A material uniformity management system and method

Abstract

The application provides a material completeness management system and method, which comprises a completeness verification module. The completeness verification module can determine the material demand of each work order one by one for a work order list, allocate materials according to the real-time inventory of various materials, and finally generate a completeness detail containing the material type, demand quantity, allocated quantity and gap quantity according to the allocation result. The module automatically allocates materials for work orders according to real-time inventory, realizes fine and automatic real-time completeness verification, and provides visual gap details to assist workers in quickly locating the shortage materials and specific quantity, significantly improving the completeness verification efficiency and accuracy, and enhancing the explainability.

Description

Technical Field

[0001] This application relates to the field of discrete manufacturing technology, and in particular to a material kitting management system and method. Background Technology

[0002] In discrete manufacturing, material shortages are a major cause of production line shutdowns and order delivery delays. Traditional material management relies on manual data comparison from multiple systems such as ERP (Enterprise Resource Planning) and WMS (Warehouse Management System), which is inefficient, error-prone, and unable to handle complex bill of materials (BOM) structures and supply chain uncertainties, thus failing to provide early warnings and rapid responses.

[0003] Traditional solutions are mainly divided into two categories: one is static inventory verification in ERP or APS (Advanced Planning and Scheduling) systems, which only makes a binary judgment based on the current inventory and work order demand, without considering the uncertainty of in-transit procurement, work orders in progress, and time, and has limited early warning capabilities; the other is simple early warning based on statistical methods, which triggers early warning through safety stock, but relies on assumptions and does not utilize multi-dimensional features, resulting in a high false alarm and false negative rate.

[0004] It is evident that traditional anti-analysis generally lacks foresight, has low precision, is not aware of the resource competition optimization capabilities among multiple work orders, and has poor interpretability, failing to provide root cause analysis and decision-making suggestions, resulting in an overall insufficient level of intelligence. Summary of the Invention

[0005] This application provides a material completeness management system and method to solve the problems of lag, low accuracy and low intelligence in traditional material completeness management methods.

[0006] Specifically, this application provides the following technical solution: Firstly, this application provides a material completeness management system, which includes a completeness verification module. The completeness verification module is used to perform current completeness verification on work orders in the following manner: For each work order in the list of work orders to be verified, determine the material requirements for that work order; Based on the material requirements of each work order and the real-time inventory of various materials, the work order is allocated materials. Based on the material allocation results of each work order, determine the completeness status of the work order at the current time and generate completeness details. The completeness status includes complete, partially complete and incomplete. The completeness details include the types of materials required for the current work order, the required quantity of each material, the allocated quantity and the shortage quantity.

[0007] Secondly, this application provides a method for managing material completeness, the method comprising: For each work order in the list of work orders to be verified, determine the material requirements for that work order; Based on the material requirements of each work order and the real-time inventory of various materials, the work order is allocated materials. Based on the material allocation results of each work order, determine the completeness status of the work order at the current time and generate completeness details. The completeness status includes complete, partially complete and incomplete. The completeness details include the types of materials required for the current work order, the required quantity of each material, the allocated quantity and the shortage quantity.

[0008] Thirdly, this application provides an electronic device, comprising: A memory, one or more processors; the memory is coupled to the processors; wherein the memory stores computer program code, the computer program code including computer instructions, and when the computer instructions are executed by the processor, the electronic device performs the method described above.

[0009] Fourthly, this application provides a computer-readable storage medium including computer instructions that, when executed on an electronic device, cause the electronic device to perform the method described above.

[0010] Fifthly, this application provides a computer program product that, when run on a computer, causes the computer to perform the method described above.

[0011] The technical solution provided in this application has the following beneficial effects: This application provides a material completeness management system, including a completeness verification module. This module determines the material requirements for each work order in a list of work orders, allocates materials based on real-time inventory levels, and generates a completeness detail list including material type, required quantity, allocated quantity, and shortage quantity. This application automatically allocates materials to work orders based on real-time inventory, achieving refined and automated completeness verification, providing a visualized shortage detail list, helping staff quickly locate missing materials and their specific quantities, quantifying the relationship between inventory and demand, and providing a data foundation for subsequent handling of resource competition among multiple work orders and determining priority allocation.

[0012] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0013] The accompanying drawings, which are incorporated in and form part of this application, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0014] Figure 1 A schematic diagram of the material completeness management system provided in an embodiment of this application; Figure 2 A flowchart illustrating the material completeness management method provided in this application embodiment; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0015] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application.

[0016] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used herein are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the corresponding listed items.

[0017] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0018] The technical solution and its beneficial effects of this application will be described in detail below through specific embodiments.

