An intelligent production scheduling optimization method and system for a multi-order multi-constraint scenario

Abstract

The application discloses a kind of intelligent production scheduling optimization method and system for multi-order multi-constraint scene, belong to production scheduling optimization and manufacturing informatization technical field, including the following steps: S1, instruction receiving and pre-processing;S2, semantic analysis and constraint completion;S3, multi-order clustering grouping;S4, finished product inventory matching determination;S5, BOM disassembly and material complete set verification;S6, production resource availability evaluation;S7, multi-constraint scheduling model construction and intelligent solution;S8, core resource locking and inventory pre-control: lock the key equipment, skilled personnel and scarce materials in scheduling plan, execute raw material inventory pre-deduction locking, after production is completed, update finished product, work-in-process and raw material inventory data simultaneously;S9, scheduling result visual output, the present application has natural language convenient interaction, multi-constraint unified modeling, intelligent automatic scheduling, disturbance adaptive rearrangement, resource collaborative optimization, decision logic traceable technical effect.

Description

Technical Field

[0001] This invention belongs to the field of production scheduling optimization and manufacturing information technology, and in particular relates to an intelligent production scheduling optimization method and system for scenarios with multiple orders and multiple constraints. Background Technology

[0002] Manufacturing is a crucial foundation for the national economy. Small and medium-sized manufacturing enterprises (SMEs) are numerous and widely distributed across discrete manufacturing scenarios such as machining and electronic assembly, playing a vital role in supply chain support and regional economic development. As market demand increasingly exhibits characteristics of "multiple varieties, small batches, short lead times, and frequent changes," enterprise production organization is gradually shifting from relatively stable batch manufacturing to order-driven flexible manufacturing. In this process, how to quickly generate executable production plans for multiple orders, multiple SKUs, and multiple processes under limited equipment, personnel, and materials, and how to promptly adjust production schedules in response to dynamic disturbances such as order insertions, material shortages, equipment failures, and staff absences, has become a key issue in the production management of SMEs. The rationality of production scheduling not only directly affects the timeliness of order delivery, equipment utilization, and work-in-process levels, but also relates to inventory costs, deferred costs, and overall production efficiency, which is of great significance for enterprises to respond quickly to market changes and maintain stable and healthy operations.

[0003] To improve production scheduling efficiency, some small and medium-sized manufacturing enterprises have introduced information systems such as ERP and MES, or adopted automatic scheduling methods based on priority rules and heuristic algorithms to achieve order sorting, process allocation, and resource scheduling. However, the production scheduling of these small and medium-sized manufacturing enterprises currently suffers from the following shortcomings: Manual scheduling relies on experience and is inefficient. Existing information systems lack adaptability: ERP, MES, and conventional automated scheduling methods can only handle static explicit constraints such as equipment capacity and process sequence, focusing on a single optimization goal and failing to simultaneously accommodate complex constraints such as order priority, personnel skill differences, equipment maintenance windows, and material procurement lead times. When faced with dynamic disturbances such as emergency order insertions and equipment failures, it is difficult to achieve staggered scheduling at the process level and real-time plan reconstruction, making it unsuitable for complex real-world production scenarios. Human-computer interaction is rigid: traditional scheduling systems use menu navigation, fixed form entry, and parameter configuration item by item, resulting in lengthy operation paths for planners. Emergency scheduling on-site suffers from cumbersome input, unintuitive demand expression, and inability to flexibly supplement missing information, leading to low emergency response efficiency. Summary of the Invention

[0004] To address the problems existing in the prior art, this invention provides an intelligent production scheduling optimization method and system for scenarios with multiple orders and multiple constraints. It has the advantages of convenient interaction with natural language, unified modeling of multiple constraints, intelligent automatic scheduling, adaptive rearrangement under disturbances, collaborative optimization of resources, and traceable decision logic. It solves the problems of high dependence on manual labor, incomplete constraint coverage, untimely response to on-site disturbances, and cumbersome system operation in the prior art.

