An artificial intelligence-based supply chain management method

Abstract

This invention discloses an artificial intelligence-based supply chain management method, relating to the field of AI-based supply chain management. The method includes: acquiring supply chain operation data; identifying supply chain disturbances, determining candidate supply chain adjustment behaviors, extracting secondary disturbance information caused by these adjustments, and controlling these adjustments based on the disturbance and secondary disturbance information. The result is not merely an immediate response to the current anomaly, but a supply chain management outcome that balances the ability to mitigate the current disturbance with subsequent operational stability. This allows for mitigation of current supply chain anomalies while suppressing subsequent anomalies caused by the adjustment behaviors themselves, reducing issues such as fluctuations in procurement and material expediting schedules, repeated changes in work order priorities, restructuring of resource allocation relationships, and instability in order commitment expectations. Ultimately, this reduces the probability of repeated rescheduling, frequent rescheduling, and multiple rounds of oscillations during supply chain operation, improving the overall stability and controllability of the supply chain.

Description

Technical Field

[0001] This invention relates to the field of supply chain management based on artificial intelligence, and more particularly to a supply chain management method based on artificial intelligence. Background Technology

[0002] As the level of digitalization in supply chain management continues to improve, more and more enterprises are adopting AI-based supply chain management methods to collect, analyze, and process multi-source data in real time, including procurement arrivals, inventory changes, production takt time, logistics status, and order demand. Through this approach, the system can dynamically assess material availability, resource allocation relationships, and order delivery capabilities, and quickly trigger recalculations when abnormal fluctuations occur to update production plans, procurement strategies, and order commitments. This improves the speed of response to anomalies, enhances resource utilization efficiency, and strengthens supply chain collaboration.

[0003] Existing technologies typically employ a "real-time data collection - dynamic analysis - trigger recalculation - update results" approach when handling supply chain disruptions. This means that when the system detects procurement delays, logistical anomalies, production line fluctuations, quality release lags, or changes in customer demand, it recalculates material supply status, production priorities, and order commitment dates based on the latest inputs to quickly generate new scheduling results. This approach can rapidly provide new execution plans when dealing with single anomalies or localized changes, and is therefore widely used in current supply chain management systems.

[0004] However, existing technologies primarily focus on recalculating optimal results based on the latest disturbances, paying less attention to whether the adjustment itself will have a repercussions on the subsequent system state. In other words, while existing systems can quickly generate new procurement, planning, or commitment schemes in response to current anomalies, they typically do not incorporate the secondary impacts of this adjustment on subsequent procurement expediting schedules, work order priorities, resource allocation, and customer commitment expectations into unified control. As a result, each seemingly reasonable local adjustment by the system may further alter the operational state of subsequent stages and generate new anomalies in the next period.

[0005] In actual operation, the above-mentioned problems will bring a series of chain reactions. For example, when the system frequently adjusts the order commitment results, customer expectations will fluctuate continuously; when the system frequently changes the work order sorting, the number of resource switching on the manufacturing side will increase significantly; when the system frequently updates the procurement priority, the procurement side's material expediting rhythm and delivery organization methods will also fluctuate accordingly. These changes caused by the adjustment behavior itself are not directly caused by the original disturbance, but rather by secondary disturbances generated by the system during repeated recalculations. These secondary disturbances will re-enter the system, becoming the input for the next round of analysis and recalculation, thus causing the system to gradually fall into a state where "the more local optimization, the more volatile the whole; the more frequent the recalculation, the more unstable the operation."

[0006] The aforementioned problems are particularly pronounced in supply chain scenarios involving multiple suppliers, materials, work orders, and orders operating in parallel. Due to the strong coupling between different levels, a local adjustment often propagates to multiple stages, including procurement, planning, manufacturing, and delivery, thereby amplifying the overall instability of the system. While existing technologies can continuously improve computing speed and response frequency, the lack of identification and constraint on "secondary disturbances caused by adjustment behavior" often only enables rapid responses to current anomalies, making it difficult to ensure the overall stability of the supply chain system during continuous operation. Summary of the Invention

[0007] The purpose of this invention is to provide an artificial intelligence-based supply chain management method to solve the above-mentioned technical problems.

[0008] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: a supply chain management method based on artificial intelligence, comprising the following steps: Acquire supply chain operation data; and identify supply chain disturbances based on the supply chain operation data; At least one supply chain adjustment behavior is preset for the aforementioned supply chain disturbance; Analyze the changes in information about the subsequent operation of the supply chain caused by the supply chain adjustment behavior, and determine the secondary disturbance information corresponding to the supply chain adjustment behavior; Based on the information about supply chain disturbances and secondary disturbances, the supply chain adjustment behavior is controlled to suppress the formation of new abnormal inputs in subsequent periods and output the corresponding supply chain management results.