[0019] This application provides a material completeness management system, such as Figure 1 As shown, the system includes a kitting verification module, and may further include an inventory simulation module and a reinforcement learning module.

[0020] Among them, the completeness verification module is based on deterministic rules and is used to verify the completeness status of work orders in real time and accurately, and output detailed material shortage details; it is also used to intelligently determine the material allocation order of each work order based on multiple factors such as work order priority, planned start time, and delivery date when inventory resources are limited, so as to maximize the overall production efficiency. The inventory simulation module is used to simulate the inventory of various materials at different points in the future based on timeline events, thereby enabling proactive and accurate prediction of future material shortage risks and issuing tiered early warnings in advance; it can also provide interpretable root cause analysis and specific action suggestions for material shortage warning results, improving the interpretability of the system; The kitting verification module can also work in conjunction with the inventory simulation module to predict the kitting status of work orders at the target future time, such as the planned start time, based on the simulated inventory of various materials output by the inventory simulation module. The reinforcement learning module can work in conjunction with the inventory simulation module and the completeness verification module. When the completeness status prediction result output by the completeness verification module is not ideal, the reinforcement learning mechanism can optimize and adjust the relevant parameters of the inventory simulation module in the inventory simulation process and the graded early warning process.

[0021] This application combines deterministic rules, timeline simulation, and reinforcement learning to construct a hybrid material kitting management system that combines reliability, foresight, and self-learning capabilities.

[0022] Below is a comparison Figure 1 Each module of the material completeness management system will be introduced separately.

[0023] (a) Completeness verification module The completeness verification module is used to perform real-time and accurate verification of the completeness status of work orders at the current point in time, and output detailed completeness information. For example... Figure 1 As shown, the inputs to the module at this time are the real-time inventory of various materials in the material warehouse, the work orders to be verified, and the bill of materials. The outputs are the completeness status and completeness details of the work orders.

[0024] Specifically, this module can implement the following (11) to (13): (11) For each work order in the list of work orders to be verified, determine the material requirements for that work order.

[0025] Specifically, based on the list of work orders to be verified and the bill of materials, the material requirements for each work order are determined. These material requirements include the types and quantities of materials needed for the work order. The list of work orders to be verified is a collection of production work orders, each containing information such as the product code, production quantity, and planned start time. The bill of materials is a product structure document that details all raw materials and components required to produce one unit of product, along with their quantity relationships. It may also include information such as hierarchical structure and loss rate.

[0026] By combining the products to be produced and their quantities indicated in each work order in the work order list to be verified with the material composition corresponding to each product in the bill of materials, the types and quantities of various materials required for each work order can be obtained, thus providing a data foundation for subsequent completeness verification.

[0027] Optionally, the types and quantities of materials required for each work order can be obtained in the following ways: By executing a BOM recursive expansion algorithm, starting from the product to be manufactured described in the work order, the BOM tree (formatted data of the bill of materials) is recursively traversed: for each level, the required quantity of child materials (i.e., the direct subordinate materials that make up the parent item) is calculated based on the current parent item's (i.e., the expanded material) required quantity and the BOM usage; if the child material is a semi-finished product (i.e., there is still a subordinate BOM), its subordinate BOM is expanded, until all bottom-level materials are obtained. Finally, all material types and their required quantities for each work order are summarized.

[0028] Furthermore, when determining the required quantity of materials, the loss rate needs to be considered. For each material, first determine its theoretical required quantity, and then determine the actual required quantity based on the loss rate. That is, the required quantity in the aforementioned material requirements is specifically the actual required quantity calculated based on the loss rate and the theoretical required quantity.

[0029] Finally, a material requirement is output for each work order. This material requirement records the following information for each material required by the work order: material code / material name, required quantity (after considering loss rate), requirement level (for recursive expansion), and whether it is a critical material.

[0030] (12) Based on the material requirements of each work order and the real-time inventory of various materials, allocate materials to the work order.

[0031] Optionally, firstly, based on the core variables of the work order and according to the preset priority rules, calculate the comprehensive priority score of the work order, and then use the order of comprehensive priority scores from high to low as the material allocation order, and allocate materials to each work order in sequence according to the material allocation order.

[0032] The aforementioned priority rules include rules for quantifying and weighting each core variable. Specifically, core variables may include any one or more of the following: work order priority, planned start time, and VIP customer identifier.