[0005] This invention is implemented as follows: an intelligent production scheduling optimization method for scenarios with multiple orders and multiple constraints, comprising the following steps: S1. Command Reception and Preprocessing: Receive production scheduling commands from users in natural language or structured form, perform redundancy removal, remove function words and invalid words from the commands, and obtain standardized command text. S2. Semantic parsing and constraint completion: Perform industrial semantic parsing on standardized instruction text to extract product model, production quantity, delivery date, priority, alternative material permission, emergency order permission and other implicit constraints. Automatically identify missing key fields and guide users to complete them through human-computer interaction to form a complete production scheduling request. S3. Multi-order clustering and grouping: For scenarios with multiple orders and multiple SKUs running in parallel, the order set is sorted or clustered and grouped based on delivery date, order priority, process similarity, shared materials and equipment, and changeover costs. S4. Finished Goods Inventory Matching and Judgment: Query the real-time inventory of the target SKU. If the inventory is sufficient, generate an inventory delivery plan directly. If the inventory is partially sufficient, split the gap into the quantity to be produced. If there is no inventory, directly enter the production scheduling process. S5. BOM Decomposition and Material Matching Verification: Retrieve the BOM structure list of the product to be produced, decompose the raw materials and parts details, and perform matching verification in combination with real-time inventory and materials in transit. If the matching is not complete, a procurement replenishment plan will be generated and production will be triggered after the materials arrive. S6. Production resource availability assessment: Analyze equipment operating status, maintenance window, personnel shift skills, mold occupancy status, and combine historical data to predict equipment load, changeover time, and order delay risk; S7. Multi-constraint scheduling model construction and intelligent solution: Integrate multi-dimensional constraints of orders, materials and resources, and pre-train a scheduling decision model based on historical production data to perform global scheduling optimization of multi-order and multi-process production tasks and generate a production task timeline plan. S8. Core Resource Lock-in and Inventory Pre-control: Lock in key equipment, skilled personnel and scarce materials in the production schedule, implement raw material inventory pre-deduction and lock, and update finished product, work-in-process and raw material inventory data synchronously after production is completed. S9. Visualized Production Scheduling Results Output: The optimized production plan, resource load, and process layout are visualized in the form of time axis, Gantt chart, and statistical charts, supporting plan viewing, export, and production deployment.

[0006] As a preferred embodiment of the present invention, it further includes a dynamic disturbance response step: for real-time online monitoring of disturbance events at the production site, the disturbance events including emergency order insertion, raw material shortage, sudden equipment failure, temporary staff absence, and delayed delivery of outsourced parts; when a disturbance event is detected, the constraint parameters are updated, the affected production tasks are partially rearranged or the global plan is reconstructed, and the adjusted optimized production scheduling scheme is output.

[0007] This setting allows for real-time detection of various abnormal disturbances on the production site. Without the need for manual reconfiguration of all parameters, it automatically updates constraints and adaptively adjusts plans, thereby improving the emergency dispatching capabilities and robustness of production scheduling schemes.

[0008] As a preferred embodiment of the present invention, in step S7, multi-dimensional data such as order demand constraints, material availability status, and production resource availability are integrated. A production scheduling decision model is pre-trained based on historical production data. A multi-constraint production scheduling model is constructed with the objectives of minimizing delivery delays, maximizing equipment utilization, minimizing changeover costs, and minimizing work-in-process inventory. Global scheduling optimization is performed on multi-order, multi-process production tasks to generate a production task timeline plan that includes the start time, end time, equipment allocation, personnel allocation, and material occupancy relationships of each process.

[0009] This setting enables multi-objective collaborative optimization, taking into account delivery timeliness, equipment utilization, production costs, and inventory control. The production scheduling results are more in line with the actual production and operation needs of enterprises and are more executable.

[0010] An intelligent production scheduling optimization system for multi-order, multi-constraint scenarios, applicable to the aforementioned intelligent production scheduling optimization method for multi-order, multi-constraint scenarios, includes an instruction receiving and interaction module, a semantic parsing and constraint completion module, a multi-order clustering and grouping module, a finished goods inventory matching module, a material completeness verification module, a resource availability assessment module, an intelligent scheduling solution module, a dynamic disturbance module, a resource locking and control module, and a visualization output module.

[0011] As a preferred embodiment of the present invention, the instruction receiving and interaction module is used to receive production scheduling instructions in natural language or structured form from users, complete instruction input and interactive pop-up prompts, the semantic parsing and constraint completion module is used to preprocess the scheduling instructions, perform semantic parsing, extract key constraint information, and identify missing fields for interactive completion, and the multi-order clustering and grouping module is used to sort and cluster multiple orders and multiple SKUs according to delivery date, priority, process and resource sharing characteristics.

[0012] This setting supports lightweight natural language input, automatically fills in missing constraints, and organizes multiple orders into groups to reduce production changeover losses, laying a data foundation for subsequent intelligent production scheduling.

