Vehicle frame production information management method and system

Abstract

The invention discloses a vehicle frame production information management method and system, and relates to the technical field of production optimization. A vehicle frame production information management system comprises a sensing module, a data storage module, a data processing module, a production optimization model module, a plan management module, a quality inspection management module and a data analysis and decision support module. Based on a machine learning model and a neural network technology, historical operation data and real-time parameters of the equipment are analyzed, and the future fault probability and the fault type of the equipment are accurately predicted; and when the equipment fault risk exceeds the preset threshold value, early warning is triggered, and a maintenance or replacement suggestion is generated, so that the shutdown risk caused by the equipment fault in the production process is remarkably reduced, and intelligent and predictive maintenance of equipment management is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of production optimization, and in particular to a frame production information management method and system. Background Art

[0002] The frame production process is a complex manufacturing process involving multiple links, multiple equipment, and multiple processes. In the existing technology, the relevant information management system usually includes modules such as material management, equipment management, and work order management, but there is a lack of efficient linkage and data sharing mechanism between the modules. Especially in the face of dynamic production needs and complex production environments, the existing system has the following problems: The existing system does not fully collect information on equipment operation status and material flow. Data mainly relies on manual entry or single-point monitoring, which lacks real-time and accuracy. Equipment operation monitoring and maintenance mainly rely on fixed-cycle inspections and empirical judgments, making it difficult to predict equipment failures in a timely manner. Existing production plans are usually based on static process flows and fixed resource allocations, and cannot dynamically respond to emergencies such as equipment failures and material shortages. The existing quality inspection management system mainly focuses on the pass rate statistics of production results, and lacks in-depth analysis of defect types and distribution, resulting in the inability to effectively feed back quality inspection results to the production process optimization link, making it difficult to form a closed-loop improvement mechanism. The existing system's analysis of production data mostly stays in a single dimension, lacks the correlation analysis and prediction functions of multi-module data, and is difficult to provide in-depth support for management decisions.

[0003] The existing technology has significant deficiencies in the comprehensiveness, real-time, intelligence and closed-loop control of frame production information management. These problems seriously limit the further improvement of frame production efficiency and product quality. Summary of the invention

[0004] A frame production information management method comprises the following steps: Step S1, equipment and production data collection: collect equipment operating parameters in real time through sensors in the perception layer; collect material flow information through RFID technology, and record operators, equipment status and production progress data during the execution of production work orders; Step S2, data storage and processing: upload the collected equipment and production data to the data processing layer, store them based on the distributed database, and remove noise and abnormal data through the data cleaning and preprocessing module; Step S3, equipment failure prediction and optimization suggestions: The historical operation data of the equipment is trained through the production optimization model, and the failure probability and failure type of the equipment in future operation are predicted in combination with the real-time collected operation parameters to obtain the operation status and operation results of the equipment; when the predicted failure risk exceeds the preset threshold, an early warning is triggered, and a comprehensive optimization suggestion based on operation efficiency, maintenance cost and equipment replacement price is generated; Step S4, production plan optimization: according to the equipment operation status prediction results and material inventory information, dynamically adjust the production plan in the case of failure risk or insufficient inventory; by analyzing the equipment maintenance needs, reallocate tasks to other equipment, and adjust the production schedule based on the production demand priority; Step S5, quality inspection management and process optimization: generate a quality inspection plan based on the production plan, including the frequency of quality inspection tasks, the setting of quality inspection points and the allocation of personnel; record the quality inspection results in real time, analyze the quality inspection data to generate a quality report, and optimize the process flow based on the quality inspection defect type; Step S6, comprehensive data analysis and report generation: Integrate equipment status, production progress, material flow and quality inspection data through the data processing layer to generate daily, weekly and monthly production efficiency reports; provide production bottleneck analysis based on historical data, equipment replacement suggestions and optimization decision support for personnel skill training needs.

[0005] As a preferred technical solution of the present invention, the data collection includes: Equipment operation parameter collection: Through the perception layer, various sensors, including temperature sensors, vibration sensors, current sensors, and energy consumption monitors, are arranged on the frame production equipment to obtain the equipment's operating status data in real time; the collected operating parameters are stored in the form of time series data, and are preliminarily processed through the edge computing module to remove abnormal fluctuations; Material flow information collection: On the frame production line, each batch of materials is bound with a unique RFID tag. The RFID reading equipment arranged at key nodes records the material flow information, including the material entry time, exit time, flow path and current process status. At the same time, it supports QR code tag scanning as a supplementary technology to ensure the comprehensiveness of material tracking. Production work order execution data collection: During the execution of the production work order, the operator information is collected through the operator identity authentication system and bound to the production task; combined with the equipment operating parameters and the production work order plan, the equipment start and stop time, task completion time, production quantity and workstation status are recorded; the production progress data is updated in real time through the production monitoring terminal and uploaded to the data processing layer.