[0009] As a preferred embodiment of the artificial intelligence-based supply chain management method described in this invention, the supply chain operation data includes procurement arrival data, inventory change data, production cycle data, logistics status data, and order demand data; The step of identifying supply chain disturbances based on the supply chain operation data includes: Construct a preset operating state corresponding to the supply chain operation data. By comparing the differences between the supply chain operation data and the corresponding preset operation states, offset information is determined. Based on the difference information in the offset information that meets preset conditions, supply chain disturbances are identified; The preset operating states include at least one of the following: preset arrival state corresponding to procurement arrival data, preset inventory state corresponding to inventory change data, preset production state corresponding to production cycle data, preset logistics state corresponding to logistics status data, and preset demand state corresponding to order demand data.

[0010] As a preferred embodiment of the artificial intelligence-based supply chain management method described in this invention, the supply chain adjustment behavior specifically includes: The supply chain disturbances are categorized by type. Based on the type of disturbance in the supply chain, at least one candidate supply chain adjustment behavior is determined from a preset set of supply chain adjustment behaviors; Before implementing the preset supply chain adjustment behaviors, it is also necessary to determine the affected links corresponding to each type of disturbance; extract the executable adjustment actions that act on the affected links; combine historical disturbance handling records or business rules to screen adjustment actions that have a mitigating effect on the corresponding disturbances; and construct the preset set of supply chain adjustment behaviors based on the screening results. The types of supply chain disturbances include at least one of the following: procurement fulfillment disturbances, inventory supply disturbances, production execution disturbances, logistics delivery disturbances, and demand fluctuation disturbances.

[0011] As a preferred embodiment of the artificial intelligence-based supply chain management method described in this invention, the determination of the secondary disturbance information includes: Obtain baseline status information corresponding to the subsequent operating status of the supply chain when the aforementioned supply chain adjustment behavior has not been executed; Obtain the change status information corresponding to the subsequent operating status of the supply chain after the execution of the supply chain adjustment behavior; Compare the differences between the baseline state information and the changed state information; Based on the differences, determine the information changes in the subsequent operational status of the supply chain caused by the supply chain adjustment behavior; Based on the changes in the information, determine the secondary disturbance information corresponding to the supply chain adjustment behavior; The subsequent operational status of the supply chain includes the pace of subsequent procurement and material expediting, work order priority, resource utilization, and expected order commitments.

[0012] As a preferred embodiment of the artificial intelligence-based supply chain management method described in this invention, controlling the supply chain adjustment behavior refers to determining the disturbance suppression value and secondary disturbance value corresponding to each candidate supply chain adjustment behavior based on the supply chain disturbance and secondary disturbance information, and determining the behavior control value corresponding to each candidate supply chain adjustment behavior based on the disturbance suppression value and the secondary disturbance value. The behavior control value is determined according to the following formula: , in, This represents the behavior control value corresponding to the i-th candidate supply chain adjustment behavior. This represents the disturbance suppression value of the i-th candidate supply chain adjustment behavior on the current supply chain disturbance. Let represent the secondary disturbance value corresponding to the i-th candidate supply chain adjustment behavior. This represents the secondary disturbance suppression coefficient; and the behavioral control value is used to implement control over the adjustment behavior of each candidate supply chain in order to determine the target supply chain adjustment behavior.

[0013] As a preferred embodiment of the artificial intelligence-based supply chain management method described in this invention, the disturbance suppression value is determined based on the degree of change in supply chain disturbance before and after the execution of the corresponding candidate supply chain adjustment behavior, and is used to characterize the ability of the candidate supply chain adjustment behavior to weaken the current supply chain disturbance. The secondary disturbance value is determined based on the degree of change in the subsequent operating status of the supply chain caused by the corresponding candidate supply chain adjustment behavior, and is used to characterize the degree of impact of the candidate supply chain adjustment behavior on the formation of new abnormal inputs in subsequent periods after its execution. The degree of change in the subsequent operation status of the supply chain is determined based on at least one of the following: changes in the pace of procurement and material expediting, changes in work order priority, changes in resource utilization relationships, and changes in expected order commitments.

[0014] As a preferred embodiment of the artificial intelligence-based supply chain management method described in this invention, the secondary disturbance information; for each of the supply chain adjustment behaviors, analyze the impact of the supply chain adjustment behavior on the subsequent operating status of the supply chain; The execution control of each candidate supply chain adjustment behavior based on the behavior control value means: when the behavior control value is greater than a preset execution threshold, the corresponding candidate supply chain adjustment behavior is executed; when the behavior control value is less than or equal to the preset execution threshold and greater than a preset restriction threshold, the execution of the corresponding candidate supply chain adjustment behavior is restricted; when the behavior control value is less than or equal to the preset restriction threshold, the execution of the corresponding candidate supply chain adjustment behavior is prohibited.

[0015] As a preferred embodiment of the AI-based supply chain management method described in this invention, the restriction execution includes at least one of the following: delayed execution, partial execution, reduced adjustment range, and replacement with alternative adjustment behaviors that have a smaller impact on the subsequent operation of the supply chain. After determining the target supply chain adjustment behavior, the method also includes outputting the corresponding supply chain management results based on the target supply chain adjustment behavior.

