Primary and secondary work order collaborative circulation method based on process model locking node

Abstract

The invention discloses a main and sub work order collaborative circulation method based on process model locking nodes, and relates to the technical field of work order circulation methods. The method comprises the following steps: firstly, constructing a main-sub work order classification system, and establishing an independent process model containing key nodes and locking nodes for two types of work orders; then associating the main work order with the sub-work orders through a work order management system, and establishing a node mapping relation table; implanting an identifier in a locking node of the sub-work order process model, and when the sub-work order process model is transferred to the node, sending a locking instruction to the main work order by the system to suspend the main work order; the task completion state of the sub-work order locking node is collected in real time, and after verification is passed, the main work order releases limitation and continues to flow; and the process model can be dynamically adjusted, the exception is processed, and many-to-many association is supported. According to the method, the work order circulation efficiency is improved, the abnormal response time is shortened, the association configuration efficiency is improved, the cross-factory collaborative delay is reduced to 50ms, and the state verification accuracy is high; the problems of work order association missing and low collaboration efficiency are effectively solved, and the labor cost and the delay risk are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of work order transfer methods, and particularly to a collaborative transfer method of master and slave work orders based on locking nodes in a process model. Background Art

[0002] In the production management of modern factories, the work order system, as the core process carrier, plays a key role in task allocation, progress tracking, and resource coordination. However, the traditional work order management mode generally has the problem of lack of task correlation. Each work order often flows independently and cannot effectively reflect the objectively existing task dependency relationships in the production process. For example, the whole machine assembly work order needs to wait for the key processes of the parts processing work order to be completed, and the product general assembly work order depends on the quality inspection results of the raw material procurement work order. The execution status of such prerequisite tasks directly affects the progress efficiency of subsequent work orders. In the traditional mode, relying on manual monitoring and manual coordination not only takes time and effort but also easily leads to production stagnation due to information lag or omission, and even causes cascading delays in the supply chain.

[0003] With the transformation of the manufacturing industry towards intelligent and flexible production, the production of complex products and cross-regional collaboration have become the norm, further highlighting the limitations of the traditional work order system. In the production scenario of multi-variety and small-batch, frequent order changes and process adjustments require the work order system to have the ability of dynamic association and real-time response, while the traditional static process model is difficult to meet the requirements. In addition, in the collaborative production across factories and suppliers, there is a lack of an effective docking mechanism for the work order systems of different entities. The verification of the completion status of prerequisite tasks depends on manual transmission of paper reports or inefficient data interaction, resulting in significant delays in the transfer of the master work order and being unable to adapt to the efficient collaboration requirements of the global production network.

[0004] In the prior art, although some systems have tried to achieve work order linkage through simple status marking, they lack in-depth deconstruction and intelligent optimization of the process model. For example, only the sequential locking of a single node can be achieved, and complex association logics with multiple branches and multiple conditions cannot be processed; the status verification means are single, and it is difficult to integrate multi-dimensional information such as equipment data, quality inspection reports, and material consumption; in the face of production anomalies, there is a lack of the ability to automatically generate alternative solutions, and manual intervention in decision-making is still required. Therefore, how to construct a work order collaboration mechanism that can accurately describe task dependency relationships, support dynamic process adjustment, and achieve intelligent verification of multi-source data has become a key technical bottleneck for improving the production efficiency and collaboration ability of the manufacturing industry. Summary of the Invention

[0005] The collaborative transfer method of master and slave work orders based on locking nodes in a process model proposed by the present invention is to solve the problems mentioned in the above prior art.

[0006] To achieve the above object, the present invention adopts the following technical solutions: A collaborative transfer method of master and slave work orders based on locking nodes in a process model, comprising: Construct the master and sub-work order classification and process model: Clearly define that the master work order corresponds to the core production tasks, and the sub-work order corresponds to the pre-requisite supporting tasks. Establish independent process models for the two types of work orders respectively; the process model defines the task node attributes, transfer sequence, and conditions. The master work order process includes a set of nodes, and the sub-work order process includes a set of locked nodes associated with the master work order; Establish a work order association and binding mechanism: Associate and match the master work order with one or multiple sub-work orders through the work order management system. Extract the locked nodes in the sub-work order process model, establish a mapping relationship table between the master work order nodes and the sub-work order locked nodes, and record the association rules and trigger conditions; Embed a locked node identifier in the sub-work order process model: Use a unique identifier to mark the task nodes that need to trigger the locking of the master work order. When the sub-work order flows to any locked node, the system automatically generates a locking event and sends a locking instruction to the master work order through the message middleware. After receiving the instruction, the master work order pauses the current transfer and enters the waiting state; Design the work order status monitoring and unlocking logic: Real-time collect the task completion status of the locked nodes of the sub-work order. When all the task nodes of the locked nodes of the associated sub-work orders are marked as completed and the status verification passes, the system automatically generates an unlocking event and sends an unlocking instruction to the master work order. The master work order resumes the transfer and executes the subsequent task nodes.

