System and method for optimizing production process of manufacturing enterprise based on intelligent coding

Abstract

The invention provides a system and method for optimizing the production process of a manufacturing enterprise based on intelligent coding, and the system employs a static part-dynamic part-semantic relation coding structure to carry out the corresponding coding of materials, equipment, exchange units, departments, and personnel needed by the production of the enterprise. Real-time acquisition and updating of five types of object states are realized through an industrial sensor and edge calculation, and object state changes are directly reflected through dynamic coding; natural language processing is combined with a rule engine, a knowledge graph between coding objects is established based on a domain exclusive rule, a semantic network is formed, a production relation between the objects is established, the method acts on production planning, material purchasing, warehouse management, manufacturing management and logistics management links in enterprise production, and an enterprise is helped to formulate a scientific production plan. According to the invention, by adopting an innovative intelligent coding mode, various resources are seamlessly integrated with each module in the production cycle of an enterprise, so that decision making of the enterprise in production and operation is more accurate and reasonable.

Description

Technical Field

[0001] The present invention involves artificial intelligence and machine learning, the Internet of Things, blockchain, edge computing, digital twins, data coding and parsing technology, and is applied to the field of enterprise intelligent decision support. It is a system and method for optimizing the production process of manufacturing enterprises based on intelligent coding. Background Art

[0002] Faced with the growing scale and complexity of business, enterprise intelligent upgrading and transformation is imperative and urgent.

[0003] The existing system solutions for manufacturing enterprises have obvious deficiencies, which are particularly prominent in practical applications. First, the design of logistics information coding related to production module activities within the enterprise lacks unified specifications, and it is difficult to achieve seamless connection between the data of each module. The problem of information islands is serious, resulting in insufficient efficiency of data sharing and overall business collaboration; secondly, the coding system cannot dynamically reflect the changes in the state of objects in each link of production scheduling, and cannot effectively adapt to the dynamically changing business needs, resulting in a lag in response speed, and cannot meet the business needs of flexible, personalized, and flexible production orders; in addition, there is still a blank in the design of coding semantic relationships, and the coding cannot carry the deep association between business logic and objects. The linkage ability between modules is weak, and intelligent collaboration is almost impossible to achieve. In terms of data processing, most enterprise production and operation systems are still based on traditional parsing and storage methods, with limited support capabilities for real-time and complexity, and cannot meet the high requirements of modern enterprises for accuracy and timeliness. More importantly, the existing technical architecture lacks the use of cutting-edge technologies, resulting in limited intelligence of enterprises. These problems restrict the further development of enterprises in intelligent management.

[0004] Therefore, designing a system based on intelligent coding to optimize the production process of manufacturing enterprises has important application value and helps to improve the intelligence and production and operation efficiency of enterprises. Summary of the invention

[0005] The objective of the present invention is to propose a system and method for optimizing the production process of a manufacturing enterprise based on intelligent coding, so as to solve the problems raised in the above-mentioned background technology.

[0006] In one aspect, the present invention proposes a system for optimizing the production process of a manufacturing enterprise based on intelligent coding, comprising the following modules:

[0007] Coding module: Use the coding structure of static part-dynamic part-semantic relationship to encode the materials, equipment, business units, departments and personnel required for enterprise production;

[0008] Production planning module: forecast and match material demand, schedule production tasks, and allocate personnel tasks based on material, equipment, and personnel codes;

[0009] Material procurement module: complete the formulation of procurement demand plan, supplier selection, and use blockchain technology to record the entire process data information;

[0010] Warehouse management module: complete dynamic warehouse allocation, real-time inventory counting, and use blockchain technology to save inventory change records;

[0011] Manufacturing management module: Use artificial intelligence and IoT technologies to optimize the manufacturing process, and respond and handle production process anomalies and equipment anomalies in a timely manner;

[0012] Logistics management module: Optimize logistics scheduling and transportation routes through material coding to carry out intelligent logistics scheduling.

[0013] Furthermore, industrial sensor networks and tools are used to obtain the real-time status of the encoded objects, and edge computing gateways are deployed to transmit object status information back to achieve static and dynamic encoding of objects.

[0014] Furthermore, natural language processing (NLP) is combined with a rule engine, based on static and dynamic coding of objects, and machine learning algorithms are used based on domain-specific rules to establish a knowledge graph of relationships between objects, build semantic coding logic, and express the attributes of the object itself, dynamic status, and relationships between objects through the semantic coding structure, so that the coding has identification functions and supports intelligent interaction between systems.

[0015] Furthermore, the production plan predicts future material needs based on material codes, and automatically matches suitable equipment and production personnel when issuing tasks based on equipment codes and personnel codes.

[0016] Furthermore, the material procurement module uses material codes to dynamically generate stage procurement plans; collects suppliers' past supply data to generate suppliers' comprehensive credit ratings, and automatically selects the best supplier procurement combination when the company purchases from multiple parties.

[0017] Furthermore, each purchase order in the material procurement module is written into the blockchain through a smart contract, making the procurement data authentic and tamper-proof.

[0018] Furthermore, the warehouse management module maps the materials, equipment codes and inventory status to the digital twin platform to form a virtual mirror of the warehouse, uses unmanned vehicles to realize automated inventory, uses AGV robots to plan the optimal transportation route according to the material codes, and collaborates to complete complex sorting and distribution tasks.

[0019] Furthermore, the warehouse management module combines blockchain technology to record the entry, exit and usage of each batch of materials, achieving full-process monitoring.

[0020] Furthermore, the manufacturing management module updates the status in real time based on the equipment code and personnel code, reasonably arranges equipment replacement or maintenance, and dynamically adjusts the manufacturing task equipment and personnel allocation.