[0016] The beneficial effects of this invention are: This invention identifies supply chain disturbances, determines candidate supply chain adjustment behaviors, extracts secondary disturbance information caused by these adjustments, and controls these adjustments based on the information. The result is not merely an immediate response to the current anomaly, but a supply chain management outcome that balances the ability to mitigate the current disturbance with subsequent operational stability. This allows for the mitigation of current supply chain anomalies while suppressing subsequent anomalies caused by the adjustment behaviors themselves. It reduces issues such as fluctuations in procurement and material expediting schedules, repeated changes in work order priorities, restructuring of resource allocation relationships, and instability in order commitments and expectations. Ultimately, this lowers the probability of repeated rescheduling, frequent rescheduling, and multiple rounds of oscillations during supply chain operations, improving the overall stability and controllability of the supply chain. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation on the scope of this application.

[0018] Figure 1 This is a flowchart illustrating the AI-based supply chain management method in the embodiments. Detailed Implementation

[0019] To make the technical means, creative features, and achieved objectives and effects of this invention easier to understand, the invention is further described below with reference to specific embodiments. However, the following embodiments are merely preferred embodiments of this invention and not all of them. Other embodiments obtained by those skilled in the art based on the embodiments described herein without creative effort are all within the protection scope of this invention. Unless otherwise specified, the experimental methods in the following embodiments are conventional methods, and the materials and reagents used in the following embodiments are commercially available unless otherwise specified.

[0020] like Figure 1As shown, the "supply chain management method based on artificial intelligence" refers to a management method that uses data processing, status recognition and behavior control mechanisms to identify, respond to and constrain abnormal situations that occur during the operation of the supply chain. Its purpose is not only to make immediate adjustments to the current abnormality, but also to suppress the propagation of subsequent abnormalities caused by the adjustment behavior itself while adjusting the current abnormality.

[0021] An artificial intelligence-based supply chain management method includes the following steps: Acquire supply chain operation data; and identify supply chain disturbances based on the supply chain operation data; At least one supply chain adjustment behavior is preset for the aforementioned supply chain disturbance; Analyze the changes in information about the subsequent operation of the supply chain caused by the supply chain adjustment behavior, and determine the secondary disturbance information corresponding to the supply chain adjustment behavior; Based on the information regarding supply chain disturbances and secondary disturbances, the supply chain adjustment behavior is controlled to prevent it from generating new abnormal inputs in subsequent periods, and corresponding supply chain management results are output. This implementation uses a multi-stage collaborative supply chain scenario in a discrete manufacturing enterprise as an example. The enterprise's supply chain typically includes multiple stages: procurement, inventory, production, logistics, and order fulfillment. In existing technologies, when an anomaly occurs in a certain stage, the system often directly regenerates procurement plans, production schedules, or order commitment plans based on the current anomaly. While this can alleviate the immediate problem, the lack of constraints on the subsequent impact of adjustments can easily lead to new anomalies recurring in later periods. For example, while a work order rescheduling can temporarily bypass work orders with material shortages, it may cause multiple subsequent work orders to change their priorities simultaneously; while an inventory reallocation can ensure the current critical orders, it may disrupt the inventory holding relationships of other orders; while an order rescheduling can relieve current production pressure, it may cause continuous fluctuations in order commitment expectations. Therefore, this implementation does not directly use "adjustment behavior" as the final output, but further determines whether the adjustment behavior will generate new anomaly inputs after execution, thereby obtaining a supply chain management result that balances current mitigation capabilities and subsequent stability.

[0022] In this embodiment, supply chain operation data is acquired; and supply chain disturbances are identified based on the supply chain operation data. First, supply chain operation data is acquired. Here, "supply chain operation data" refers to a set of data that characterizes the current state of each operational link in the supply chain, providing input for subsequent disturbance identification and behavioral control. The supply chain operation data includes at least procurement arrival data, inventory change data, production cycle time data, logistics status data, and order demand data. Among them, "procurement arrival data" refers to data reflecting whether materials arrive as planned, including planned arrival time, actual arrival time, quantity, and batch; "inventory change data" refers to data reflecting the inventory status of materials, including inventory balance, available inventory, frozen inventory, and inventory occupancy; "production cycle data" refers to data reflecting production execution speed and status, including output per unit time, work order execution progress, equipment operating status, and process completion status; "logistics status data" refers to data reflecting the transportation and delivery process, including dispatch time, node arrival time, transit time, and receipt time; and "order demand data" refers to data reflecting customer demand and fulfillment requirements, including order quantity, order priority, order release rhythm, and promised delivery date.

[0023] After acquiring the supply chain operation data, supply chain disturbances are identified based on the supply chain operation data. Here, "supply chain disturbance" refers to an abnormal event in which the operating state of the supply chain deviates from the expected state, and this deviation, once it reaches a preset condition, can affect the subsequent operation of the supply chain.