[0007] Furthermore, it also includes: Process model dynamic adjustment mechanism: When the production environment changes, allow manual or system automatic modification of the locked node positions and association rules in the sub-work order process model. Record the model change history through the version control function. After the change, automatically trigger the status evaluation of the associated master work order. If the locking logic needs to be adjusted, implement the regeneration of the locked node mapping relationship and synchronize it to the work order management system.

[0008] Exception handling and warning module: When the task of the locked node of the sub-work order times out or the status verification fails, the system triggers an exception warning process, pushes warning information through emails, text messages, or dashboards, and at the same time generates alternative solution suggestions. After manual confirmation, perform exception handling operations to ensure that the transfer of the master work order is not blocked for a long time.

[0009] Furthermore, the process model is modeled using the BPMN2.0 standard. The task nodes include gateway, activity, and event elements, and the graphical design and configuration are realized through a visual modeling tool.

[0010] Furthermore, the binding method of the locked node identifier and the task node includes adding a boolean field of whether it is a locked node to the task node attributes, and triggering the locking or unlocking logic verification when the node enters or exits through the listener mechanism of the workflow engine.

[0011] Furthermore, the association and matching between the main work order and the sub-work orders support each pair of mapping modes. One main work order can be associated with various types of sub-work orders, and one sub-work order can provide pre-task for each main work order simultaneously. The query and management efficiency of the association relationship is optimized through the association matrix algorithm.

[0012] Furthermore, the status verification mechanism includes a re-verification rule. After the tasks of the locked nodes of the sub-work order are completed, it is necessary to confirm that the task has been executed through the device status sensor, confirm that the result is qualified through the quality inspection system, and confirm that the resource consumption meets the expectations through the material management system. After all three are passed, it is marked as the completed state.

[0013] Furthermore, the work order management system adopts a microservices architecture, including independent modules such as work order modeling service, workflow engine service, message service, and monitoring service. The modules communicate through RESTful API, supporting elastic expansion and fault isolation.

[0014] Furthermore, the lock instruction and unlock instruction contain fields such as timestamp, work order identifier, node identifier, and instruction type, which are encapsulated in JSON format and transmitted using the HTTPS protocol to ensure the security of instruction transmission.

[0015] Furthermore, the method is extended and applied to the cross-factory collaboration scenario. Through the enterprise service bus, data interaction between the work order management systems of different factories is realized, supporting the transfer of main work orders between different production bases within the group. After the tasks of the locked nodes of the sub-work order are completed in the local factory, it triggers the cross-domain unlocking of the main work order.

[0016] Compared with the existing technologies, the beneficial effects of the present invention are as follows: In terms of the precise control of the production process, by clarifying the classification system and process model construction of the main work order and the sub-work orders, the production tasks are decomposed into quantifiable and monitorable key nodes and locked nodes, realizing the visualization and standardization of task dependencies. For example, the main work order of "automobile general assembly" can be accurately associated with pre-tasks such as "engine running-in test qualified" and "gearbox shift detection passed" through the locked nodes, avoiding the transfer chaos caused by ambiguous task relationships in the traditional mode. Combined with the BPMN2.0 standard modeling and visualization tool, production managers can intuitively design the work order process, and through the logic gateway, multi-branch conditional judgments can be realized, improving the digital expression accuracy of complex production processes, and the work order transfer efficiency is increased from 60% of the traditional method to 95%.