[0013] As a preferred embodiment of the present invention, the finished product inventory matching module is used to query the finished product SKU inventory, complete the inventory deliverability determination and split the production task quantity; the material completeness verification module is used to retrieve the BOM list, query the raw materials and in-transit material inventory, perform completeness verification and generate replenishment procurement suggestions; and the resource availability assessment module is used to analyze the status of equipment, personnel and mold resources, predict equipment load, changeover time and delay risk.

[0014] This setting allows for layer-by-layer verification and screening from three dimensions: finished goods inventory, material availability, and production resources. It helps identify risks such as material shortages and insufficient capacity in advance, avoids ineffective production scheduling, and improves the feasibility of plans.

[0015] As a preferred embodiment of the present invention, the intelligent scheduling solution module is used to construct a multi-constraint, multi-objective scheduling model, and complete the collaborative optimization solution of process tasks through the trained scheduling decision model to generate an initial scheduling plan. The dynamic disturbance module is used to monitor disturbance events on the production site in real time, triggering constraint parameter updates and partial or global rescheduling of the production plan. The resource locking and control module is used to lock the core resources of the scheduling plan, perform raw material pre-deduction, and update the corresponding "finished product" category inventory data after production is completed. The visualization output module is used to visualize the scheduling plan, resource load, and process sequence in the form of charts, Gantt charts, and timelines.

[0016] This setup enables integrated intelligent solution, dynamic optimization, resource management, and visualization, achieving a high degree of automation throughout the entire process and reducing the cost of manual intervention.

[0017] As a preferred embodiment of the present invention, the system is equipped with an AI Agent intelligent scheduling unit and a CoT logic recording unit. The AI ​​Agent intelligent scheduling unit is used to autonomously complete the entire process of demand analysis, constraint completion, production scheduling solution and disturbance response automated scheduling. The CoT logic recording unit is used to record the entire logical link of the system's production scheduling decision process, and retain the production scheduling basis for easy traceability and manual review.

[0018] This setting enables autonomous and intelligent scheduling of the entire production process, while retaining traces of decision-making logic, facilitating production management traceability, review, and compliance management.

[0019] Compared with the prior art, the beneficial effects of the present invention are as follows: By comprehensively considering various constraints such as order delivery dates, process sequences, equipment availability, personnel shifts and skills, material inventory, and delivery cycles, and responding to and rescheduling dynamic events such as order insertions, material shortages, equipment failures, and personnel absences, the system improves its adaptability to multi-order, multi-constraint scheduling scenarios. Through unified modeling and automatic solution of order tasks, process relationships, and resource status, it reduces reliance on repeated manual adjustments, shortens scheduling time, reduces resource conflicts, minimizes plan delays, and improves the efficiency and executability of scheduling plan generation. By supporting user interaction with scheduling requirements using natural language, it reduces reliance on fixed menu navigation and parameter configuration, improves human-computer interaction efficiency, lowers the barrier to entry, and enhances the system's ease of use and flexibility. Through collaborative optimization of equipment, personnel, and material resources, it supports parallel scheduling of multiple orders and staggered scheduling at the process level, effectively improving equipment utilization and order response timeliness. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the method logic flow provided in the embodiments of the present invention; Figure 2 This is a schematic diagram of the system framework structure provided in an embodiment of the present invention. Detailed Implementation

[0021] To further understand the invention's content, features, and effects, the following embodiments are provided, and detailed descriptions are given in conjunction with the accompanying drawings.

[0022] The structure of the present invention will now be described in detail with reference to the accompanying drawings.