[0024] The step of identifying supply chain disturbances based on the supply chain operation data includes: constructing a preset operation state corresponding to the supply chain operation data, comparing the differences between the supply chain operation data and the corresponding preset operation state, determining offset information, and identifying supply chain disturbances based on the difference information in the offset information that meets preset conditions. To ensure clear identification criteria for supply chain disturbances, this embodiment first constructs preset operating states corresponding to the supply chain operation data. These "preset operating states" refer to pre-defined expected baseline states for operational aspects such as procurement, inventory, production, logistics, and order demand, serving as a comparison reference to determine whether current operations have deviated. The preset operating states can be constructed based on historical operation records, planning and scheduling results, business rule thresholds, or target execution parameters.

[0025] The preset operating state includes at least one of the following: preset arrival state corresponding to procurement arrival data, preset inventory state corresponding to inventory change data, preset production state corresponding to production cycle data, preset logistics state corresponding to logistics status data, and preset demand state corresponding to order demand data. Specifically, the preset operating status corresponding to procurement and delivery data can be preset delivery time, preset delivery quantity, or preset delivery batch; the preset operating status corresponding to inventory change data can be preset safety stock minimum, preset available inventory range, or preset inventory occupancy relationship; the preset operating status corresponding to production cycle time data can be preset production cycle time, preset completion time, or preset work order execution order; the preset operating status corresponding to logistics status data can be preset transportation time, preset node arrival time, or preset delivery order; and the preset operating status corresponding to order demand data can be preset demand fluctuation range, preset order release rhythm, or preset promised delivery window.

[0026] Subsequently, the supply chain operation data is compared with the corresponding preset operation state to determine the deviation information. Here, "deviation information" refers to the difference between the actual operation state and the preset operation state, used to characterize the degree to which the current operation deviates from expectations. If the deviation information meets preset conditions, it is identified as a supply chain disturbance. Here, "preset conditions" refers to judgment conditions used to distinguish between normal fluctuations and effective anomalies. Specifically, it can be that the deviation magnitude exceeds a preset threshold, the deviation duration reaches a preset duration, or the deviation has already affected subsequent operation links. For example, when the actual arrival time of a purchase order is delayed by more than 2 hours compared to the preset arrival time, it can be identified as a purchase fulfillment disturbance; when the available inventory of a key material is more than 10% lower than the safety stock limit, it can be identified as an inventory supply disturbance; when the actual production cycle of a production line is more than 15% lower than the preset production cycle, it can be identified as a production execution disturbance; when the transit time of a logistics task exceeds 20% of the preset transportation time, it can be identified as a logistics delivery disturbance; when the demand of a customer order deviates from the preset demand fluctuation range by more than 15%, it can be identified as a demand fluctuation disturbance.

[0027] The aforementioned supply chain adjustment behaviors specifically include: classifying the supply chain disturbances by type; determining at least one candidate supply chain adjustment behavior from a preset set of supply chain adjustment behaviors based on the disturbance type; determining the affected links corresponding to each disturbance type before the preset supply chain adjustment behaviors; extracting executable adjustment actions that act on the affected links; screening adjustment actions that have a mitigating effect on the corresponding disturbances by combining historical disturbance handling records or business rules; and constructing the preset set of supply chain adjustment behaviors based on the screening results; the disturbance types of the supply chain disturbances include at least one of procurement fulfillment disturbances, inventory supply disturbances, production execution disturbances, logistics delivery disturbances, and demand fluctuation disturbances. After identifying supply chain disturbances, at least one supply chain adjustment action is determined based on the disturbances. Here, "supply chain adjustment action" refers to a response action taken in response to a supply chain disturbance to mitigate its adverse effects on operations. It should be noted that the supply chain adjustment action in this embodiment is not a final action to be executed directly, but rather an action that enters a candidate set and awaits subsequent control. To facilitate the determination of candidate actions, the supply chain disturbances are first categorized. Here, "categorization" refers to classifying the identified disturbances according to their location and impact characteristics, so as to match corresponding response actions from different sets of adjustment actions. The disturbance types of the supply chain disturbances include at least one of the following: procurement fulfillment disturbances, inventory supply disturbances, production execution disturbances, logistics delivery disturbances, and demand fluctuation disturbances. Among them, "procurement fulfillment disturbance" refers to disturbances caused by delayed delivery, insufficient delivery, or delayed release of incoming materials; "inventory supply disturbance" refers to disturbances caused by insufficient inventory, unbalanced inventory distribution, or conflicting inventory holdings; "production execution disturbance" refers to disturbances caused by decreased cycle time, equipment stagnation, or delayed work order execution; "logistics delivery disturbance" refers to disturbances caused by transportation delays, node blockages, or abnormal delivery sequence; and "demand fluctuation disturbance" refers to disturbances caused by order insertion, order cancellation, or sudden increase in demand.