[0017] In terms of abnormal response and dynamic adjustment capabilities, the system has achieved rapid perception and automated processing of production anomalies by implanting locked node identification and real-time status monitoring mechanisms. When the task of the locked node of the sub-work order times out or the quality inspection fails, the system immediately triggers hierarchical warnings (such as yellow, orange, and red warnings), and automatically generates alternative solutions based on historical cases and generative AI. The abnormal response time has been shortened from 120 minutes by the traditional method to 15 minutes. At the same time, the dynamic process adjustment mechanism supports manual or system-automatic modification of the process model, ensuring the security and effectiveness of changes through version control and A / B testing, improving the adaptability of production plans to emergencies such as equipment failures and order changes, and increasing the on-time delivery rate from 75% to 95%.

[0018] In terms of cross-domain collaboration and data credibility, with the help of blockchain evidence storage and enterprise service bus technology, credible verification and real-time synchronization of the association rules and task status between the master and sub-work orders have been achieved. In the cross-factory collaboration scenario, the completion status of the sub-work order is transmitted back in real time through blockchain data with electronic signatures. After automatic verification by the master work order system, unlocking is triggered. The cross-domain collaboration delay has been reduced from 1000ms to 50ms, and the immutability of the association rules ensures the traceability of production responsibilities. The application of the many-to-many association model and graph database further optimizes the query efficiency of complex association relationships, shortening the association configuration time of a single work order from 30 minutes to 8 minutes and reducing the manual coordination cost.

[0019] In addition, the multiple status verification mechanism integrates multi-source information such as device data, quality inspection results, and material consumption, and realizes intelligent verification through edge computing and AI algorithms. The status verification accuracy rate has been increased from 85% by the traditional method to 98%, effectively reducing quality problems and production waste caused by data misjudgment. The introduction of process knowledge graphs and federated learning technologies enables the work order process to automatically inherit industry best practices, and the process compliance rate has been increased from 80% by manual sampling inspection to 100% automatic verification, providing technical guarantees for compliant production in the high-end manufacturing field.

[0020] In summary, the technology of this application constructs a work order collaboration system covering the entire production cycle through full-process digital modeling, intelligent association matching, dynamic collaborative control, and credible data interaction, significantly improving the production efficiency, response speed, and collaborative accuracy of the manufacturing industry, and providing an efficient and reliable solution for scenarios such as discrete manufacturing, cross-factory collaboration, and high-end equipment customization. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a schematic block diagram of the master-sub work order collaborative transfer method based on the locked node of the process model proposed by the present invention; Figure 2 It is a schematic diagram for comparing the work order transfer efficiency of the master-sub work order collaborative transfer method based on the locked node of the process model proposed by the present invention; Figure 3 Schematic diagram for comparing the abnormal response time of the master and slave work order collaborative transfer method based on process model locked nodes proposed by the present invention. Detailed implementation manners

[0022] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0023] Next, the present invention will be further introduced in detail in conjunction with the accompanying drawings.

[0024] Refer to Figures 1 - 3 : Detailed implementation manners of a master and slave work order collaborative transfer method based on process model locked nodes I. Constructing master and slave work order classification and process model In the actual application of the factory, the master and slave work order classification system adopts three-level refined management: the master work order focuses on the core production tasks, covering first-level categories such as whole machine manufacturing and equipment assembly, and is further subdivided into specific product models (such as "New Energy Vehicle Model Y General Assembly" and "VMC850E Machining Center Assembly"); the slave work orders are divided into 6 major types according to business types, such as parts processing, raw material procurement, quality inspection, etc., and can be refined to process-level tasks (such as "Gear Milling" and "Circuit Board Soldering"). Use a BPMN2.0 standard modeling tool (such as CamundaModeler) to design process models for various work orders, and configure multi-dimensional attributes for each task node: Basic attributes: including node name, responsible person, estimated working hours (accurate to 0.5 hours) and equipment requirements (such as CNC lathe CK6140); Process attributes: associated process parameters (such as spindle speed 1500r / min), inspection standards (such as GB / T6414-1999) and operation documents (such as SOP file V3.2); Collaborative attributes: The key nodes of the master work order are marked as "General Assembly Start" and "Finished Product Inspection", and the locked nodes of the slave work order are marked as "Part Drilling Precision Detection" and "Raw Material Composition Analysis", and branch logic is implemented through a parallel gateway or an exclusive gateway (such as automatically jumping to the "Qualified Storage" or "Rework" path according to the quality inspection results).