[0033] For example, firstly, each core variable is quantified. For instance, the VIP customer identifier is assigned a value of 1 (yes) or 0 (no). The work order priority is converted into a numerical value according to a preset level (e.g., 100 for level 1 priority and 80 for level 2 priority). The planned start time is converted into a time score according to its distance from the current time (e.g., the earlier the start time, the higher the score). Secondly, a preset weight coefficient is assigned to each variable, with the VIP customer identifier having the highest weight, followed by the work order priority, and then the planned start time. Then, the quantified values ​​are weighted and summed with their corresponding weight coefficients to obtain the comprehensive priority score for each work order. Finally, the material allocation order of the work orders is determined according to the comprehensive priority scores from high to low.

[0034] Through the aforementioned priority mechanism, when inventory resources are limited, multiple factors such as work order priority and planned start time can be comprehensively considered to effectively coordinate the competition of multiple work orders for limited inventory, ensuring the certainty, uniqueness and fairness of material allocation, and maximizing overall production efficiency.

[0035] In this application, real-time inventory is divided into three categories based on its status: available inventory, locked inventory, and reserved inventory. The following is a definition of these three types of inventory: Available inventory refers to inventory that actually exists in the warehouse, meets quality standards, and is not currently occupied by any work orders, and can be freely allocated. In practical applications, materials are marked as "available" by default after they are received into the warehouse and pass quality inspection; when a work order is canceled or the inventory is released, the inventory status will be rolled back to "available".

[0036] Reserved inventory refers to inventory that has been allocated to a specific work order by the system or algorithm but has not yet been actually issued. This inventory is physically still in the warehouse, but logically it has been reserved and cannot be allocated to other work orders. In practical applications, when a multi-work order inventory allocation algorithm allocates inventory to a work order according to priority rules, the corresponding inventory is marked as "reserved".

[0037] Locked inventory refers to inventory that actually exists in the warehouse but is frozen for some reason and temporarily cannot be used for work order allocation. In practical applications, when material quality abnormalities occur (such as pending inspection or non-conforming), during warehouse inventory counts, or when special management freezes occur, the corresponding inventory is marked as "locked".

[0038] Optionally, when allocating materials to each work order based on real-time inventory, only available inventory is allocated; once inventory is allocated to a work order, its status is changed from available inventory to reserved inventory.

[0039] For example, when allocating materials to each work order according to the material allocation order, the available inventory of the materials required by each work order is checked one by one, and the quantity of the material required by the work order is deducted accordingly. If the available inventory of a certain material is insufficient, only the actually available portion is allocated, and the remaining required quantity is recorded as a shortage. After allocation, the status of the allocated inventory is changed from "available" to "reserved" to ensure that it will not be repeatedly used in subsequent work order allocations. This mechanism ensures the consistency and accuracy of the inventory allocation process by locking the inventory as it is allocated.

[0040] Optionally, the "alternative material rule" may be used as a reference during the material allocation process.

[0041] The substitution rule refers to the system's ability to allocate alternative materials (substitutes) with similar functions or specifications when the inventory of a certain material (main material) is insufficient, based on preset substitution relationships. In the allocation algorithm, when the available inventory of the main material is insufficient, the system automatically searches for available substitutes that meet the rules. If found, the substitute is included in the allocation scope according to the rules, thereby improving the completeness rate of work orders. The usage of substitutes is also recorded in the allocation results for subsequent traceability and analysis.

[0042] Optionally, during the material allocation process, a "critical material veto rule" is implemented. That is, if any critical material required by any work order is missing, no material allocation will be performed on that work order, and the work order's completeness status will be directly determined as "incomplete".

[0043] Among them, critical materials refer to materials that have a decisive impact on production progress or product quality, and they enjoy a special judgment mechanism during the allocation process. If, before or during the allocation process, it is detected that the real-time inventory of a critical material is insufficient, or the available inventory is insufficient, or the real-time inventory is insufficient and there is no substitute material, or the available inventory is insufficient and there is no substitute material, the system will immediately terminate further allocation for that work order, directly mark it as "incomplete set", and no other materials will be allocated to that work order.

[0044] The aforementioned veto mechanism aims to prevent production interruptions or resource waste due to the lack of key materials, and to ensure that limited inventory is prioritized for work orders that can be completed, thereby improving the feasibility and execution efficiency of the overall production plan.

[0045] (13) Based on the allocation results of each work order, determine the kitting status of each work order at the current time and generate kitting details. Kitting details include the types of materials required for the work order, the required quantity of each material, the allocated quantity, and the shortage quantity.

[0046] Complete sets include complete sets, partially complete sets, and incomplete sets. The following is an explanation of each of these three complete sets: Full kit indicates that all materials required for the work order are ready and can be put into production. In practical applications, when the allocated quantity of all materials under the work order equals the required quantity (i.e., there is no shortage), it will be marked as "full kit".