[0021] Furthermore, the logistics management module binds the material code with the logistics node, updates the transportation information in real time, adjusts the logistics priority based on the coding information, and ensures that urgent orders are given priority; using blockchain technology, the logistics process information is stored on the chain to control the status of each link in real time.

[0022] Furthermore, the above module includes the following functions and contents:

[0023] The coding module: All coding adopts the structure of static part-dynamic part-semantic relationship, and the coding objects include five categories: materials, equipment, trading units, departments, and personnel.

[0024] Specifically, the semantic relationship is used to describe the association and business logic between the above-mentioned coding objects, and the dynamic part includes the real-time status and function fields of the above-mentioned coding objects.

[0025] Specifically, the material code includes basic code (static part), dynamic code (dynamic part), and semantic layer code (semantic relationship). The basic code includes the material category, material type, material batch, and material unique identification number; specifically, the dynamic code includes material inventory, function field, real-time estimated production plan quantity of this type of material, and future procurement plan of this type of material; according to production and operation needs, specific materials are bound to material codes using RFID tags for material inventory and positioning, so as to realize intelligent scheduling of this type of material.

[0026] Specifically, the equipment coding includes basic coding, dynamic coding, and semantic layer coding. The basic coding includes basic equipment information, including category code, serial number code, and manufacturer code; specifically, the dynamic coding includes equipment operation status field, operation parameters, health level generated by AI based on operation data, special process identification that can be completed, and estimated production time of the current task, and dynamic parameter acquisition is completed using industrial sensors and AI big models.

[0027] Specifically, the business unit code includes basic code, dynamic code, and semantic layer code. The basic code includes basic information of the unit, including category and serial number; specifically, the dynamic code includes cooperation status and timestamp, and cooperation credit level generated based on past delivery cycle and supply capacity AI model.

[0028] Specifically, the department code includes basic code, dynamic code and semantic layer code. The basic code includes basic information of the department, including category and serial number; specifically, the dynamic code includes the current department's production task completion rate and the number of available production personnel.

[0029] Specifically, the personnel code includes basic code, dynamic code and semantic layer code. The basic code includes the personnel code, including work number, department, skill identification code and skill level identification; specifically, the dynamic code includes the current status identification of the personnel and the intelligent performance extension code, which is derived from the personnel's working hours and workload this month.

[0030] Preferably, the above-mentioned coding specific naming fields, such as equipment category code, business unit category serial number, department number and other fields are formulated with specific rules based on the specific production and operation environment within the enterprise.

[0031] Furthermore, based on the basic coding and dynamic coding of five types of objects, namely materials, equipment, business units, departments and personnel, semantic layer coding is designed to express the attributes, dynamic states and relationships between objects through semantic coding structures, so that the coding not only has identification functions, but also supports intelligent interaction between systems. Preferably, natural language processing (NLP) technology is used to extract relevant information in the daily production and operation of enterprises, and machine learning algorithms are used to analyze and predict the dynamic relationships between objects, converting them into coding structures that can be understood by machines, and using graph theory and matrix operation technology, by building a relationship network to reflect the multi-dimensional connections between different materials, equipment, personnel, departments, etc., generating relationship graphs and matrices to describe the dependency or collaboration relationship between objects.

[0032] The production planning module: The production planning is the core module of the manufacturing enterprise, which directly affects the material scheduling, equipment utilization and personnel allocation. The production planning module is deeply integrated through five types of object coding to realize a dynamic, efficient and intelligent production process.

[0033] Specifically, the production planning module predicts material demand through a deep learning algorithm based on the LSTM model, and matches the optimal production resources based on equipment and personnel coding.

[0034] Specifically, production planning needs to predict the types and quantities of required materials in advance to avoid insufficient or excessive inventory. Based on material coding, current inventory is counted, future procurement plans and logistics delivery information are predicted. RFID technology is used to monitor the material coding status, collect and analyze the life cycle fields in the materials, including the basic coding and dynamic coding of the materials, count the current inventory, use the deep learning LSTM model to predict future material demand, and combine coding to quickly respond to changes.

[0035] Specifically, the production plan should dynamically match the best equipment according to the production activities. Based on the operation status field and operation parameter field of the equipment code, as well as the equipment health level calculated based on the operation data, the special process identification that can be completed, and the estimated production time of the current task, the equipment priority schedule suitable for the current production task is generated, and the task schedule is further generated to reasonably arrange the production tasks and improve the production efficiency of the enterprise.

[0036] Preferably, a reinforcement learning algorithm is used to train the matching model, and the equipment scheduling strategy is optimized through continuous iteration so that each scheduling decision maximizes production efficiency.

[0037] Preferably, in order to avoid the impact of insufficient skills or task assignment errors on the production progress, the skill level field of the personnel code is combined with the production plan task requirements to automatically assign tasks using genetic algorithms to ensure efficient use of personnel resources. In the complex process flexible production mode, the system automatically assigns equipment with special process capabilities through equipment coding, and dispatches employees with specific skills to assist in production in combination with personnel coding.

[0038] Preferably, when urgent order production occurs, the material coding is used to quickly check whether the inventory is sufficient. If insufficient, the material procurement module is triggered, and the equipment coding is used to assign priorities in real time to ensure that key equipment is given priority in production. The personnel coding is called to arrange highly skilled personnel to handle complex processes.

[0039] Preferably, the production task-equipment-personnel matching is combined to generate semantic coding and construct relationship graph data. The coding clearly expresses resource configuration and task logic, which helps to achieve intelligent scheduling, optimal resource allocation and cross-module collaboration. Further training through machine learning models can greatly improve production efficiency and production intelligence.