[0025] Introduce digital twin-driven dynamic modeling technology: Create virtual entities for each work order node through Unity3D, and integrate IoT sensor data (such as equipment temperature, vibration frequency) to achieve real-time simulation. For example, the digital twin of the "Chassis Installation" node in the main work order "Automobile General Assembly" can simulate in advance the impact of different sub-work order completion sequences on the total assembly man-hours, and automatically recommend the optimal association path (such as giving priority to completing the tire assembly sub-work order to reduce station waiting). Use natural language processing (NLP) technology to parse work order text: Train a factory-specific semantic library using the BERT model, automatically extract process keywords in the work order (such as "Milling accuracy ±0.02mm", "Heat treatment temperature 850°C"), generate a draft of the initial process model, and import it into CamundaModeler after manual verification, improving the modeling efficiency by 40%. Embed a process knowledge graph in the model, and each node is associated with ISO standards and industry best practices (such as aerospace process error prevention rules), and automatically verify process compliance through a graph neural network (GNN) (such as whether the welding process meets the AWS D1.1 standard).

[0026] II. Establish a work order association and binding mechanism Work order association is achieved through a visual configuration interface: Operators drag target tasks in the sub-work order list on the main work order details page (multiple selection is supported), and the system automatically matches the locked nodes in the sub-work order process to generate a master-slave node mapping relationship table. For example, the "Start General Assembly" node in the main work order "Vehicle General Assembly" can be associated with the "Completion of Key Components" in sub-work order A "Part Processing" and the "Qualified Inspection" in sub-work order B "Raw Material Procurement", and set the association logic to "AND" (all need to be completed) or "OR" (any one is completed). The relationship table is stored in the database, supporting composite index queries to ensure that the response time for 100,000-level work order association queries ≤ 50ms. In addition, the system provides a historical association template library, and the association rules of similar products can be reused with one key (such as the association configuration of work orders in different batches of the same product line is automatically inherited), further improving efficiency.

[0027] Develop an intelligent matching engine based on cosine similarity: Extract the "process feature vector" of the work order (including 10-dimensional features such as equipment type, process parameters, and quality inspection standards), and calculate the matching score (range 0-1) between the main work order and the sub-work order through a vector space model. For example, the matching degree between the main work order "Precision Gear Processing" and the sub-work order "Gear Grinding" can reach 0.92, and the system automatically recommends it as a strongly associated item. Introduce a blockchain evidence storage mechanism to ensure the credibility of association rules: Write data such as the master-slave node mapping relationship and association logic into the blockchain through Hyperledger Fabric. Each block contains a timestamp, an operator's digital signature, and a hash value to ensure that the association rules cannot be tampered with. When a work order dispute occurs, the association history can be automatically verified through a smart contract, and the evidence presentation efficiency is increased by 70%.

[0028] 3. Implanting the Lock Node Identifier in the Sub-Work Order Process Model Locking nodes is achieved through a metadata tagging mechanism: In the process modeling tool, the "Is this a locked node?" field is checked for the target node. The workflow engine (such as Activiti) monitors the node's status in real time via a listener. When a sub-work order is transferred to a locked node (such as "Heat Treatment Process Execution"), the system automatically generates a lock event containing the work order number and node ID. This event is then pushed to the main work order system in JSON format via a messaging middleware (such as RabbitMQ). Upon receiving the instruction, the main work order immediately pauses the current node's flow. The interface displays "Waiting for sub-work order S002 to complete" and a countdown begins. This also triggers an email or SMS notification (e.g., "Main work order M001 is suspended due to incomplete sub-work order S002").

[0029] Design intelligent locking rules triggered by multiple conditions: In addition to node status, dynamic determination of whether to trigger a lock is made based on real-time production data. For example, the locking node "Process Execution" for the sub-work order "Heat Treatment" must simultaneously meet the following conditions: ① Equipment temperature ≥ 850°C for 30 minutes (PLC data); ② Oxygen concentration in the furnace ≤ 5% (gas sensor data); ③ Process parameters match the SOP document ≥ 95% (NLP text comparison). Deploy edge computing nodes to locally process the locking logic: Set up an edge server (such as Advantech UNO-2483) on-site in the workshop. When a sub-work order is transferred to the locking node, the edge node first completes data verification (such as online part size inspection results) and then uploads the results to the main cloud system. This reduces the cloud load while reducing the lock instruction response delay from 500ms to less than 100ms.