[0047] Partial Kit (PARTIAL_KIT) indicates that the materials required for this work order are only partially met, and some materials are still in short supply. In practical applications, when, after material allocation, the allocated quantity of at least one material under the work order is less than the required quantity (i.e., there is a gap), but the "critical material veto rule" has not been triggered, it will be marked as "partial kit".

[0048] Not Kit / Critical Material Missing (NOT_KIT) indicates that the work order cannot proceed with production due to the complete absence of critical materials. In practical applications, when the "critical material veto rule" is triggered during material allocation, the system will terminate the material allocation for that work order and directly mark it as "not kitted".

[0049] The complete set details are a detailed record of the allocation results for each work order. Specifically, it may include the following fields: material code, material name, required quantity, allocated quantity, shortage quantity, and whether it is a critical material, which is used to support subsequent material shortage analysis and production scheduling.

[0050] The above-mentioned completeness verification module performs real-time completeness verification based on deterministic rules. All judgment logic has clear basis, and the results are 100% interpretable.

[0051] (ii) Inventory Simulation Module The inventory simulation module is used to dynamically maintain the simulated inventory of each material at future points in time based on timeline events, and to provide material shortage warnings for each material based on the simulated inventory.

[0052] Specifically, this module can implement the following (21) to (22): (21) Based on the future work order plan, purchase order plan and work-in-process work order plan within the future time window, construct a time-sorted inbound and outbound event sequence for each material; based on the inbound and outbound event sequence, determine the simulated inventory of each material at each future time point.

[0053] For a future time window (e.g., 90 days), the system obtains future work order plans, purchase order plans, and work-in-process (WPC) plans (work orders for semi-finished products currently in production but not yet completed; these semi-finished products, such as components, also fall under the category of materials). Subsequently, using time-axis event modeling and rolling inventory simulation algorithms, future work order plans are converted into future outbound events, and purchase order plans and WPC plans are converted into future inbound events, thereby determining the simulated inventory of each material at various future points in time.

[0054] Specifically, a time-ordered sequence of inbound and outbound events is constructed for each material. This sequence includes initial inventory events, future inbound events (corresponding to receipt of purchase orders or completion of work orders for in-process components), and future outbound events (corresponding to production scheduling of future work orders). Then, each event is processed sequentially according to time order, and the simulated inventory balance of the material is dynamically updated to obtain the simulated inventory of the material at various future time points.

[0055] For example, for any purchase order in the purchase order plan, assuming the purchase target of the purchase order is materials A, B, and C, this application will generate a future receipt event for materials A, B, and C respectively. The receipt time of each future receipt event on the timeline is the arrival time of the purchase order. Similarly, for any work-in-process work order in the work-in-process work order plan, assuming the production target of the work-in-process work order is materials D and E, this application's system will generate a future receipt event for materials D and E respectively. The receipt time of each future receipt event on the timeline is the completion time of the work-in-process work order.

[0056] Optionally, the inventory simulation module constructs a time-ordered sequence of inbound and outbound events for each material in the following ways: For any purchase order in the purchase order plan, a first machine learning model is used to generate a delivery time prediction result; for any work-in-process sub-parts in the work-in-process work order plan, a second machine learning model is used to generate a completion time prediction result; based on the future work order plan, purchase order plan and delivery time prediction result, and work-in-process sub-parts plan and completion time prediction result within the future time window, a time-ordered sequence of inbound and outbound events is constructed for each material.

[0057] In other words, in practical applications, the delivery time of purchase orders is affected by many factors such as suppliers and logistics, while the completion time of work-in-process orders is constrained by production scheduling and seasonal factors, both of which have a certain degree of uncertainty. Therefore, for purchase orders whose actual delivery time cannot be determined, this application uses a first machine learning model to predict the delivery time of such purchase orders; correspondingly, for work-in-process orders whose actual completion time cannot be determined, this application uses a second machine learning model to predict the completion time of such work-in-process orders.

[0058] Specifically, a training dataset is constructed based on historical purchase order receipt records and historical work-in-process component completion records, encompassing material characteristics, supplier characteristics, order characteristics, and time-series characteristics. A gradient boosting model (such as XGBoost or LightGBM) is used to train a regression model to predict the delivery time or delay days of purchase orders, and it supports outputting the time distribution through quantile regression (a regression analysis method). Similarly, a corresponding model can also be trained to predict the completion time of self-manufactured component work orders.