[0040] The material procurement module: The material procurement module determines the timeliness and cost control of material supply. Through the deep integration of five types of coding, the transparency, intelligence and efficiency of the procurement process can be significantly improved.

[0041] Specifically, the procurement plan should respond to production plan requirements and inventory changes in real time, and use the type and batch fields of the material code and the inventory status field to dynamically generate stage procurement plans to meet production task requirements.

[0042] Optimally, when purchasing materials, different suppliers often have obvious differences in price, delivery speed, etc., so we collect the supplier's past supply data, use machine learning algorithms to build a selection review model, and generate a comprehensive supplier credit rating based on the supplier's supply price, time, and inventory. When the company needs to purchase raw materials from suppliers in multiple regions, the system automatically recommends the best purchase combination by coding the supplier's real-time inventory and delivery capacity to achieve the optimal balance between delivery time and cost. We routinely identify materials in urgent shortage through material coding, and quickly lock in suppliers with emergency supply capabilities in combination with the codes of the trading units to avoid emergency resource shortages.

[0043] Preferably, each purchase order is written into the blockchain through a smart contract, and the encoded records ensure full traceability, guaranteeing the authenticity and tamper-proofing of the purchase data.

[0044] The warehouse management module: Warehouse management is an important link between material supply and production consumption. Combined with five types of coding applications, the materials, equipment codes and inventory status are mapped to the digital twin platform to form a virtual mirror of the warehouse, realizing full automation and intelligence of warehouse management.

[0045] Specifically, dynamic storage location allocation is achieved through material coding, and the material category and function fields in the coding are combined with the storage location allocation optimization algorithm to update the storage location allocation:

[0046] Preferably, the material codes are bound by RFID tags, combined with the GIS (Geographic Information System), and unmanned vehicles are used to realize automated inventory and update the inventory system in real time; multiple AGV (automatic guided vehicle) robots are used to plan the optimal transportation route according to the material codes, and collaborate to complete complex sorting and distribution tasks.

[0047] Preferably, blockchain technology is used to record the storage, delivery and usage of each batch of materials to prevent data tampering and achieve full-process monitoring.

[0048] The manufacturing management module: The manufacturing management module is the core of achieving the company's production goals. Through the deep integration of five major types of codes, it can optimize the scheduling of production resources and process flow.

[0049] Specifically, the equipment status is updated in real time based on the equipment code, the health field is combined with the Bayesian network model to predict the possible failure time, equipment replacement or maintenance is reasonably arranged, and the manufacturing task allocation is dynamically adjusted to avoid production interruptions due to equipment failure.

[0050] Specifically, manufacturing tasks are accurately matched with resources, and the equipment operating status and personnel skill level fields in the code are used in combination with process requirements to complete a quick match, allowing production work to run efficiently.

[0051] The logistics management module: uses RFID and IoT technologies to bind material codes to logistics nodes, updates transportation information in real time, combines blockchain technology to store logistics process information on the chain, and controls the status of each link in real time; adjusts logistics priorities based on coding information to ensure that urgent orders are given priority; uses reinforcement learning algorithms, combines coding information and map data (high-precision GIS system), and dynamically updates the codes to intelligently decompose logistics tasks, optimize transportation routes and resource allocation, reduce transportation costs, reduce delays, and optimize warehouse utilization.

[0052] Preferably, the object information capture technology includes: a sensor network that uses RFID tags, NFC scanning, and infrared sensor technology to detect the status and location of materials and equipment. Each material is equipped with an RFID tag, and dynamic change information is obtained through access control scanning equipment when entering and leaving the warehouse. Industrial-grade sensors are installed on the equipment to collect operating parameters such as equipment operating time and temperature in real time. When employees enter and exit the production workshop using NFC cards, the system automatically updates their location and attendance status.

[0053] Preferably, an edge computing gateway is deployed to perform local processing on the collected coded data and reduce the transmission of redundant information to the central system.

[0054] Preferably, data transmission adopts MQTT protocol and LoRaWAN protocol, uses Apache Kafka tool to manage real-time data stream, packages each change data into an event message (including encoding, timestamp, change field), and passes the message stream directly to the parsing module.

[0055] Preferably, when parsing dynamic information, natural language processing (NLP) is used in combination with a rule engine to establish a knowledge graph of materials, equipment and other objects based on domain-specific rules. By comparing the state changes of old values ​​and new values, it is determined whether business logic operations are triggered, thereby forming a semantic coding logic.

[0056] In another aspect of the present invention, the present invention provides the following technical solution: a method for optimizing the production process of a manufacturing enterprise based on intelligent coding, the method comprising the following steps:

[0057] S1. Use the coding structure of static part-dynamic part-semantic relationship to encode materials, equipment, business units, departments and personnel respectively;

[0058] S2. Collect and integrate the coded data of the above five categories of objects to complete material demand forecasting and matching, production task scheduling, and personnel task allocation in the production plan;

[0059] S3. Collect and integrate the coded data of the above five categories of objects, complete the formulation of procurement requirements and supplier selection in material procurement, and use blockchain technology to record the entire process data information;

[0060] S4. Collect and integrate the coded data of the above five types of objects, complete dynamic storage location allocation and real-time inventory counting in warehouse management, and use blockchain technology to save inventory change records;

[0061] S5. Collect and integrate the coded data of the above five categories of objects, use artificial intelligence and Internet of Things technologies in manufacturing management to optimize the production process, and respond and handle production process anomalies and equipment anomalies in a timely manner;

[0062] S6. Collect and integrate the coded data of the above five categories of objects to realize intelligent logistics scheduling in logistics management.