[0023] refer to Figures 1 to 2 As shown in the figure, an intelligent production scheduling optimization method for multi-order, multi-constraint scenarios provided by an embodiment of the present invention includes the following steps: S1. Command Reception and Preprocessing: Receive production scheduling commands from users in natural language or structured form, perform redundancy removal, remove function words and invalid words from the commands, and obtain standardized command text. S2. Semantic parsing and constraint completion: Perform industrial semantic parsing on standardized instruction text to extract product model, production quantity, delivery date, priority, alternative material permission, emergency order permission and other implicit constraints. Automatically identify missing key fields and guide users to complete them through human-computer interaction to form a complete production scheduling request. S3. Multi-order clustering and grouping: For scenarios with multiple orders and multiple SKUs running in parallel, the order set is sorted or clustered and grouped based on delivery date, order priority, process similarity, shared materials and equipment, and changeover costs. S4. Finished Goods Inventory Matching and Judgment: Query the real-time inventory of the target SKU. If the inventory is sufficient, generate an inventory delivery plan directly. If the inventory is partially sufficient, split the gap into the quantity to be produced. If there is no inventory, directly enter the production scheduling process. S5. BOM Decomposition and Material Matching Verification: Retrieve the BOM structure list of the product to be produced, decompose the raw materials and parts details, and perform matching verification in combination with real-time inventory and materials in transit. If the matching is not complete, a procurement replenishment plan will be generated and production will be triggered after the materials arrive. S6. Production resource availability assessment: Analyze equipment operating status, maintenance window, personnel shift skills, mold occupancy status, and combine historical data to predict equipment load, changeover time, and order delay risk; S7. Multi-constraint scheduling model construction and intelligent solution: Integrate multi-dimensional constraints of orders, materials and resources, and pre-train a scheduling decision model based on historical production data to perform global scheduling optimization of multi-order and multi-process production tasks and generate a production task timeline plan. S8. Core Resource Lock-in and Inventory Pre-control: Lock in key equipment, skilled personnel and scarce materials in the production schedule, implement raw material inventory pre-deduction and lock, and update finished product, work-in-process and raw material inventory data synchronously after production is completed. S9. Visualized Production Scheduling Results Output: The optimized production plan, resource load, and process layout are visualized in the form of time axis, Gantt chart, and statistical charts, supporting plan viewing, export, and production deployment.

[0024] Specifically, it also includes a dynamic disturbance response step: used to monitor disturbance events on the production site in real time. The disturbance events include emergency order insertion, raw material shortage, sudden equipment failure, temporary staff absence, and delayed delivery of outsourced parts. When a disturbance event is detected, the constraint parameters are updated, the affected production tasks are partially rearranged or the global plan is reconstructed, and the adjusted optimized production scheduling plan is output.

[0025] By adopting the above solution, various abnormal disturbances on the production site can be detected in real time. Without the need for manual reconfiguration of all parameters, the system can automatically update constraints and adaptively adjust plans, thereby improving the emergency dispatching capability and the robustness of the production scheduling plan.

[0026] Specifically, in step S7, multi-dimensional data such as order demand constraints, material availability status, and production resource availability are integrated. A production scheduling decision model is pre-trained based on historical production data. A multi-constraint production scheduling model is constructed with the objectives of minimizing delivery delays, maximizing equipment utilization, minimizing changeover costs, and minimizing work-in-process inventory. Global scheduling optimization is performed on multi-order, multi-process production tasks, generating a production task timeline plan that includes the start time, end time, equipment allocation, personnel allocation, and material occupancy relationships of each process.

[0027] By adopting the above solution, multi-objective collaborative optimization can be achieved, taking into account delivery timeliness, equipment utilization, production costs and inventory control. The production scheduling results are more in line with the actual production and operation needs of enterprises and are more executable.

[0028] An intelligent production scheduling optimization system for multi-order, multi-constraint scenarios, applicable to the aforementioned intelligent production scheduling optimization method for multi-order, multi-constraint scenarios, includes an instruction receiving and interaction module, a semantic parsing and constraint completion module, a multi-order clustering and grouping module, a finished goods inventory matching module, a material completeness verification module, a resource availability assessment module, an intelligent scheduling solution module, a dynamic disturbance module, a resource locking and control module, and a visualization output module.

[0029] Specifically, the instruction receiving and interaction module is used to receive production scheduling instructions from users in natural language or structured form, complete instruction input and interactive pop-up prompts, the semantic parsing and constraint completion module is used to preprocess the scheduling instructions, perform semantic parsing, extract key constraint information, and identify missing fields for interactive completion, and the multi-order clustering and grouping module is used to sort and cluster multiple orders and multiple SKUs according to delivery date, priority, process and resource sharing characteristics.

[0030] The above solution supports lightweight natural language input requirements, automatically fills in missing constraints, and organizes and groups multiple orders to reduce production changeover losses, laying a data foundation for subsequent intelligent production scheduling.

[0031] Specifically, the finished goods inventory matching module is used to query the finished goods SKU inventory, complete the inventory deliverability determination and split the production task quantity; the material completeness verification module is used to retrieve the BOM list, query the raw materials and in-transit materials inventory, perform completeness verification and generate replenishment procurement suggestions; and the resource availability assessment module is used to analyze the status of equipment, personnel and mold resources, predict equipment load, changeover time and delay risk.