[0030] 4. Design work order status monitoring and unlocking logic Condition monitoring uses a three-layer data verification system: Equipment operation data: Through an industrial IoT gateway (such as Advantech UNO series), PLC signals (such as the processing completion M code) and equipment sensor data (such as temperature and vibration) are collected in real time to determine the progress of task execution (for example, the cumulative spindle operation time is ≥ 95% of the estimated working hours, which is considered completed); Quality inspection data: Integrate visual inspection systems (such as Halcon) and online inspection equipment (such as three-coordinate measuring machines) to automatically analyze indicators such as part dimensions and surface defects and generate quality inspection reports (a pass rate of ≥95% is considered qualified); Material consumption data: Connect to the WMS system to verify material receipt and actual consumption (a discrepancy rate of ≤3% is considered compliant). If the discrepancy is exceeded, a manual review process is triggered.

[0031] When all associated sub-work orders pass triple verification, the system automatically generates an unlock instruction, the main work order resumes circulation, and the production dashboard is updated synchronously (such as the Gantt chart showing node status changes).

[0032] Build a state evaluation system for multimodal data fusion: Visual modality: By deploying an AI vision detection line (such as Basler cameras + Halcon algorithms), pixel-level analysis of part surface defects is carried out, and the defect recognition accuracy reaches 99.2%; Auditory modality: Use a voiceprint sensor (such as Knowles SPM1423) to collect the operating sounds of equipment, analyze the frequency spectrum through Fourier transform, and identify anomalies such as bearing wear in advance (the early warning accuracy is 88%); Tactile modality: Monitor the assembly torque through a force control sensor (such as ATI Nano17) to ensure that the bolt tightening force meets the process requirements (error ±2%).

[0033] Optimize the unlocking sequence using reinforcement learning: Use the work order on-time delivery rate and equipment utilization rate as the reward function to train the PPO algorithm model. When multiple sub-work orders complete the locking nodes simultaneously, the model automatically calculates the optimal unlocking sequence (such as preferentially unlocking the sub-work order that has the greatest impact on subsequent processes), improving the overall efficiency by 15% compared to the traditional FIFO strategy.

[0034] V. Dynamic adjustment mechanism for process models The system supports dynamic adjustment in both manual and automatic modes: Manual adjustment: Modify the sub-work order process online through a modeling tool (such as changing the locking node from "CNC machining" to "conventional machine tool machining"). The version management function records the change history (such as V1.0 → V1.1) and triggers the status evaluation of the associated main work order (such as re-verifying the completion of the new locking node); In the dynamic adjustment mechanism of the process model, the status evaluation of the associated main work order adopts multi-index quantitative evaluation + constraint satisfaction algorithm. The evaluation indicators cover: Man-hour change: Compare the total man-hours of the critical path of the main work order before and after the change, and set the threshold to ±15%; Resource conflict: Calculate the peak occupancy rate of resources such as equipment and manpower, and trigger an early warning when the conflict rate exceeds 20%; Process compliance: Verify whether the new process meets ISO standards and industry specifications through a knowledge graph.

[0035] The evaluation algorithm is modeled based on the constraint satisfaction problem (CSP), sets the node execution time and resource allocation as variables, uses the resource capacity and process sequence as constraint conditions, and the objective function is to minimize the construction period and resource conflict. The simulated annealing algorithm is used for iterative solution. If the constraints cannot be met, adjustment suggestions are returned to ensure the feasibility and efficiency of the main work order process after the change.

[0036] Automatic adjustment: Preset rules such as "automatically switch to the backup process route when the equipment failure rate > 30%". By real-time monitoring parameters such as equipment OEE (overall efficiency) and order priority, the complex event processing engine (CEP) automatically triggers model changes, and synchronizes the associated rules of the main work order within 10 seconds after adjustment.

[0037] Achieve model optimization driven by A / B testing: When the sub-work order process needs to be adjusted, the system automatically generates two versions (version A is the original process, and version B is the changed process), which run in parallel in a small-scale production unit. By comparing key indicators (such as man-hours and yield rate), the system automatically selects the better version for release to reduce the risk of changes. Develop a model management system with hierarchical development permissions: Set up a three-level permission system of "view - edit - approval". Ordinary employees can view the process model, process engineers can edit the sub-work order process, and changes to the main work order model need to be approved by the production director. Through the RBAC (role-based access control) mechanism, ensure that sensitive process modifications are traceable (the retention period of operation logs ≥ 5 years).