[0063] According to the above content, the present invention has the following advantages and beneficial effects:

[0064] 1. The overall solution covers five major types of coding: materials, equipment, business units, departments and personnel. Through the three-layer structure design of static part, dynamic part and semantic relationship, the above five major types of resources are seamlessly integrated with the production planning, material procurement, warehouse management, manufacturing management and logistics management modules in the enterprise production cycle. This globally optimized coding system achieves a high degree of unification of internal resources, information and processes of the enterprise, eliminates the problem of information islands between modules, builds a collaborative management platform for the entire production cycle for the enterprise, and provides timely and effective resource information guarantee.

[0065] 2. The system uses dynamic coding technology to achieve real-time collection and update of the status of five major types of objects through IoT devices and edge computing, and directly reflects the changes in object status through dynamic coding. When the state of the object changes, the system will use the decision model to analyze and generate targeted optimization strategies. Such a dynamic response mechanism ensures that the company can quickly adjust each link in the production, and effectively improves the flexibility and adaptability of the company in dealing with complex production environments. Since all modules in the system support dynamic response, through the combination of dynamic coding and real-time data, it can improve the existing production system to better adapt to the flexible production needs of manufacturing companies.

[0066] 3. Using semantic hierarchical coding, the system realizes the semantic expression and linkage analysis of the relationship between various coded objects. By forming a semantic network through NLP and knowledge graph technology, the system can not only understand the complex relationship between multiple objects such as tasks, equipment, and personnel, but also provide semantic-driven intelligent support for resource allocation and process optimization. Compared with traditional systems, this semantic-based intelligent coding significantly improves the accuracy and rationality of decision-making in enterprise production and operation. Therefore, the coding module constructs semantic coding logic based on NLP technology, and uses knowledge graphs to describe the relationship between materials, equipment, personnel, trading units, and departments to achieve intelligent interaction.

[0067] 4. Through relationship graph data, task flow tracking and real-time status visualization, the system significantly improves the transparency of enterprise operations. Managers can grasp production tasks, equipment status, personnel allocation and logistics flow in real time, thereby achieving more accurate predictions and more scientific production planning.

[0068] 5. The solution integrates a variety of cutting-edge technologies, including the Internet of Things (IoT), edge computing, artificial intelligence decision-making models, knowledge graphs, and semantic analysis. This technology integration not only provides innovative solutions for the current system, but also leaves enough room for future expansion. Enterprises can further optimize or upgrade system functions according to their needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 It is a system structure block diagram of the present invention.

[0070] Figure 2 The figure is a flow chart of the method steps of the present invention. DETAILED DESCRIPTION

[0071] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0072] The present invention provides Figure 1 A system for optimizing the production process of a manufacturing enterprise based on intelligent coding is shown, including a coding module, a production planning module, a material procurement module, a warehouse management module, a manufacturing management module, and a logistics management module.

[0073] In this embodiment, further, the operation of other system modules such as the production planning module of the system depends on the coding module. The coding module uses a coding structure of static part-dynamic part-semantic relationship to perform corresponding encoding on five types of objects: materials, equipment, business units, departments, and personnel, and generates coded data to be called by other system modules. The data between system modules are interconnected and called by each other.

[0074] In this embodiment, the material code includes basic code (static part), dynamic code (dynamic part), and semantic layer code (semantic relationship). The basic code includes the material category (01 represents raw materials, 02 represents parts), the material type (A represents standard parts, B represents non-standard parts), the material batch (202402 represents the production batch of the material is February 2024), and the material unique identification number (such as 01A023). For example, the material basic code 01A023-02-A-202402 means that the independent number of the material is 01A023, which is a standard part, and the production batch is February 2024. In other words, the basic code of a certain material consists of the material unique identification number, material category, material type and material batch. Specifically, dynamic coding includes material inventory (inventory changes are monitored in real time by IoT devices), functional fields (i.e., purpose, from recent production product plans to material remark labels for quick selection and use), real-time estimated production plan quantity of this type of material (input data of each module is obtained by the prediction model), and future procurement plan of this type of material (input data of each module is obtained by the prediction model). For example, the dynamic material code 500-building support-200-300 means that the inventory quantity of this material is 500 copies, the purpose is building support, 200 copies are planned to be produced in the future, and 300 copies are planned to be purchased in the future; by adopting material coding that combines basic coding and dynamic coding, specific materials can be bound to material codes using RFID tags according to production and operation needs, which can be used for material inventory and positioning, thereby realizing intelligent scheduling of this type of materials.

[0075] In a specific embodiment, taking steel plate materials as an example, the material code of a batch of steel plates is 01A045-01-A-202403|800-Building template-500-200, and the code indicates that the material number of this batch of steel plates is 01A045, the category is raw materials, the type is standard parts, and the production batch is March 2024; the current inventory is 800 sheets, and the main function of this batch of steel plate materials is building templates. It is expected to produce 500 sheets in the future, and the procurement plan is 200 sheets.

[0076] The equipment coding includes basic coding, dynamic coding and semantic layer coding. The basic coding includes basic equipment information (category code, serial number code, manufacturer code). In this embodiment, the enterprise-defined basic code MT01E001-A represents a machine tool equipment (MT) with a production number of 01E001 produced by manufacturer A; the dynamic coding includes equipment operation status fields (WORK, MAINT, REST), operation parameters (temperature, load, continuous working time), health level generated by AI based on operation data, special process identification that can be completed, and the estimated production time of the current task. The dynamic parameter acquisition is completed using industrial sensors and AI big models.

[0077] In a specific embodiment, taking a machine tool device as an example, a device code is MT01E001-A|WORK-75-85-4-90-5AXIS-8, which indicates that this is a machine tool device, numbered 01E001, and manufactured by A; the current status of the device is in operation, the temperature is 75°C, the load is 85%, it has been working continuously for 4 hours, the health level is 90, it supports five-axis machining technology, and the current task is expected to take 8 hours to complete.

