the actual delivery of the product to the customer; Max a (DL a -KD a ) indicates that among all orders, the demand exceeds the maximum value of the current inventory.

5. The method for intelligent supply chain management based on artificial intelligence according to claim 4, characterized in that: The calculation formula of the inventory balance adjustment valuation unit is as follows: ； YK=(DK+CG); in: PT is the inventory balance adjustment value; YK is the expected inventory value; DK is the current inventory value, CG is the purchase quantity; Used to account for the relative sizes of order quantities and inventory quantities in inventory balancing adjustments.

6. The method for intelligent supply chain management based on artificial intelligence according to claim 5, characterized in that: The evaluation of the inventory balance adjustment value PT output by the inventory balance adjustment evaluation unit is as follows: S1. Setting an expected range of the inventory balance adjustment value PT by the system module to determine the inventory status and its expected range; S2. If the inventory balance adjustment value PT exceeds the preset expected range, it is considered that the inventory is in excess; S3. If the inventory balance adjustment value PT is lower than the preset expected range, and the expected inventory value YK is insufficient to meet the demand in the future, the inventory is considered insufficient; S4. If the inventory balance adjustment value PT is within the expected range, the inventory is considered to be in a balanced state.

7. The method for intelligent supply chain management based on artificial intelligence according to claim 6, characterized in that: The automated decision and execution module makes the following adjustments based on the analysis of the inventory balance adjustment evaluation unit: S1. Observe whether the expected inventory value YK meets the demand in the future. If the expected inventory value YK does not meet the demand, analyze which orders contribute the most, that is, which orders have the highest DL. a -KD a The value of is the largest; S2, among which, For those analyzed orders whose demand exceeds the current inventory and whose delivery deadline is tight, their priority will be increased; S3. If the inventory balance adjustment value PT shows insufficient inventory and increasing the order priority is not enough to solve the problem, adjust the procurement plan, specifically increase the procurement volume CG, and negotiate with suppliers to shorten the delivery time and find alternative suppliers; S4. The adjusted purchasing plan will affect the future expected inventory value YK, as well as the subsequent order filling strategy, that is, the evaluation result of the subsequent order filling optimization evaluation unit.

8. The method for intelligent supply chain management based on artificial intelligence according to claim 4, characterized in that: The additional inventory value AQ is specifically the inventory amount maintained in addition to the normal inventory, which is used to deal with uncertain factors, including out-of-stock risks caused by demand fluctuations and supply delays, and the additional inventory value AQ does not directly participate in meeting daily demand.

Images