[0059] In addition to gradient boosting models, other machine learning algorithms, such as deep neural networks, random forests, or time series models (such as LSTM and Transformer), can also be used, and this application does not limit them.

[0060] (22) During the inventory simulation process, for each future outbound event, the shortage level of the target material is determined based on the quantitative relationship between the demand quantity of the target material for the future outbound event, the simulated inventory of the target material, and the safety stock threshold of the target material, and a corresponding shortage warning is generated.

[0061] Specifically, when processing events in chronological order, whenever a future outbound event is encountered, the system determines the warning level for the target material based on the simulated inventory of the target material according to the following rules: Red alert (definitely short of materials): Simulated inventory is less than the quantity of the target material required for the future outbound event; Yellow alert (risk of material shortage): The simulated inventory minus the demand is below the safety stock threshold of the target material; Green alert (safe): Other situations.

[0062] Optionally, the material shortage alerts generated by this application include standardized root cause codes (such as supplier delays, unexpected consumption) and suggested action codes (such as order expediting, activating alternative materials), which are automatically converted into natural language descriptions for easy understanding and response by relevant personnel. The alert records also link corresponding future outbound events and future work orders to trace the specific source of the material shortage demand.

[0063] By attaching root cause codes and suggested action codes to material shortage warnings, a structured expression of material shortage risks is achieved. Simultaneously, the standardized codes are converted into natural language descriptions, significantly reducing communication costs during the warning response process and improving material management efficiency in production. Furthermore, associating material shortage risks with corresponding future outbound events and future work orders allows for precise tracing of the specific demand source, facilitating rapid problem identification and targeted measures by staff. This enhances the clarity, interpretability, and operability of material shortage warnings, shortens warning response time, reduces decision-making delays caused by ambiguous information or difficulty in location, and provides more efficient and accurate early warning support for enterprise material security.

[0064] Based on the inventory simulation module's ability to output simulated inventory levels for various materials, the completeness verification module in this application can also predict the completeness status of future work orders. For example... Figure 1 As shown, the inputs to the module are the simulated inventory of various materials, future work orders, and bill of materials. The output is the predicted kitting status of future work orders at the target future time.

[0065] Specifically, the completeness check module can perform the following (14): (14) Based on the simulated inventory of various materials, predict the availability status of future work orders at the target future time (including but not limited to the planned start time) and obtain the availability status prediction results.

[0066] Specifically, the system receives and responds to a kitting status prediction instruction for future work orders. In practical applications, this prediction instruction can specify any future time, such as the planned start time, for example: predicting the kitting status of future work order F at a future time G.

[0067] In response to the prediction instruction, the system obtains the simulated inventory of various materials at the target future time from the inventory simulation module. Then, based on the material requirements of the future work order, the system performs simulated material allocation for the future work order based on the obtained simulated inventory of various materials at the target future time, thereby obtaining the kitting status prediction result of the future work order at the target future time.

[0068] The above-mentioned completeness prediction results can include the completeness probability. Furthermore, when the completeness probability is lower than the completeness probability threshold, a probabilistic completeness risk warning can be generated to alert staff that there is a risk of incompleteness in future work orders.

[0069] The process of determining the material requirements for future work orders, as well as the process of simulating material allocation for future work orders, can be referred to in the previous text and will not be repeated here.

[0070] (III) Reinforcement Learning Module Based on the use of the completeness verification module to predict the completeness status of future work orders, this application further introduces a reinforcement learning module.

[0071] like Figure 1 As shown, the reinforcement learning module is triggered when the matching state prediction result does not meet the preset standard. Specifically, this module can achieve the function described in (31): (31) When the prediction result of the complete set state does not meet the preset standard, the safety stock threshold and / or the length of the future time window of the inventory simulation module are adjusted through the reinforcement learning mechanism.

[0072] Specifically, a reinforcement learning mechanism is introduced to establish a "prediction result - actual result" feedback loop. When the prediction accuracy does not meet the requirements, the system dynamically adjusts the relevant parameters of the inventory simulation module through this feedback loop, including the safety stock threshold and the length of the future time window, thereby achieving adaptive optimization of the model parameters and forming a complete closed loop of "early warning - feedback - optimization".

[0073] For example, the reward function for reinforcement learning can be designed as follows: R = a. Precision + b. Recall - c. False Alarm Rate - d. Inventory Cost - e. Downtime Penalty Among them, a, b, c, d, and e are the weights of each reward or penalty item; accuracy rate refers to the proportion of correct warnings, and improving accuracy rate will result in positive rewards; recall rate refers to the proportion of actual delays that are successfully warned, and improving recall rate will also result in positive rewards; false alarm rate refers to the proportion of false warnings, and an increase in false alarm rate will result in negative penalties; inventory cost refers to the cost increase caused by rising inventory levels, which is a negative indicator; and production stoppage penalty refers to production stoppage caused by material shortages, which is a serious negative indicator.