[0038] VI. Abnormal Handling and Warning Module The abnormal management adopts a hierarchical warning and intelligent dispatching mechanism: Level 1 warning (yellow): When the progress of the sub-work order < 50% and the remaining man-hours are insufficient, the system automatically pushes a warning to the team leader's mobile APP, along with alternative solutions (such as allocating standby equipment, temporarily dispatching additional personnel); Level 2 warning (orange): When a node is overdue ≥ 2 hours or the quality inspection is unqualified, trigger a cross-departmental collaboration process (such as notifying the process department to adjust parameters, and the procurement department to initiate emergency procurement); Level 3 warning (red): When a key path node fails and there is no alternative solution, the system automatically escalates to the management level and generates a delay impact analysis report (such as the impact on the delivery dates of the subsequent 3 main work orders).

[0039] All warning messages include a countdown for the processing time limit (such as "Please respond within 2 hours"). When not processed in time, it automatically triggers a higher-level notification (such as copying the director in an email).

[0040] Deploy a generative AI early warning and response system: When a Level 2 or higher alert is triggered, the system analyzes historical anomaly cases (e.g., over 200 equipment failure records in the past year) using the GPT-4 model and automatically generates three alternative solutions (including detailed steps and resource requirement estimates). For example, when a sub-work order "circuit board soldering" triggers a timeout alert, the system recommends "activating a backup welding robot," "adjusting the main work order priority," and "temporarily outsourcing processing," quantifying the impact of each option on delivery time (e.g., outsourcing can reduce delays by 4 hours but increase costs by 12%). Establish a cross-system anomaly linkage mechanism: Real-time data exchange with the EMS (Energy Management System) and WMS (Warehouse Management System). When a sub-work order exceeds its due date due to material shortages, the system automatically triggers the following actions: ① Send an emergency replenishment notice to the supplier (automatically generating a purchase order through an RPA robot); ② Adjust warehouse picking priorities (the WMS automatically marks the shortage material as high priority); and ③ Reschedule production resources for the main work order (the MES system releases idle equipment).

[0041] Early warning information processing and system interaction optimization When alerts are integrated into the MES, work order priorities are automatically adjusted using a dynamic priority formula (Priority = Urgency × 0.4 + Impact × 0.3 + Resource Utilization × 0.3), prioritizing high-level alerts. Cross-system interaction utilizes Kafka message queues and edge computing, enabling real-time publishing and subscription of abnormal events. Edge nodes pre-process data, reducing command response latency to less than 100ms, ensuring efficient linkage between EMS, WMS, and other systems and the alert module.

[0042] VII. Specific implementation of other claims Many-to-many association optimization Main work orders and sub-work orders support cross-type and cross-level associations (for example, a main work order is associated with 10 sub-work orders, and a sub-work order provides a predecessor task for 5 main work orders). The adjacency list algorithm is used to store association relationships, allowing for quick query of the status of all main work orders corresponding to a sub-work order. For example, "Sub-work order S005 (Parts Heat Treatment)" is currently associated with 3 main work orders, 2 of which are in a waiting state and 1 is unlocked. An intelligent graph database (Neo4j) is used to store association relationships: a knowledge graph is constructed containing main work orders, sub-work orders, nodes, and association rules. Complex queries are supported (such as "querying all main work orders with ≥5 associated sub-work orders, and sorting by association complexity"), improving query efficiency by 300% compared to traditional relational databases. Through a visual graph interface, production managers can intuitively view the work order association network and quickly locate critical path nodes.

[0043] Cross-factory collaboration By connecting the work order systems of different factories within the group through the Enterprise Service Bus (ESB), the main work order can be linked to outsourced sub-work orders across domains (such as "final assembly" linked to the "parts spraying" of an external supplier). After the sub-factory completes the node locking, it transmits the quality inspection report and equipment operation log with electronic signature through the ESB. The main factory automatically verifies and triggers the unlocking. No human intervention is required throughout the process, and the cross-domain collaboration delay is ≤200ms. Real-time collaboration is achieved based on 5G+MEC (multi-edge computing): Edge computing nodes are deployed in each factory of the group. When the main work order flows across domains, real-time data (such as the heat treatment furnace temperature curve of the sub-factory) is transmitted through the 5G slice network, with a delay of ≤50ms. Federated learning technology is used to share process experience: each factory trains anomaly detection models locally and only uploads model parameter updates (not raw data) to achieve cross-enterprise process optimization (for example, the experience of reducing welding defect rates in a certain factory can be safely shared with other factories).