[0078] The business unit code includes basic code, dynamic code and semantic layer code. The basic code includes basic unit information (category, serial number); the dynamic code includes cooperation status and timestamp, and cooperation credit level generated based on past delivery cycle and supply capacity AI model.

[0079] In a specific embodiment, taking a certain cooperative unit as an example, the unit code is SUP0015|ACTIVE-2024-12-16-85, which indicates that this is a trading unit with a category of supplier and a serial number of 0015; the supplier is currently in active status, and the status was last updated on December 16, 2024. AI evaluates its cooperative credibility level as 85.

[0080] The department code includes basic code, dynamic code and semantic layer code. The basic code includes basic information of the department (category, serial number); the dynamic code includes the completion rate of the current department's production tasks and the number of available production personnel.

[0081] In a specific embodiment, taking a certain department as an example, the department code is PRD020|80-15, which indicates that this is a production department, and the serial number is 020; the current task completion rate of the department is 80%, and there are currently 15 production personnel available.

[0082] The personnel code includes basic code, dynamic code and semantic layer code. The basic code includes the personnel code (work number, department), skill identification code and skill level identification; specifically, the dynamic code includes the current status identification of the personnel (WORK, REST, HOLS) and the intelligent performance extension code, which is derived from the personnel's working hours and workload this month.

[0083] In a specific embodiment, taking a certain person as an example, the person code is E1005-PRD020-WELD-LV3|WORK-300-1200, and the code indicates that the employee with the work number E1005 belongs to the production department PRD020, has welding skills, and the level is 3; the person is currently working, with a cumulative working time of 300 hours this month, and a comprehensive workload score of 1200.

[0084] What needs to be specifically explained in this embodiment is that the specific naming fields of the above codes, such as equipment category code, business unit category serial number, department number and other fields, can be formulated according to specific rules based on the specific production and operation environment within the enterprise.

[0085] In this embodiment, further, based on the basic coding and dynamic coding of five types of objects, namely materials, equipment, business units, departments and personnel, semantic layer coding is designed to express the attributes, dynamic states and relationships between objects through semantic coding structures, so that the coding not only has identification functions, but also supports intelligent interaction between systems. Natural language processing (NLP) technology is used to extract relevant information in the daily production and operation of enterprises, and machine learning algorithms are used to analyze and predict the dynamic relationships between objects, converting them into coding structures that can be understood by machines. Graph Theory and matrix operation technology are used to build a relationship network to reflect the multi-dimensional connections between different materials, equipment, personnel, departments, etc., and generate relationship diagrams and matrices to describe the dependency or collaboration relationship between objects.

[0086] In specific embodiment one, taking a batch of steel materials as an example, the semantic code of this batch of steel materials is generated by collecting data equipment and collection methods: building template-X203-D3021-S001-D01 / W02. The semantic code indicates that the function of this batch of steel plates is for building templates, the task number is X203, and it needs to be processed by cutting machine D3021. The procurement priority supplier is S001, and the management departments are manufacturing department D01 and warehousing department W02. The basic code and dynamic code are superimposed to form a complete coding structure 01A045-01-A-202403|800-building template-500-200|building template-X203-D3021-S001-D01 / W02. This complete coding structure is bound to the material coding through RFID to achieve rapid positioning and inventory of steel plate stocks; based on the association between dynamic coding and semantic layer, it intelligently matches task requirements, equipment capabilities and procurement plans to optimize resource utilization; it uses dynamic data to predict future demand, accurately formulate procurement plans, and avoid excessive inventory or shortages.

[0087] In the second specific embodiment, taking the semantic coding of a certain person as an example: Senior Welding Technician-WELD-TASK-004-P1, it means that the person is a senior welding technician, currently responsible for welding task 004, and the task priority is the highest (P1). The basic coding and dynamic coding are superimposed to form a complete coding structure E1005-PRD020-WELD-LV3|WORK-300-1200|Senior Welding Technician-WELD-TASK-004-P1. This complete coding structure automatically prioritizes welding-related high-priority tasks (WELD-TASK-004) in the production planning module based on the basic skills (welding skill LV3) and dynamic performance (this month's working hours, comprehensive scores); the intelligent performance extension code (300-1200) in the dynamic coding provides the HR system with detailed assessment basis for identifying efficient employees and optimizing human resource allocation; the functional description and current task information in the semantic layer coding enable the task dispatching module to quickly locate suitable employees (such as senior welding technicians) and monitor task progress in real time; the semantic layer coding clarifies employee tasks and priorities (P1), and the system can prioritize the allocation of key resources during cross-departmental collaboration to ensure that tasks are completed quickly.

[0088] In the third specific embodiment, a certain production task requires the use of machine tool equipment MTB1E081, and the required materials include standard parts 01A02 and tools and fixtures 82801. The production requires support from the production department PR0101, and personnel F101001 is responsible for the operation. The semantic coding structure is shown in Table 1. A large number of relationship tables are maintained in the system to construct a relationship network, and relationship diagrams and matrices are generated to describe the dependency or collaboration relationship between objects.

[0089] Table 1. Semantic encoding structure

[0090] Object coding equipment MT01E001 Supplies 1 01A02-01-02-20231125 Supplies 2 02B01-01-01-20231125 department PR0101 personnel F101001

[0091] In this embodiment, it is necessary to further explain that the coding changes of the object are captured through data collection, and the data collection scope includes:

[0092] Material code:

[0093] Static information captured: material category (parts, raw materials, tools), type, origin, batch number; dynamic information captured: current status (in warehouse, out of warehouse, in production), inventory quantity, circulation location, and usage time.