[0074] In addition to reinforcement learning, traditional operations research methods (such as linear programming) can also be used for parameter tuning, and this application does not limit this to either.

[0075] This concludes the introduction to the three modules.

[0076] As can be seen from the above technical solutions, this application adopts a three-module progressive hybrid architecture design, which organically combines the completeness verification module (based on deterministic rules), the inventory simulation module (based on time axis rules) and the reinforcement learning module (based on reinforcement learning dynamic optimization rules) to achieve a complete capability closed loop from real-time verification to intelligent prediction.

[0077] Specifically, in the kitting compliance verification module, the material requirements of the work order are determined through a BOM recursive expansion algorithm. Priority ranking and a veto mechanism for critical materials ensure the accuracy and fairness of kitting compliance verification. The inventory simulation module employs time-axis event modeling and rolling inventory simulation algorithms to simulate the inventory of various materials at future points in time, dynamically extrapolating future supply and demand balance, and thus achieving tiered early warning of material shortage risks. To address the issue of time uncertainty, this module uses a gradient boosting model to train a regression model, constructing a predictive model for procurement arrival time and work order completion time, and supports quantile regression. By transmitting the simulated inventory to the kitting compliance verification module, it assists in achieving kitting status prediction and kitting risk early warning based on prediction probability. When kitting prediction is inaccurate, the reinforcement learning module uses reinforcement learning to construct a "prediction-actual" feedback loop, achieving adaptive dynamic optimization of model parameters such as early warning thresholds and time windows.

[0078] It is worth mentioning that the three modules could be further decoupled, or even deployed as independent microservices. This loosely coupled architecture facilitates independent upgrades and maintenance. Despite the existence of the above alternatives, the three-module progressive hybrid architecture adopted in this application complements each other in terms of deterministic rules ensuring reliability, timeline simulation providing foresight, machine learning prediction improving accuracy, and reinforcement learning achieving adaptability. It is the best practice for achieving high-precision, high-reliability, and highly intelligent early warning systems under current technological conditions.

[0079] Through the above-described technical solution, this application can achieve the following beneficial effects: First, in terms of accuracy, the system can achieve a current kitting accuracy rate of ≥99%, a future material shortage warning accuracy rate of ≥90%, and a median warning lead time of ≥24 hours, which can accurately locate production risks and leave sufficient time for business intervention.

[0080] Secondly, in terms of performance, the system is highly efficient. A single full verification (taking 5,000 work orders as an example) can be completed in minutes, and the average response time of the near real-time API interface is ≤500 milliseconds, which meets the needs of high-frequency queries.

[0081] Third, in terms of intelligence, the system uses machine learning to predict the delivery time of purchases, with the average absolute error of the prediction controlled within 1.5 days. It also introduces probabilistic early warning and adaptive optimization, which significantly improves the foresight and adaptability of the completeness risk warning.

[0082] Fourth, in terms of interpretability and operability, the system provides structured root cause analysis and specific action suggestions for each warning, and supports complete decision trajectory auditing, which greatly reduces the difficulty for business personnel to locate problems and make decisions.

[0083] Fifth, in terms of business value, after the system goes live, it can effectively reduce production downtime caused by material shortages, improve overall equipment efficiency and on-time order delivery rate, and reduce inventory holding costs and emergency procurement costs by optimizing inventory allocation.

[0084] This application can be widely used in discrete manufacturing industries such as electronic manufacturing and mechanical assembly, providing enterprises with real-time perception, accurate prediction and intelligent decision support for material supply.

[0085] Based on the same inventive concept, this application provides a method for managing material completeness.

[0086] like Figure 2 As shown, the method specifically includes: Step 210: For each work order in the list of work orders to be verified, determine the material requirements for that work order; Step 220: Based on the material requirements of each work order and the real-time inventory of various materials, allocate materials for the work order. Step 230: Based on the material allocation results of each work order, determine the material availability status of the work order at the current time and generate a material availability list. The material availability status includes complete, partially complete, and incomplete. The material availability list includes the types of materials required for the current work order, the required quantity of each material, the allocated quantity, and the shortage quantity.

[0087] Optionally, the method specifically allocates materials to work orders in the following ways: According to the preset priority rules, the comprehensive priority score of the work order is calculated based on the core variables of the work order, wherein the core variables include any one or more of the following: work order priority, planned start time, VIP customer identifier; the comprehensive priority scores are used as the material allocation order from high to low, and the materials are allocated to each work order in sequence according to the material allocation order.