[0006] As a preferred technical solution of the present invention, the data storage includes: Distributed database storage: The data processing layer stores the collected equipment operation data, material flow data, and production work order data based on the distributed database, supporting efficient access to massive data and fault tolerance under the distributed storage architecture; Data classification and hierarchical storage: The stored data is classified into equipment status data, material flow data, quality inspection data and production work order data, and stored in different data partitions in chronological order; equipment status data includes real-time monitoring data of temperature, vibration, current and energy consumption; material flow data includes information on material inbound, outbound and process flow; quality inspection data includes quality inspection results and defect types; production work order data includes task execution time, operator and equipment allocation information; Historical data archiving and index management: Archiving and storing historical production data, and establishing multi-dimensional indexes to support quick query based on time, equipment number, work order number or material batch number; archived data is used for subsequent trend analysis, process improvement and training of equipment failure prediction models; Real-time data caching: The real-time data generated during the production process is temporarily stored through the cache module to reduce the database access pressure and ensure the rapid processing of real-time data and the efficiency of subsequent data analysis.

[0007] As a preferred technical solution of the present invention, the data processing includes: Data cleaning and exception handling: Clean the collected equipment operating parameters, material flow data and production work order data to filter out missing, duplicate or abnormal data; use statistical rules and anomaly detection algorithms to identify and correct abnormal values ​​to ensure data accuracy and reliability; Data preprocessing: normalize and standardize the cleaned data; slice the time series data into fixed time windows to generate a data format suitable for analysis and prediction; perform structured processing on different types of data and unify the storage format to support subsequent processing; Data integration and association analysis: Integrate equipment status data, material flow data and production work order data, and establish logical associations between data through association rule analysis algorithms; identify potential impact patterns between equipment status and production work order completion efficiency and material flow; Model input data generation: Generate input data sets for production optimization model training based on the integrated data, including feature vectors and labels required for equipment failure prediction model and production optimization model; extract key variables through data feature selection algorithm, remove redundant information, and improve model training efficiency; Real-time data stream processing: Use the stream processing framework to instantly calculate and update the real-time collected data, including equipment status change trends, material inventory status and production progress, to support real-time production plan adjustments and triggering of equipment maintenance warnings.

[0008] As a preferred technical solution of the present invention, the structure of the production optimization model includes: Input layer: receives equipment operating parameters, material flow information, production work order execution data and quality inspection result data, and normalizes the data as input variables, including time series data and multi-dimensional feature vectors; Feature extraction module: extracts features from input time series data through convolutional neural networks to capture potential key features in equipment operation status and production processes; combines fully connected networks to fuse static features; Prediction module: Use recurrent neural network to process the extracted time series features to predict the future equipment operation status, production efficiency and material consumption; use multi-task learning model to simultaneously generate equipment failure probability and production optimization suggestions; Decision-making module: Based on the prediction results, combined with the rule-based optimization algorithm, it generates production plan adjustment suggestions, including equipment task allocation optimization, material procurement plan adjustment, and operator rescheduling suggestions; when the risk of equipment failure is higher than the threshold, it triggers an early warning and generates an equipment replacement plan; Output layer: Convert the output results of the production optimization model into executable production plan adjustment instructions, and send them to relevant equipment and operators in real time through the application layer. At the same time, generate a visual report of optimization suggestions for managers' reference.

[0009] As a preferred technical solution of the present invention, the training of the production optimization model includes: Training data preparation: extract historical equipment operation data, material flow data, production work order data and quality inspection data, and generate feature vectors according to time series; normalize different types of data, and annotate data labels based on actual fault records and production efficiency; Data enhancement: For the time series data of equipment operating parameters, interpolation methods are used to fill missing values, and data enhancement methods are used to improve the generalization ability of the model; Model construction: Build a production optimization model based on a deep learning framework, including feature extraction module, prediction module and decision module; the feature extraction module uses a convolutional neural network to extract key features of the equipment status, the prediction module uses a recurrent neural network to process time series data, and the decision module combines a rule-based optimization algorithm to generate production optimization suggestions; Training process: Use supervised learning methods to train the model, iteratively update the model weights based on the optimized loss function; use batch gradient descent algorithm to improve the model convergence speed and prediction accuracy; Model verification and optimization: Verify the trained model through the reserved verification set to evaluate the accuracy of the model in equipment failure prediction, production efficiency prediction and optimization suggestion generation; adjust the model hyperparameters according to the verification results to improve the model performance; Model deployment: Integrate the trained model into the data processing layer and application layer for real-time analysis of equipment operating status, prediction of production efficiency, and generation of optimization suggestions; regularly retrain the model with new data to ensure prediction accuracy and applicability.