[0094] Device Code:

[0095] Static information captured: equipment model, manufacturer, equipment category, maintenance cycle; dynamic information captured: operating status, working hours, key parameters.

[0096] Contact unit code:

[0097] Static information captured: company name, region, contact information, business type; dynamic information captured: current transaction status (cooperating, not cooperating), credit rating.

[0098] Department Code:

[0099] Static information captured: department name, scope of responsibilities, location; dynamic information captured: current production task completion rate, number of available personnel.

[0100] Personnel code:

[0101] Static information captured: employee identity (worker, technician, administrator), skill category; dynamic information captured: attendance status, work assignment, current production tasks.

[0102] What needs to be further explained in this embodiment is that the information capture technology includes: a sensor network, which uses RFID tags, NFC scanning, and infrared sensor technology to detect the status and location of materials and equipment. Each material is equipped with an RFID tag, and dynamic change information is obtained through access control scanning equipment when entering and leaving the warehouse. Industrial-grade sensors are installed on the equipment to collect operating parameters such as equipment operating time and temperature in real time. When employees enter and exit the production workshop using NFC cards, the system automatically updates their location and attendance status.

[0103] Deploy edge computing gateways to locally process the collected coded data and reduce the transmission of redundant information to the central system.

[0104] Data transmission adopts MQTT protocol and LoRaWAN protocol, and uses Apache Kafka tool to manage real-time data stream, package each change data into an event message (including encoding, timestamp, change field), and pass the message stream directly to the parsing module.

[0105] When parsing dynamic information, we use natural language processing (NLP) combined with a rule engine to build a knowledge graph of materials, equipment and other objects based on domain-specific rules. By comparing the state changes of old and new values, we can determine the business logic operations that are triggered, thereby forming a semantic coding logic.

[0106] In this embodiment, the production planning module, the production planning is the core module of the manufacturing enterprise, which directly affects the material scheduling, equipment utilization and personnel allocation. The production planning module is deeply integrated through five types of object coding to achieve a dynamic, efficient and intelligent production process.

[0107] Production planning requires predicting the types and quantities of required materials in advance to avoid insufficient or excessive inventory. Based on material coding, statistics are collected on current inventory, future procurement plans, and logistics delivery information are predicted. RFID technology is used to monitor the material coding status, collect and analyze the life cycle fields in the materials, including the basic coding and dynamic coding of the materials, and count the current inventory. The deep learning LSTM model is used to predict future material demand, and the coding is combined to quickly respond to changes.

[0108]

[0109] P plan (i) represents the planned demand, S inventory (i) represents the current inventory, and M demand represents the material procurement demand plan (demand).

[0110] The production plan should dynamically match the best equipment according to the production activities. Based on the operation status fields (such as WORK, MAINT) and operation parameter fields (such as temperature, load) of the equipment code, as well as the equipment health level, special process identification that can be completed, and the estimated production time of the current task calculated based on the operation data, the equipment priority schedule suitable for the current production task is generated, and the task schedule is further generated to reasonably arrange the production tasks to improve the production efficiency of the enterprise.

[0111] Preferably, a reinforcement learning algorithm is used to train the matching model, and the equipment scheduling strategy is optimized through continuous iteration so that each scheduling decision maximizes production efficiency.

[0112] In order to prevent the production progress from being affected by insufficient skills or mistakes in task assignment, the skill level field of the personnel code is combined with the production plan task requirements to automatically assign tasks using genetic algorithms to ensure efficient use of personnel resources. In the complex process flexible production mode, the system automatically assigns equipment with special process capabilities through equipment coding, and combines personnel coding to dispatch employees with specific skills to collaborate in production.

[0113] A task = argmax (Q skill matching · P availability);

[0114] Q 技能匹配 : The matching degree between the skills required for the task and the skills of the personnel, P 可用性 :Current personnel idle status, A 任务 Indicates whether tasks can be matched with human resources that meet skill requirements.

[0115] Preferably, when urgent order production occurs, the material coding is used to quickly check whether the inventory is sufficient. If insufficient, the material procurement module is triggered, and the equipment coding is used to assign priorities in real time to ensure that key equipment is given priority in production. The personnel coding is called to arrange highly skilled personnel to handle complex processes.

[0116] Preferably, the production task-equipment-personnel matching is combined to generate semantic coding and construct relationship graph data. The coding clearly expresses resource configuration and task logic, which helps to achieve intelligent scheduling, optimal resource allocation and cross-module collaboration. Further training through machine learning models can greatly improve production efficiency and production intelligence.

[0117] In this embodiment, the material procurement module: The material procurement module determines the timeliness and cost control of material supply, and through the deep integration of five types of codes, the transparency, intelligence and efficiency of the procurement process can be significantly improved.

[0118] The procurement plan must respond to production plan requirements and inventory changes in real time, and use the material coding type and batch fields and inventory status fields to dynamically generate stage procurement plans to meet production task requirements.

[0119] Optimally, when purchasing materials, different suppliers often have obvious differences in price, delivery speed, etc., so we collect the supplier's past supply data, use machine learning algorithms to build a selection review model, and generate a comprehensive supplier credit rating based on the supplier's supply price, time, and inventory. When the company needs to purchase raw materials from suppliers in multiple regions, the system automatically recommends the best purchase combination by coding the supplier's real-time inventory and delivery capacity to achieve the optimal balance between delivery time and cost. We routinely identify materials in urgent shortage through material coding, and quickly lock in suppliers with emergency supply capabilities in combination with the codes of the trading units to avoid emergency resource shortages.