[0088] Optionally, the real-time inventory includes available inventory, locked inventory, and reserved inventory, and the method specifically allocates materials to work orders in the following ways: Based on the available inventory of various materials, materials are allocated to work orders, and the status of the inventory allocated to the work orders is changed from available inventory to reserved inventory.

[0089] Optionally, the method specifically allocates materials to work orders in the following ways: Based on the alternative material rules, allocate materials to work orders.

[0090] Optionally, the method further includes: The critical material veto rule is implemented as follows: if the real-time inventory of the critical material required by any work order is less than the quantity of the critical material required by the work order, then no material allocation will be made for the work order, and the completeness status of the work order will be directly determined as incomplete.

[0091] Optionally, the method further includes: Inventory simulation for various materials is performed as follows: Based on future work order plans, purchase order plans, and work-in-process work order plans within future time windows, a time-ordered sequence of inbound and outbound events is constructed for each material. The inbound and outbound event sequence includes future outbound events determined based on future work order plans and future inbound events determined based on purchase order plans or work-in-process work order plans. Based on the inbound and outbound event sequence, the simulated inventory of each material at various future time points is determined.

[0092] Optionally, the method specifically constructs a time-ordered sequence of inbound and outbound events for each material in the following manner: For any purchase order in the purchase order plan, the first machine learning model is used to generate a delivery time prediction result; for any work-in-process work order in the work-in-process work order plan, the second machine learning model is used to generate a completion time prediction result; based on the future work order plan, purchase order plan and delivery time prediction result, work-in-process work order plan and completion time prediction result within the future time window, a time-sorted inbound and outbound event sequence is constructed for each material.

[0093] Optionally, the method further includes: Shortage warnings for various materials are generated in the following manner: For each future outbound event, the shortage level of the target material is determined based on the quantitative relationship between the required quantity of the target material for that future outbound event, the simulated inventory of the target material, and the safety stock threshold of the target material, and a corresponding shortage warning is generated.

[0094] Optionally, the method generates a material shortage warning in the following manner: Generate a material shortage warning that includes a root cause code and a suggested action code, and associate it with the corresponding future outbound event or future work order.

[0095] Optionally, the method further includes: The following method is used to predict the future availability of work orders: Based on the material requirements of future work orders and the simulated inventory of various materials, the availability status of future work orders at the target future time is predicted to obtain the availability status prediction result.

[0096] Optionally, the method further includes: When the predicted results of the completeness state do not meet the preset standards, the safety stock threshold and / or the size of the future time window of the inventory simulation module are adjusted through a reinforcement learning mechanism.

[0097] The material completeness method provided in this application is applied to the material completeness management described above. Therefore, the beneficial effects achieved by this method can be referenced from the beneficial effects of the material completeness system described above, and will not be repeated here.

[0098] This application provides an electronic device that may include a memory and one or more processors. The memory stores computer program code, including computer instructions. When the processor executes the computer instructions, the electronic device can perform various functions or steps of the above-described method embodiments.

[0099] The structure of this electronic device can be referenced. Figure 3 The structure of the electronic device 100 shown.

[0100] For example, the processor mentioned above can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0101] This application provides a computer-readable storage medium including computer instructions that, when executed on an electronic device, cause the electronic device to perform the various functions or steps of the above-described method embodiments.

[0102] The aforementioned computer-readable storage media include, but are not limited to, any of the following: USB flash drive, portable hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, and other media capable of storing program code.

[0103] This application provides a computer program product that, when run on a computer, causes the computer to perform various functions or steps of the above-described method embodiments.

[0104] The electronic devices, computer-readable storage media, and computer program products provided in the above embodiments are all used to execute the corresponding methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0105] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0106] The specific implementation process of the functions and roles of each unit / module in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0107] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units / modules described as separate components may or may not be physically separate. The components shown as units / modules may or may not be physical units / modules, that is, they may be located in one place or distributed across multiple network units / modules. Some or all of the units / modules can be selected to achieve the purpose of this application according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0108] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A material completeness management system, characterized in that, The material completeness management system includes a completeness verification module, which is used to perform current completeness verification on work orders in the following manner: For each work order in the list of work orders to be verified, determine the material requirements for that work order; Based on the material requirements of each work order and the real-time inventory of various materials, the work order is allocated materials. Based on the material allocation results of each work order, determine the completeness status of the work order at the current time and generate completeness details. The completeness status includes complete, partially complete and incomplete. The completeness details include the types of materials required for the current work order, the required quantity of each material, the allocated quantity and the shortage quantity.