[0028] After classifying the disturbance types, at least one candidate supply chain adjustment behavior is determined from a pre-set set of supply chain adjustment behaviors. The "pre-set set of supply chain adjustment behaviors" refers to a set of response actions maintained by the system in advance, capable of corresponding to different disturbance types; "candidate supply chain adjustment behaviors" refer to one or more actions selected from this set that may be used to mitigate the current disturbance. For procurement fulfillment disturbances, candidate supply chain adjustment behaviors may include expediting materials, adjusting procurement priorities, or splitting arrival batches; for inventory supply disturbances, candidate supply chain adjustment behaviors may include inventory reallocation, releasing safety stock, or calling up alternative inventory; for production execution disturbances, candidate supply chain adjustment behaviors may include work order rearrangement, priority switching, resource switching, or partial skipping; for logistics delivery disturbances, candidate supply chain adjustment behaviors may include adjusting shipping order, adjusting transit nodes, or shipping in batches; for demand fluctuation disturbances, candidate supply chain adjustment behaviors may include order splitting, adjusting fulfillment order, or adjusting order commitments. It is important to emphasize that the system does not directly execute the candidate supply chain adjustment behaviors at this stage, but rather further analyzes the impact of these behaviors on subsequent periods.

[0029] The analysis examines the impact of supply chain adjustment actions on subsequent supply chain operational status, determining the secondary disturbance information corresponding to the adjustment actions. Determining the secondary disturbance information includes: acquiring baseline status information corresponding to the subsequent supply chain operational status before the adjustment actions were implemented; acquiring changed status information corresponding to the subsequent supply chain operational status after the adjustment actions were implemented; comparing the differences between the baseline status information and the changed status information; determining the information changes in the subsequent supply chain operational status based on the differences; and determining the secondary disturbance information corresponding to the adjustment actions based on the information changes. The subsequent supply chain operational status specifically includes subsequent procurement and material expediting schedules, work order priorities, resource utilization relationships, and order commitment expectations. Next, the impact of the supply chain adjustment behavior on the subsequent operation of the supply chain is analyzed, and the secondary disturbance information corresponding to the supply chain adjustment behavior is determined. The "subsequent operation of the supply chain" refers to the key operational status items that may be affected in subsequent periods after a candidate supply chain adjustment behavior is executed. These status items include at least one of the following: subsequent procurement expediting rhythm, work order priority, resource utilization relationship, and order commitment expectation. Specifically, "procurement expediting rhythm" refers to the order and frequency of expedited procurement items in subsequent periods; "work order priority" refers to the ranking relationship of each work order in subsequent production execution; "resource utilization relationship" refers to the allocation and utilization relationship of resources such as equipment, tooling, inventory, and workstations in subsequent periods; and "order commitment expectation" refers to the customer's expected state regarding order delivery date or fulfillment results. The "secondary disturbance information" refers to the information on changes in the subsequent operation status caused by the candidate supply chain adjustment behavior, used to characterize whether the adjustment behavior itself will generate new abnormal inputs in subsequent periods. It should be specifically noted that the secondary disturbance information is not a continuation of the original supply chain disturbance, but rather a secondary impact information caused by the supply chain adjustment behavior itself. This is an important basis for the present invention to differ from existing direct recalculation schemes.

[0030] Specifically, first, baseline state information corresponding to the subsequent operating state of the supply chain before the candidate supply chain adjustment behavior is executed is obtained. Here, "baseline state information" refers to the state information that the subsequent operating state of the supply chain should have remained unchanged without executing the current candidate supply chain adjustment behavior. Then, changed state information corresponding to the subsequent operating state of the supply chain after executing the candidate supply chain adjustment behavior is obtained. Here, "changed state information" refers to the state information after the subsequent operating state item changes following the execution of the candidate supply chain adjustment behavior. Next, the baseline state information and the changed state information are compared to determine the information change in the subsequent operating state of the supply chain. Here, "information change" refers to the difference between the baseline state and the changed state, used to reflect the impact of the current candidate supply chain adjustment behavior on the subsequent operating state. For example, if the subsequent work order order order sequence is ABC before work order reordering is performed, but becomes BAC after work order reordering, it can be determined that the work order priority information has changed; if material M1 is occupied by order O1 before inventory reallocation is performed, but becomes occupied by order O2 after inventory reallocation, it can be determined that the resource occupancy relationship information has changed; if the commitment date of order O3 is March 20th before order commitment adjustment is performed, but becomes March 22nd after commitment adjustment, it can be determined that the order commitment expectation information has changed; if the expediting adjustment is performed before material expediting is performed, and the target material expediting order is expedited ahead of schedule after expediting is performed, it can be determined that the material expediting rhythm information has changed. Based on the above information changes, the secondary disturbance information corresponding to the candidate supply chain adjustment behavior is determined.

[0031] Based on the information on supply chain disturbances and secondary disturbances, the supply chain adjustment behavior is controlled to suppress the formation of new abnormal inputs in subsequent periods and output the corresponding supply chain management results. After obtaining information on supply chain disturbances and secondary disturbances, the supply chain adjustment behavior is controlled. "Controlling the supply chain adjustment behavior" here does not mean simply executing or not executing it, but rather comprehensively determining the execution priority and method of each candidate supply chain adjustment behavior based on the need to mitigate the current supply chain disturbance and the subsequent risks reflected in the secondary disturbance information. To achieve this, this embodiment determines the disturbance suppression value and secondary disturbance value corresponding to each candidate supply chain adjustment behavior. The "disturbance suppression value" refers to the ability of a candidate supply chain adjustment behavior to weaken the current supply chain disturbance, which can be determined based on the degree of change in supply chain disturbance before and after the candidate behavior is executed; for example, if a delivery delay disturbance would originally have caused a 4-hour production stoppage, but after inventory reallocation it only affects 1 hour of production, then the inventory reallocation action corresponds to a higher disturbance suppression value. The "secondary disturbance value" mentioned here refers to the degree of impact of a candidate supply chain adjustment behavior on new abnormal inputs in subsequent periods after its execution. It can be determined based on the degree of change in the subsequent operating status of the supply chain caused by the candidate behavior. For example, if a work order rescheduling behavior can alleviate the current material shortage risk, but causes changes in the priority of multiple work orders, changes in multiple resource occupancy relationships, and adjustments to order commitments, then the work order rescheduling behavior corresponds to a higher secondary disturbance value.