[0032] By adopting the above approach, we can verify and screen the inventory of finished products, the availability of materials, and production resources layer by layer to identify risks such as material shortages and insufficient production capacity in advance, avoid ineffective production scheduling, and improve the feasibility of the plan.

[0033] Specifically, the intelligent scheduling solution module is used to construct a multi-constraint, multi-objective scheduling model, and complete the collaborative optimization solution of process tasks through the trained scheduling decision model to generate an initial scheduling plan. The dynamic disturbance module is used to monitor disturbance events on the production site in real time, triggering constraint parameter updates and partial or global rescheduling of the production plan. The resource locking and control module is used to lock the core resources of the scheduling plan, perform raw material pre-deduction, and update the corresponding "finished product" category inventory data after production is completed. The visualization output module is used to visualize the scheduling plan, resource load, and process sequence in the form of charts, Gantt charts, and timelines.

[0034] The above solution integrates intelligent solution, dynamic optimization, resource management and visualization, achieving a high degree of automation throughout the process and reducing the cost of manual intervention.

[0035] Specifically, the system is equipped with an AI Agent intelligent scheduling unit and a CoT logic recording unit. The AI ​​Agent intelligent scheduling unit is used to autonomously complete the entire process of demand analysis, constraint completion, production scheduling solution and disturbance response automated scheduling. The CoT logic recording unit is used to record the entire logical link of the system's production scheduling decision process, and retain the production scheduling basis for easy traceability and manual review.

[0036] By adopting the above solution, autonomous and intelligent scheduling of the entire production process can be achieved, while retaining traces of decision-making logic to facilitate production management traceability, review, and compliance management.

[0037] This invention lowers the barrier to entry for system use through natural language interaction, automatically completing constraint completion and multi-order clustering. Combined with BOM breakdown, material completeness verification, and resource availability assessment, it proactively mitigates scheduling risks. Relying on a multi-objective scheduling model, it achieves collaborative optimization of order delivery, equipment utilization, cost control, and inventory management. Coupled with a dynamic disturbance response mechanism, it can quickly respond to various on-site anomalies and adjust plans accordingly. Core resource locking and pre-management of inventory ensure the implementation of scheduling plans. The AI ​​Agent intelligent scheduling unit achieves full-process automation, and the CoT logic recording unit ensures decision traceability. Overall, it significantly improves scheduling efficiency, resource utilization, and order delivery timeliness, while reducing labor costs, delay costs, and inventory costs, providing reliable technical support for flexible production in small and medium-sized manufacturing enterprises.

[0038] 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 process, method, article, or apparatus.

[0039] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A smart production scheduling optimization method for scenarios with multiple orders and multiple constraints, characterized in that, Includes the following steps: S1. Command Reception and Preprocessing: Receive production scheduling commands from users in natural language or structured form, perform redundancy removal, remove function words and invalid words from the commands, and obtain standardized command text. S2. Semantic parsing and constraint completion: Perform industrial semantic parsing on standardized instruction text to extract product model, production quantity, delivery date, priority, alternative material permission, emergency order permission and other implicit constraints. Automatically identify missing key fields and guide users to complete them through human-computer interaction to form a complete production scheduling request. S3. Multi-order clustering and grouping: For scenarios with multiple orders and multiple SKUs running in parallel, the order set is sorted or clustered and grouped based on delivery date, order priority, process similarity, shared materials and equipment, and changeover costs. S4. Finished Goods Inventory Matching and Judgment: Query the real-time inventory of the target SKU. If the inventory is sufficient, generate an inventory delivery plan directly. If the inventory is partially sufficient, split the gap into the quantity to be produced. If there is no inventory, directly enter the production scheduling process. S5. BOM Decomposition and Material Matching Verification: Retrieve the BOM structure list of the product to be produced, decompose the raw materials and parts details, and perform matching verification in combination with real-time inventory and materials in transit. If the matching is not complete, a procurement replenishment plan will be generated and production will be triggered after the materials arrive. S6. Production resource availability assessment: Analyze equipment operating status, maintenance window, personnel shift skills, mold occupancy status, and combine historical data to predict equipment load, changeover time, and order delay risk; S7. Multi-constraint scheduling model construction and intelligent solution: Integrate multi-dimensional constraints of orders, materials and resources, and pre-train a scheduling decision model based on historical production data to perform global scheduling optimization of multi-order and multi-process production tasks and generate a production task timeline plan. S8. Core Resource Lock-in and Inventory Pre-control: Lock in key equipment, skilled personnel and scarce materials in the production schedule, implement raw material inventory pre-deduction and lock, and update finished product, work-in-process and raw material inventory data synchronously after production is completed. S9. Visualized Production Scheduling Results Output: The optimized production plan, resource load, and process layout are visualized in the form of time axis, Gantt chart, and statistical charts, supporting plan viewing, export, and production deployment.