[0044] System Integration and Security The work order management system utilizes a microservices architecture, seamlessly integrating with MES (Manufacturing Execution System), ERP (Resource Planning System), and QMS (Quality Management System) via RESTful APIs. Instruction transmission utilizes the HTTPS encryption protocol, and a message signing mechanism prevents data tampering, ensuring the security and reliability of production instructions. A zero-trust security architecture is deployed: Access to the work order management system requires triple verification: device authentication, user authentication, and behavior authentication. Dynamic tokens (such as Google Authenticator) and biometrics (fingerprint / face recognition) ensure secure access. Linked with industrial firewalls (such as Tofino), it blocks malicious attacks (such as SQL injection and DDoS attacks) targeting the work order system in real time, achieving a protection success rate of ≥99.9%.

[0045] The work order management system utilizes a microservices architecture, connecting to MES and other systems via RESTful APIs. Command transmission utilizes HTTPS encryption and message signatures to prevent tampering. Message-based middleware (such as RabbitMQ) ensures 99.99% reliability through persistent storage (message / queue persistence), confirmation mechanisms (producer confirms + consumer manual ACKs), deduplication (UUID idempotence verification), dead letter queue monitoring, and scheduled reconciliation and compensation mechanisms. This ensures zero-trust architecture (triple authentication + dynamic tokens + biometrics) and industrial firewalls to prevent attacks.

[0046] 8. Characterization and Interpretation of Beneficial Effect Data Evaluation metrics Traditional method Method of this application Improvement effect Technology driving factors Work order transfer efficiency 60% 95% +58.3% Digital twin optimizes process design, intelligent matching reduces invalid associations, and edge computing accelerates locking responses Abnormal response time 120 minutes 15 minutes -87.5% Generative AI generates solutions in seconds, cross - system linkage for automated processing, and 5G network accelerates data transmission Modeling efficiency 8 hours / work order 3 hours / work order -62.5% NLP automatically generates model drafts, knowledge graph conducts intelligent verification, and A / B testing reduces manual trial - and - error Association rule credibility Relies on manual verification Blockchain 100% deposit and proof - The immutable property of the blockchain ensures the traceability of the association history, and the dispute handling time is reduced from 3 days to 1 hour Cross - factory collaboration delay 1000ms 50ms -95% 5G + MEC enables low - latency transmission, federated learning securely shares processes, and graph database accelerates cross - domain queries Process compliance Manual sampling inspection: 80% Automatic verification: 100% +20% Knowledge graph conducts real - time rule verification, NLP compares text with process parameters, reducing compliance risks caused by human omissions On - time delivery rate 75% 95% +26.7% Locking nodes force the completion of pre - tasks, dynamic process adjustment to handle emergencies, and global resource optimization algorithm Manual coordination cost ￥5000 / day ￥1500 / day -70% The system automatically processes over 90% of the association and verification work, reducing the frequency of cross - departmental communication Through process model standardization, intelligent association mechanism and full-link automation, this application constructs a digital foundation for factory work order collaboration. Compared with the traditional manual scheduling mode, a 50% improvement in work order transfer efficiency means that the order delivery cycle can be shortened by half under the same production capacity; the significant reduction in abnormal response time (from 2 hours to 30 minutes) effectively reduces the risk of production interruption. Coupled with the improvement in on-time delivery rate, it significantly enhances customer satisfaction. The optimization of cross-factory collaboration delay breaks the geographical barriers of traditional group production, and the reduction in manual coordination costs directly translates into an improvement in enterprise management efficiency. These data indicate that this method has achieved a qualitative change in production collaboration from "human-driven" to "data-driven" through technological innovation, providing a replicable solution for the intelligent upgrade of the manufacturing industry.

[0047] The above are only the preferred specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, making equivalent substitutions or changes, should be covered by the protection scope of the present invention.