[0032] Controlling the supply chain adjustment behavior refers to determining the disturbance suppression value and secondary disturbance value corresponding to each candidate supply chain adjustment behavior based on the supply chain disturbance and secondary disturbance information, and determining the behavior control value corresponding to each candidate supply chain adjustment behavior based on the disturbance suppression value and the secondary disturbance value. The behavior control value is determined according to the following formula: ,in, This represents the behavior control value corresponding to the i-th candidate supply chain adjustment behavior. This represents the disturbance suppression value of the i-th candidate supply chain adjustment behavior on the current supply chain disturbance. Let represent the secondary disturbance value corresponding to the i-th candidate supply chain adjustment behavior. This represents the secondary disturbance suppression coefficient; and the execution control of each candidate supply chain adjustment behavior is performed based on the behavior control value to determine the target supply chain adjustment behavior; the disturbance suppression value is determined based on the degree of supply chain disturbance change before and after the execution of the corresponding candidate supply chain adjustment behavior, and is used to characterize the ability of the candidate supply chain adjustment behavior to weaken the current supply chain disturbance; the secondary disturbance value is determined based on the degree of change in the subsequent operating state of the supply chain caused by the corresponding candidate supply chain adjustment behavior, and is used to characterize the degree of impact of the candidate supply chain adjustment behavior on the formation of new abnormal inputs in subsequent periods after execution; the degree of change in the subsequent operating state of the supply chain is determined based on at least one of the following: changes in procurement expediting rhythm, changes in work order priority, changes in resource utilization relationship, and changes in order commitment expectations; In this embodiment, based on the disturbance suppression value and the secondary disturbance value, a behavior control value corresponding to each candidate supply chain adjustment behavior is further determined. Here, "behavior control value" refers to a comprehensive control indicator used to measure whether a candidate supply chain adjustment behavior is worth implementing. It reflects not only the ability of the behavior to mitigate current supply chain disturbances but also the degree of impact of the behavior on subsequent operational stability. The behavior control value is determined according to the following formula: , in, This represents the behavior control value corresponding to the i-th candidate supply chain adjustment behavior. This represents the disturbance suppression value of the i-th candidate supply chain adjustment behavior on the current supply chain disturbance. Let represent the secondary disturbance value corresponding to the i-th candidate supply chain adjustment behavior. This represents the secondary disturbance suppression coefficient; and based on the behavioral control value, execution control is applied to each candidate supply chain adjustment behavior to determine the target supply chain adjustment behavior. The "secondary disturbance suppression coefficient" mentioned here refers to a control parameter used to adjust the system's sensitivity to secondary disturbances. When enterprises prioritize overall operational stability, this coefficient can be appropriately increased to enhance the suppression of high-secondary disturbance behaviors; when enterprises prioritize the rapid mitigation of current disturbances, this coefficient can be appropriately decreased. In this formula, the numerator represents the ability of the current candidate supply chain adjustment behavior to mitigate the current anomaly, and the denominator represents the amplified cost that the candidate behavior may cause in subsequent periods. Therefore, the behavioral control value is not a simple priority score, but rather a "comprehensive control result between current effectiveness and subsequent stability."