2. The material completeness management system according to claim 1, characterized in that, The completeness verification module allocates materials to work orders in the following ways: According to the preset priority rules, the comprehensive priority score of the work order is calculated based on the core variables of the work order, wherein the core variables include any one or more of the following: work order priority, planned start time, VIP customer identifier; The material allocation order is determined by ranking the overall priority scores from highest to lowest, and materials are allocated to each work order in that order.

3. The material completeness management system according to claim 1, characterized in that, The real-time inventory includes available inventory, locked inventory, and reserved inventory. The completeness verification module allocates materials to work orders in the following ways: Based on the available inventory of various materials, materials are allocated to work orders, and the status of the inventory allocated to the work orders is changed from available inventory to reserved inventory.

4. The material completeness management system according to claim 1, characterized in that, The completeness verification module allocates materials to work orders in the following ways: Based on the alternative material rules, allocate materials to work orders.

5. The material completeness management system according to claim 1, characterized in that, The completeness verification module is also used to execute the critical material veto rule in the following ways: If the real-time inventory of the critical materials required for any work order is less than the quantity of the critical materials required by the work order, then no material allocation will be performed for the work order, and the completeness status of the work order will be directly determined as incomplete.

6. The material completeness management system according to claim 1, characterized in that, The material completeness management system also includes an inventory simulation module, which is used to simulate the inventory of various materials in the following manner: Based on the future work order plan, purchase order plan, and work-in-process work order plan within the future time window, construct a time-ordered sequence of inbound and outbound events for each material. The sequence of inbound and outbound events includes future outbound events determined according to the future work order plan and future inbound events determined according to the purchase order plan or work-in-process work order plan. Based on the inbound and outbound event sequence, determine the simulated inventory of each material at various future time points.

7. The material kitting management system according to claim 6, characterized in that, The inventory simulation module specifically constructs a time-ordered sequence of inbound and outbound events for each material in the following way: For any purchase order in the purchase order plan, the first machine learning model is used to generate a delivery time prediction result. For any work-in-process work order in the work-in-process work order plan, the second machine learning model is used to generate a completion time prediction result. Based on the future work order plan, purchase order plan and delivery time forecast, and work-in-process work order plan and completion time forecast within the future time window, construct a time-ordered sequence of inbound and outbound events for each material.

8. The material completeness management system according to claim 6, characterized in that, The inventory simulation module is also used to provide shortage warnings for various materials in the following ways: For each future outbound event, the shortage level of the target material is determined based on the quantitative relationship between the demand quantity of the target material for that future outbound event, the simulated inventory of the target material, and the safety stock threshold of the target material, and a corresponding shortage warning is generated.

9. The material completeness management system according to claim 8, characterized in that, The inventory simulation module is also used to generate shortage warnings in the following manner: Generate a material shortage warning that includes a root cause code and a suggested action code, and associate it with the corresponding future outbound event or future work order.

10. The material completeness management system according to claim 6, characterized in that, The completeness verification module is also used to predict the future completeness of work orders in the following manner: Based on the material requirements of future work orders and the simulated inventory of various materials, the availability status of future work orders at the target future time is predicted, and the availability status prediction result is obtained.

11. The material completeness management system according to claim 10, characterized in that, The material completeness management system further includes a reinforcement learning module, which is used for: When the predicted results of the completeness state do not meet the preset standards, the safety stock threshold and / or the size of the future time window of the inventory simulation module are adjusted through a reinforcement learning mechanism.

12. A method for managing material completeness, characterized in that, The method includes: For each work order in the list of work orders to be verified, determine the material requirements for that work order; Based on the material requirements of each work order and the real-time inventory of various materials, the work order is allocated materials. Based on the material allocation results of each work order, determine the completeness status of the work order at the current time and generate completeness details. The completeness status includes complete, partially complete and incomplete. The completeness details include the types of materials required for the current work order, the required quantity of each material, the allocated quantity and the shortage quantity.

13. An electronic device, characterized in that, include: A memory, one or more processors; the memory is coupled to the processors; wherein the memory stores computer program code, the computer program code including computer instructions, and when the computer instructions are executed by the processor, the electronic device performs the method as described in claim 12.

14. A computer-readable storage medium comprising computer instructions, characterized in that, When the computer instructions are executed on the electronic device, the electronic device causes the electronic device to perform the method as described in claim 12.

15. A computer program product, characterized in that, When the computer program product is run on a computer, it causes the computer to perform the method as described in claim 12.

Images