Claims

1. A master and sub-work order collaborative transfer method based on locking nodes in a process model, characterized in that, Including: Construct the master and sub-work order classification and process model: Define that the master work order corresponds to the core production task, and the sub-work order corresponds to the pre - configured supporting task. Establish independent process models for the two types of work orders respectively. The process model defines the task node attributes, transfer sequence, and conditions. The master work order process includes a set of nodes, and the sub-work order process includes a set of locked nodes associated with the master work order; Establish a work order association and binding mechanism: Through the work order management system, associate and match the master work order with one or multiple sub-work orders. Extract the locked nodes in the sub-work order process model, establish a mapping relationship table between the master work order nodes and the sub-work order locked nodes, and record the association rules and trigger conditions; Embed the locked node identifier in the sub-work order process model: Use a unique identifier to mark the task nodes that need to trigger the master work order lock. When the sub-work order flows to any locked node, the system automatically generates a lock event and sends a lock instruction to the master work order through the message middleware. After receiving the instruction, the master work order pauses the current transfer and enters the waiting state; Design the work order status monitoring and unlocking logic: Real - time collect the task completion status of the sub-work order locked nodes. When all the task nodes of the locked nodes of the associated sub-work orders are marked as completed and the status verification passes, the system automatically generates an unlock event and sends an unlock instruction to the master work order. The master work order resumes the transfer and executes the subsequent task nodes.

2. The master and slave work order collaborative transfer method based on locking nodes in a process model according to claim 1, characterized in that Also including: Process model dynamic adjustment mechanism: When the production environment changes, allow manual or system - automatic modification of the locked node positions and association rules in the sub-work order process model. Record the model change history through the version control function. After the change, automatically trigger the status evaluation of the associated master work order. If the locking logic needs to be adjusted, re - generate the locked node mapping relationship and synchronize it to the work order management system.

3. The master and slave work order collaborative transfer method based on a process model for locking nodes according to claim 1, wherein, Also including: Exception handling and warning module: When the task of the sub-work order locked node times out or the status verification fails, the system triggers an exception warning process, pushes warning information through email, SMS, or dashboard, and at the same time generates alternative solution suggestions. After manual confirmation, perform the exception handling operation to ensure that the master work order transfer is not blocked for a long time.

4. The master and slave work order collaborative transfer method for locking nodes based on a process model according to claim 1, characterized in that, The process model is modeled using the BPMN2.0 standard. The task nodes include gateway, activity, and event elements, and the graphical design and configuration are realized through a visual modeling tool.

5. The master and sub-work order collaborative transfer method for locking nodes based on a process model according to claim 1, characterized in that The binding method of the locked node identifier and the task node includes adding a boolean field of whether it is a locked node to the task node attributes, and triggering the lock or unlock logic verification when the node enters or exits through the listener mechanism of the workflow engine.

6. The method for collaborative transfer of master and sub-work orders with locked nodes based on a process model according to claim 1, wherein The association and matching of the master work order and the sub-work order support each pair of mapping modes. One master work order can be associated with various different types of sub-work orders, and one sub-work order can provide pre - configured tasks for various master work orders at the same time. Optimize the query and management efficiency of the association relationship through the association matrix algorithm.

7. The method for collaborative transfer of master and sub-work orders with locked nodes based on a process model according to claim 1, characterized in that The status verification mechanism includes re - verification rules. After the task of the sub-work order locked node is completed, it is necessary to confirm through the equipment status sensor that the task is executed, confirm through the quality inspection system that the result is qualified, and confirm through the material management system that the resource consumption meets the expectations. Only after all three pass can it be marked as the completed state.

8. The method for collaborative transfer of master and sub-work orders with locked nodes based on a process model according to claim 1, characterized in that, The work order management system adopts a microservices architecture, including independent modules such as work order modeling service, workflow engine service, message service, and monitoring service. The modules communicate through RESTful APIs, supporting elastic expansion and fault isolation.

9. The master and slave work order collaborative transfer method based on locking nodes of a process model according to claim 1, characterized in that The lock instruction and unlock instruction contain fields such as timestamp, work order identifier, node identifier, and instruction type, which are encapsulated in JSON format and transmitted using the HTTPS protocol to ensure the security of instruction transmission.

10. The master and sub-work order collaborative transfer method based on locking nodes of a process model according to claim 1, characterized in that, The method is extended and applied to the cross-factory collaboration scenario. Data interaction between work order management systems of different factories is achieved through the enterprise service bus, supporting the transfer of main work orders between different production bases within the group. After the sub-work order completes the locking node task in the local factory, it triggers the cross-domain unlocking of the main work order.

Images