[0033] The secondary disturbance information; for each of the aforementioned supply chain adjustment behaviors, analyze the impact of the supply chain adjustment behavior on the subsequent operating status of the supply chain; The execution control of each candidate supply chain adjustment behavior based on the behavior control value means: when the behavior control value is greater than a preset execution threshold, the corresponding candidate supply chain adjustment behavior is executed; when the behavior control value is less than or equal to the preset execution threshold but greater than a preset restriction threshold, the execution of the corresponding candidate supply chain adjustment behavior is restricted; when the behavior control value is less than or equal to the preset restriction threshold, the execution of the corresponding candidate supply chain adjustment behavior is prohibited. The restricted implementation includes at least one of the following: delayed implementation, partial implementation, reduced adjustment magnitude, and replacement with alternative adjustment behaviors that have a smaller impact on the subsequent operation of the supply chain; After determining the target supply chain adjustment behavior, the method also includes outputting corresponding supply chain management results based on the target supply chain adjustment behavior. Based on the aforementioned behavior control value, execution control is applied to each candidate supply chain adjustment behavior. Here, "execution control" refers to different control methods—execution, restricted execution, or prohibition—based on the behavior control value. When the behavior control value is greater than a preset execution threshold, the corresponding candidate supply chain adjustment behavior is executed; when the behavior control value is less than or equal to the preset execution threshold but greater than a preset restriction threshold, the corresponding candidate supply chain adjustment behavior is restricted; when the behavior control value is less than or equal to the preset restriction threshold, the corresponding candidate supply chain adjustment behavior is prohibited. Here, "restricted execution" means not completely rejecting the candidate behavior, but implementing it to a limited extent to reduce subsequent volatility risks. This includes at least one of the following: delayed execution, partial execution, reduced adjustment magnitude, and replacement with an alternative adjustment behavior that has a smaller impact on the subsequent operation of the supply chain. Among them, "delayed execution" means postponing the implementation of the candidate behavior to a more suitable time window; "partial execution" means implementing the behavior only in a local supply chain link, without spreading it to the whole; "reducing the adjustment range" means narrowing the scope or intensity of the candidate behavior; "replacing with an alternative adjustment behavior that has less impact on the subsequent operation of the supply chain" means replacing the current candidate behavior with another candidate behavior with similar disturbance suppression capabilities but lower secondary disturbance values.

[0034] After completing the above execution controls, the target supply chain adjustment behavior is determined. Here, "target supply chain adjustment behavior" refers to the final response action that has been screened and permitted by behavior control values. Then, based on the target supply chain adjustment behavior, the corresponding supply chain management result is output. Here, "supply chain management result" refers to the controlled management output result, including at least one of the following: procurement execution result, inventory allocation result, production scheduling result, logistics scheduling result, and order commitment result. In other words, this implementation method does not output the most sensitive and aggressive result to the current anomaly, but rather a result that can mitigate the current anomaly while minimizing anomaly input in subsequent periods.

[0035] The following example illustrates the operation of this implementation method. A batch of critical material M1 for a discrete manufacturing company was originally scheduled to arrive in 1000 units at 10:00 AM on March 18th, but had not arrived by 2:00 PM on March 18th. Simultaneously, the available inventory of this material was only 150 units, while the safety stock limit was 300 units. At this point, the system acquires procurement arrival data and inventory change data, compares them with the corresponding preset operating states, and identifies procurement fulfillment disturbances and inventory supply disturbances. Subsequently, based on these supply chain disturbances, the system determines the following candidate supply chain adjustment behaviors: Candidate action A is to expedite material delivery; candidate action B is to reallocate inventory to prioritize critical work orders; candidate action C is to reorder work orders to execute work orders that do not depend on material M1 first; and candidate action D is to directly postpone the order commitment date. Next, the system acquires the baseline state information when each candidate action was not executed, as well as the changed state information after executing each candidate action, and determines the corresponding secondary disturbance information accordingly. For example, candidate action A might cause changes in the procurement expediting rhythm but not significantly affect work order priority; candidate action B might change inventory holding relationships but have little impact on subsequent scheduling; candidate action C might cause changes in the priority and resource holding relationships of multiple work orders; and candidate action D might directly lead to changes in order commitment expectations. Further, the system determines disturbance suppression and secondary disturbance values ​​for each candidate action and substitutes them into the aforementioned behavior control value formula. For example, if candidate action B has a high disturbance suppression value and a low secondary disturbance value, its behavior control value is the highest, and the system prioritizes inventory reallocation; if candidate action C has some disturbance suppression capability but a high secondary disturbance value, the system only allows its partial execution; if candidate action D has limited mitigation of the current disturbance and a significantly high secondary disturbance value, the system prohibits its execution. Ultimately, the system outputs the following supply chain management results: priority is given to inventory reallocation to ensure the production of critical work orders; work order rescheduling is only partially implemented; moderate expediting is triggered on the procurement side but does not comprehensively disrupt the procurement rhythm, while maintaining order commitments unchanged. In this way, the system not only alleviates the current material shortage problem, but also avoids repeated rescheduling of subsequent work orders and frequent rescheduling of orders caused by excessive adjustments.

[0036] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0037] Furthermore, those skilled in the art will understand that although some embodiments herein include certain features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of this application and form different embodiments. For example, all the embodiments above can be used in any combination. The information disclosed in this background section is intended only to enhance the understanding of the general background of this application and should not be construed as an admission or in any way implying that such information constitutes prior art known to those skilled in the art. The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A supply chain management method based on artificial intelligence, characterized in that, Includes the following steps: Acquire supply chain operation data; and identify supply chain disturbances based on the supply chain operation data; At least one supply chain adjustment behavior is preset for the aforementioned supply chain disturbance; Analyze the changes in information about the subsequent operation of the supply chain caused by the supply chain adjustment behavior, and determine the secondary disturbance information corresponding to the supply chain adjustment behavior; Based on the information on supply chain disturbances and secondary disturbances, the supply chain adjustment behavior is controlled to suppress the formation of new abnormal inputs in subsequent periods and output the corresponding supply chain management results.