2. The intelligent production scheduling optimization method for multi-order, multi-constraint scenarios as described in claim 1, characterized in that, It also includes a dynamic disturbance response step: used to monitor disturbance events on the production site in real time. The disturbance events include emergency order insertion, raw material shortage, sudden equipment failure, temporary staff absence, and delayed delivery of outsourced parts. When a disturbance event is detected, the constraint parameters are updated, the affected production tasks are partially rearranged or the global plan is reconstructed, and the adjusted optimized production scheduling plan is output.

3. The intelligent production scheduling optimization method for multi-order, multi-constraint scenarios as described in claim 1, characterized in that, In step S7, multi-dimensional data such as order demand constraints, material availability status, and production resource availability are integrated. A production scheduling decision model is pre-trained based on historical production data. A multi-constraint production scheduling model is constructed with the objectives of minimizing delivery delays, maximizing equipment utilization, minimizing changeover costs, and minimizing work-in-process inventory. Global scheduling optimization is performed on multi-order, multi-process production tasks, generating a production task timeline plan that includes the start time, end time, equipment allocation, personnel allocation, and material occupancy relationships of each process.

4. An intelligent production scheduling optimization system for scenarios with multiple orders and multiple constraints, characterized in that, The intelligent production scheduling optimization method for multi-order, multi-constraint scenarios as described in any one of claims 1-3 includes an instruction receiving and interaction module, a semantic parsing and constraint completion module, a multi-order clustering and grouping module, a finished goods inventory matching module, a material completeness verification module, a resource availability assessment module, an intelligent scheduling solution module, a dynamic disturbance module, a resource locking and control module, and a visualization output module.

5. The intelligent production scheduling optimization method for multi-order, multi-constraint scenarios as described in claim 4, characterized in that, The instruction receiving and interaction module is used to receive production scheduling instructions from users in natural language or structured form, complete instruction input and interactive pop-up prompts. The semantic parsing and constraint completion module is used to preprocess the scheduling instructions, perform semantic parsing, extract key constraint information, and identify missing fields for interactive completion. The multi-order clustering and grouping module is used to sort and cluster multiple orders and multiple SKUs according to delivery date, priority, process and resource sharing characteristics.

6. The intelligent production scheduling optimization method for multi-order, multi-constraint scenarios as described in claim 4, characterized in that, The finished goods inventory matching module is used to query the finished goods SKU inventory, complete the inventory deliverability determination and split the production task quantity. The material completeness verification module is used to retrieve the BOM list, query the raw materials and in-transit materials inventory, perform completeness verification and generate replenishment procurement suggestions. The resource availability assessment module is used to analyze the status of equipment, personnel and mold resources, predict equipment load, changeover time and delay risk.

7. The intelligent production scheduling optimization method for multi-order, multi-constraint scenarios as described in claim 4, characterized in that, The intelligent scheduling solution module is used to construct a multi-constraint, multi-objective scheduling model. Through the trained scheduling decision model, it completes the collaborative optimization solution of process tasks and generates an initial scheduling plan. The dynamic disturbance module is used to monitor disturbance events on the production site in real time, triggering constraint parameter updates and partial or global rescheduling of the production plan. The resource locking and control module is used to lock the core resources of the scheduling plan, perform raw material pre-deduction, and update the corresponding "finished product" category inventory data after production is completed. The visualization output module is used to visualize the scheduling plan, resource load, and process sequence in the form of charts, Gantt charts, and timelines.

8. The intelligent production scheduling optimization method for multi-order, multi-constraint scenarios as described in claim 4, characterized in that, The system is equipped with an AI Agent intelligent scheduling unit and a CoT logic recording unit. The AI ​​Agent intelligent scheduling unit is used to autonomously complete the entire process of demand analysis, constraint completion, production scheduling solution and disturbance response, and the CoT logic recording unit is used to record the entire logical link of the system's production scheduling decision process, and retain the production scheduling basis for easy traceability and manual review.

Images