[0120]

[0121] C 价格 : Unit material cost; W 信誉 : Supplier reputation score; T 交付 : Delivery cycle; S 优选 It indicates the optimal material procurement strategy determined by the system after decision-making.

[0122] Preferably, each purchase order is written into the blockchain through a smart contract, and the encoded records ensure full traceability, guaranteeing the authenticity and tamper-proofing of the purchase data.

[0123] In this embodiment, the warehouse management module: warehouse management is an important link between material supply and production consumption. Combined with five types of coding applications, the materials, equipment codes and inventory status are mapped to the digital twin platform to form a virtual mirror of the warehouse, realizing full automation and intelligence of warehouse management.

[0124] Dynamic storage location allocation is achieved through material coding. The material category and function fields in the coding are combined with the storage location allocation optimization algorithm to update the storage location allocation:

[0125] L allocation = argmin(T access(l) + D transport(l));

[0126] T 存取 : The time required to take out materials from the current storage location; D 运输 : Transportation time from storage location to production line; L 分配 Indicates the storage location allocation plan of materials in the warehouse.

[0127] Preferably, the material codes are bound by RFID tags, combined with the GIS (Geographic Information System), and unmanned vehicles are used to realize automated inventory and update the inventory system in real time; multiple AGV (automatic guided vehicle) robots are used to plan the optimal transportation route according to the material codes, and collaborate to complete complex sorting and distribution tasks.

[0128] Preferably, blockchain technology is used to record the storage, delivery and usage of each batch of materials to prevent data tampering and achieve full-process monitoring.

[0129] In this embodiment, the manufacturing management module: The manufacturing management module is the core of achieving the production goals of the enterprise. Through the deep integration of five major types of codes, the scheduling and process flow of production resources can be optimized.

[0130] The device status is updated in real time based on the device code. The health field is combined with the Bayesian network model to predict the possible failure time, reasonably arrange equipment replacement or maintenance, dynamically adjust manufacturing task allocation, and avoid production interruptions due to equipment failure.

[0131] Accurately match manufacturing tasks with resources, use the equipment operating status and personnel skill level fields in the code in combination with process requirements to complete quick matching, and make production work run efficiently.

[0132] In this embodiment, the logistics management module: uses RFID and IoT technology to bind material codes to logistics nodes, updates transportation information in real time, combines blockchain technology to store logistics process information on the chain, and controls the status of each link in real time; adjusts logistics priorities based on coding information to ensure that urgent orders are given priority; uses reinforcement learning algorithms, combines coding information and map data (high-precision GIS system), and dynamically updates the codes to intelligently decompose logistics tasks, optimize transportation routes and resource allocation, reduce transportation costs, reduce delays, and optimize warehouse utilization.

[0133] In this embodiment, Figure 2 A method for optimizing the production process of a manufacturing enterprise based on intelligent coding includes the following steps:

[0134] S1. Use the coding structure of static part-dynamic part-semantic relationship to encode materials, equipment, business units, departments and personnel respectively;

[0135] S2. Collect and integrate the coded data of the above five categories of objects to complete material demand forecasting and matching, production task scheduling, and personnel task allocation in the production plan;

[0136] S3. Collect and integrate the coded data of the above five categories of objects, complete the formulation of procurement requirements and supplier selection in material procurement, and use blockchain technology to record the entire process data information;

[0137] S4. Collect and integrate the coded data of the above five types of objects, complete dynamic storage location allocation and real-time inventory counting in warehouse management, and use blockchain technology to save inventory change records;

[0138] S5. Collect and integrate the coded data of the above five categories of objects, use artificial intelligence and Internet of Things technologies in manufacturing management to optimize the production process, and respond and handle production process anomalies and equipment anomalies in a timely manner;

[0139] S6. Collect and integrate the coded data of the above five categories of objects to realize intelligent logistics scheduling in logistics management.

[0140] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A system for optimizing the production process of a manufacturing enterprise based on intelligent coding, characterized in that: Includes the following modules: Coding module: Use the coding structure of static part-dynamic part-semantic relationship to encode the materials, equipment, business units, departments and personnel required for enterprise production; Production planning module: forecast and match material demand, schedule production tasks, and allocate personnel tasks based on material, equipment, and personnel codes; Material procurement module: complete the formulation of procurement demand plan, supplier selection, and record the whole process data information with blockchain technology; Warehouse management module: complete dynamic storage allocation, real-time inventory counting, and use blockchain technology to save inventory change records; Manufacturing management module: manufacturing process optimization, production process anomalies and equipment anomaly response and processing; Logistics management module: Optimize logistics scheduling and transportation routes through material coding to carry out intelligent logistics scheduling.

2. The system according to claim 1, characterized in that The encoding module is used to perform the following corresponding encoding: Material code: basic code, i.e. static part, includes material category, type, batch, and unique identification number; dynamic code, i.e. dynamic part, includes material inventory, function field, real-time estimated production plan number of this type of material, and future purchase plan number of this type of material; Equipment coding: basic coding includes basic equipment information, including category code, serial number code, and manufacturer code; dynamic coding includes equipment operation status field, operation parameters, health level generated by AI based on operation data, special process identification that can be completed, and estimated production time of the current task; The basic code includes basic information of the unit, including category and serial number; the dynamic code includes cooperation status and timestamp, and cooperation credit rating generated by AI model based on past delivery cycle and supply capacity; Department code: basic code includes basic department information, including category and serial number; dynamic code includes the current department's production task completion rate and the number of available production personnel; Personnel coding, basic coding includes personnel coding, including work number, department, skill identification code and skill level identification; dynamic coding includes the personnel's current status identification and intelligent performance extension code. The intelligent performance extension code is derived from the personnel's working hours and workload this month.