2. The supply chain management method based on artificial intelligence according to claim 1, characterized in that: The supply chain operation data includes procurement and delivery data, inventory change data, production cycle data, logistics status data, and order demand data; The step of identifying supply chain disturbances based on the supply chain operation data includes: Construct a preset operating state corresponding to the supply chain operation data. By comparing the differences between the supply chain operation data and the corresponding preset operation states, offset information is determined. Based on the difference information in the offset information that meets preset conditions, supply chain disturbances are identified; The preset operating states include at least one of the following: preset arrival state corresponding to procurement arrival data, preset inventory state corresponding to inventory change data, preset production state corresponding to production cycle data, preset logistics state corresponding to logistics status data, and preset demand state corresponding to order demand data.

3. The supply chain management method based on artificial intelligence according to claim 2, characterized in that: The aforementioned supply chain adjustment behaviors specifically include: The supply chain disturbances are categorized by type. Based on the type of disturbance in the supply chain, at least one candidate supply chain adjustment behavior is determined from a preset set of supply chain adjustment behaviors; Before the preset supply chain adjustment behaviors, it is also necessary to determine the affected links corresponding to each type of disturbance; extract the executable adjustment actions that act on the affected links; combine historical disturbance handling records or business rules to screen adjustment actions that have a mitigating effect on the corresponding disturbances; and construct the preset set of supply chain adjustment behaviors based on the screening results. The types of supply chain disturbances include at least one of the following: procurement fulfillment disturbances, inventory supply disturbances, production execution disturbances, logistics delivery disturbances, and demand fluctuation disturbances.

4. The supply chain management method based on artificial intelligence according to claim 3, characterized in that: The determination of the secondary disturbance information includes: Obtain baseline status information corresponding to the subsequent operating status of the supply chain when the aforementioned supply chain adjustment behavior has not been executed; Obtain the change status information corresponding to the subsequent operating status of the supply chain after the execution of the supply chain adjustment behavior; Compare the differences between the baseline state information and the changed state information; Based on the differences, determine the information changes in the subsequent operational status of the supply chain caused by the supply chain adjustment behavior; Based on the changes in the information, determine the secondary disturbance information corresponding to the supply chain adjustment behavior; The subsequent operational status of the supply chain includes the pace of subsequent procurement and material expediting, work order priority, resource utilization, and expected order commitments.

5. The supply chain management method based on artificial intelligence according to claim 4, characterized in that: Controlling the supply chain adjustment behavior refers to determining the disturbance suppression value and secondary disturbance value corresponding to each candidate supply chain adjustment behavior based on the supply chain disturbance and secondary disturbance information, and determining the behavior control value corresponding to each candidate supply chain adjustment behavior based on the disturbance suppression value and the secondary disturbance value. The behavior control value is determined according to the following formula: ， in, This represents the behavior control value corresponding to the i-th candidate supply chain adjustment behavior. This represents the disturbance suppression value of the i-th candidate supply chain adjustment behavior on the current supply chain disturbance. Let represent the secondary disturbance value corresponding to the i-th candidate supply chain adjustment behavior. This represents the secondary disturbance suppression coefficient; and the behavioral control value is used to perform execution control on the adjustment behavior of each candidate supply chain in order to determine the target supply chain adjustment behavior.

6. The supply chain management method based on artificial intelligence according to claim 5, characterized in that: The disturbance suppression value is determined based on the degree of change in supply chain disturbance before and after the corresponding candidate supply chain adjustment behavior is executed, and is used to characterize the ability of the candidate supply chain adjustment behavior to weaken the current supply chain disturbance. The secondary disturbance value is determined based on the degree of change in the subsequent operating status of the supply chain caused by the corresponding candidate supply chain adjustment behavior, and is used to characterize the degree of impact of the candidate supply chain adjustment behavior on the formation of new abnormal inputs in subsequent periods after its execution. The degree of change in the subsequent operation status of the supply chain is determined based on at least one of the following: changes in the pace of procurement and material expediting, changes in work order priority, changes in resource utilization relationships, and changes in expected order commitments.

7. The supply chain management method based on artificial intelligence according to claim 6, characterized in that: The secondary disturbance information; for each of the supply chain adjustment behaviors, analyze the impact of the supply chain adjustment behavior on the subsequent operation of the supply chain; The execution control of each candidate supply chain adjustment behavior based on the behavior control value means: when the behavior control value is greater than a preset execution threshold, the corresponding candidate supply chain adjustment behavior is executed; when the behavior control value is less than or equal to the preset execution threshold and greater than a preset restriction threshold, the execution of the corresponding candidate supply chain adjustment behavior is restricted; when the behavior control value is less than or equal to the preset restriction threshold, the execution of the corresponding candidate supply chain adjustment behavior is prohibited.

8. The supply chain management method based on artificial intelligence according to claim 7, characterized in that: The restricted implementation includes at least one of the following: delayed implementation, partial implementation, reduced adjustment magnitude, and replacement with alternative adjustment behaviors that have a smaller impact on the subsequent operation of the supply chain; After determining the target supply chain adjustment behavior, the method also includes outputting the corresponding supply chain management results based on the target supply chain adjustment behavior.

Images