3. The system according to claim 1, characterized in that The coding module and the semantic relationship coding are based on the basic coding and dynamic coding of five types of objects: materials, equipment, business units, departments, and personnel. The semantic layer coding is designed to express the attributes, dynamic states, and relationships between objects through the semantic coding structure, so that the coding has an identification function and supports intelligent interaction between systems. Use natural language processing (NLP) technology to extract relevant information from the company's daily production and operation operations, use machine learning algorithms to analyze and predict the dynamic relationships between objects, and convert them into coding structures that can be understood by machines. Utilize graph theory and matrix operation technology to build a relationship network to reflect the multi-dimensional connections between different materials, equipment, personnel, departments, etc., and generate relationship diagrams and matrices to describe the dependency or collaboration relationship between objects.

4. The system according to claim 1, characterized in that Technologies for capturing information dynamically encoded by objects include: industrial sensor networks, RFID tags, NFC scanning, and edge computing gateways; Use RFID tags, NFC scanning, and infrared sensor technology to detect the status and location of materials and equipment. Each material is equipped with an RFID tag. The dynamic change information is obtained through the access control scanning device when entering and leaving the warehouse. The equipment is equipped with industrial-grade sensors to collect real-time operating parameters such as equipment operating time and temperature. When employees enter and exit the production workshop with NFC cards, the system automatically updates their location and attendance status; Deploy edge computing gateways to localize the collected coded data and reduce the transmission of redundant information to the central system; Data transmission uses MQTT protocol and LoRaWAN protocol, and uses Apache Kafka tool to manage real-time data streams. Each change data is packaged into an event message, including code, timestamp, and change field, and the message stream is directly passed to the parsing module; Based on the generation of dynamic coded information analysis and processing, natural language processing (NLP) is combined with a rule engine to establish a knowledge graph of materials, equipment and other objects based on domain-specific rules. By comparing the state changes of old and new values, it is determined whether business logic operations are triggered, thereby forming a semantic coding logic.

5. The system according to claim 1, wherein: The production planning module: Use RFID technology to monitor the material coding status, collect and analyze the life cycle fields in the materials, count the current inventory, use the deep learning LSTM model to predict future material demand, and combine the coding to reflect the changes; Based on the basic equipment code and dynamic code, generate the equipment priority schedule for the current production task, and continue to generate the task schedule, based on which the production task is scheduled; In the flexible production mode of complex processes, the system automatically allocates equipment with special process capabilities through equipment coding, and combines it with personnel coding to dispatch employees with specific skills to assist in production.

6. The system according to claim 1, characterized in that The material procurement module: Use material coding type, batch field and inventory status field to dynamically generate stage procurement plans to meet production task requirements; The supplier's comprehensive credit rating is generated by integrating the supplier's supply price, time, and inventory. When an enterprise purchases raw materials from suppliers in multiple regions, the system automatically recommends the best purchase combination by coding the supplier's real-time inventory and delivery capacity. The system also identifies urgently needed materials through material codes, and combines the codes of the business units to lock in suppliers with emergency supply capabilities for supply. Each purchase order is written into the blockchain via a smart contract.

7. The system according to claim 1, characterized in that The warehouse management module: Map the codes with materials, equipment and inventory status to the digital twin platform to form a virtual image of the warehouse, realize automated warehouse management, and achieve dynamic storage location allocation through material coding. The material category and function fields in the code are combined with the storage location allocation optimization algorithm to update the storage location allocation; By binding material codes with RFID tags and combining them with the GIS geographic information system, multiple AGVs are used to plan the optimal transportation route based on the material codes and collaborate to complete complex sorting and distribution tasks. Combined with blockchain technology, the storage, delivery and usage of each batch of materials are recorded.

8. The system of claim 1, wherein: The manufacturing management module: Update the equipment status in real time based on the equipment code, and use the health field combined with the Bayesian network model to predict possible failure time, arrange equipment replacement or repair, and dynamically adjust manufacturing task allocation; Use the equipment operating status and personnel skill level fields in the code in combination with process requirements to complete the matching and achieve a match between manufacturing tasks and resources.

9. The system according to claim 1, characterized in that The logistics management module: Utilize RFID and IoT technologies to bind material codes to logistics nodes, update transportation information in real time, and combine blockchain technology to store logistics process information on the chain; Adjust logistics priorities based on coded information, use reinforcement learning algorithms, combine coded information and high-precision GIS map systems, dynamically update codes, intelligently decompose logistics tasks, and configure transportation routes and resources.

10. A method for optimizing the production process of a manufacturing enterprise based on intelligent coding, the method being used to implement the system described in any one of claims 1 to 9, characterized in that: The method comprises the following steps: S1. Use the coding structure of static part-dynamic part-semantic relationship to encode materials, equipment, business units, departments and personnel respectively; S2. Collect and integrate the coded data of the above five categories of objects to complete material demand forecasting and matching, production task scheduling, and personnel task allocation in the production plan; S3. Collect and integrate the coded data of the above five categories of objects, complete the formulation of procurement requirements and supplier selection in material procurement, and use blockchain technology to record the entire process data information; S4. Collect and integrate the coded data of the above five types of objects, complete dynamic storage location allocation and real-time inventory counting in warehouse management, and use blockchain technology to save inventory change records; S5. Collect and integrate the coded data of the above five categories of objects, use artificial intelligence and Internet of Things technologies in manufacturing management to optimize the production process, and respond and handle production process anomalies and equipment anomalies in a timely manner; S6. Collect and integrate the coded data of the above five categories of objects to realize intelligent logistics scheduling in logistics management.

Images