[0010] As a preferred technical solution of the present invention, the production plan optimization includes: Production resource status monitoring: Real-time acquisition of equipment operating status, material inventory levels, and production work order execution status. Analyze equipment availability and material sufficiency through the data processing layer to provide data support for production plan adjustments. Task priority adjustment: Dynamically sort production tasks based on production demand priority and order delivery time; Dynamic equipment task allocation: When a device predicts a risk of failure or enters maintenance status, the system automatically allocates its tasks to other available devices; combining equipment performance data and task requirements to optimize equipment task allocation to improve production efficiency; Material scheduling optimization: Dynamically generate material procurement plans or internal transfer plans by analyzing material inventory data and production task requirements; adjust production sequences to avoid downtime when inventory is insufficient or logistics is delayed; Personnel scheduling management: intelligently match tasks and personnel according to the complexity of production tasks and the skill level of operators; when there is a shortage of personnel, generate temporary deployment suggestions to ensure the normal progress of production tasks; Real-time production adjustment and feedback: Combine equipment status monitoring with production order execution progress to dynamically adjust production plans and provide real-time feedback to production line operators; display the latest production plans and key adjustment points through a visual interface.

[0011] As a preferred technical solution of the present invention, the quality inspection management and process optimization include: Quality inspection plan generation: Generate a quality inspection plan based on the production plan and process flow, clarify the frequency, time, quality inspection points and corresponding operators of quality inspection tasks; the quality inspection plan includes daily quality inspection and periodic quality inspection tasks, covering key processes and equipment; Quality inspection data collection and recording: The quality inspection result data, including product qualification rate, defective product type and distribution, are collected in real time through the quality inspection terminal equipment; the operator, quality inspection time and quality inspection process of each quality inspection are recorded, and the data is uploaded to the data processing layer for storage and analysis; Quality inspection data analysis: Use the data processing layer to analyze historical quality inspection data and identify production links and defect patterns with high non-conformity rates; use statistical analysis and machine learning algorithms to locate weak links in the process flow and provide data support for process optimization; Generation of process improvement suggestions: Generate process improvement suggestions based on the results of quality inspection data analysis, including adjusting process parameters, adding quality inspection points, optimizing process sequence, etc.; combine equipment status data and production plans to ensure the feasibility of process optimization suggestions; Real-time feedback and closed-loop improvement: The quality inspection data analysis results are fed back to the production line in real time, and closed-loop control of quality inspection management and process optimization is achieved by adjusting process parameters or reminding operators to pay attention to key issues; Quality inspection report generation: Generate periodic quality inspection reports, including product qualification rate statistics, defective product type distribution and process improvement records, to provide decision support for quality management.

[0012] As a preferred technical solution of the present invention, the comprehensive data analysis includes: Data integration and correlation analysis: Integrate equipment operation status, material flow information, production work order data and quality inspection result data through the data processing layer; use multi-dimensional correlation analysis algorithms to identify the logical relationship between various data modules, including the impact of equipment status on production efficiency and quality inspection results, and the dependence of material flow on work order execution; Production efficiency analysis: Based on historical production data and real-time production progress data, analyze the task completion rate, equipment utilization rate and production bottleneck of the production line; identify the change pattern of production efficiency through trend analysis and provide targeted optimization suggestions; Equipment management analysis: Statistical analysis of equipment failure history records and operating status data is performed to generate equipment operation reliability reports; combined with energy consumption and maintenance cost data, quantitative support is provided for equipment replacement decisions to ensure optimal management of the equipment life cycle; Quality inspection data analysis and improvement: classify and count the quality inspection results, analyze the product qualification rate, defective product distribution and defect type; combine production and process data to generate quality improvement measures and optimize the production process; Personnel skills and efficiency assessment: Based on production work order records and quality inspection results, evaluate the operator's task completion efficiency and quality compliance rate; generate personnel training needs reports and provide skills improvement suggestions based on task matching; Multi-dimensional data report generation: Based on the integrated data analysis results, comprehensive reports including production efficiency, equipment utilization, quality inspection results, personnel skills and material consumption are generated; support daily, weekly and monthly multi-time dimension report generation, and display analysis results through a visual interface.

[0013] A frame production information management system includes the following modules: Perception module: used to collect equipment operating parameters through sensors; collect material flow information through RFID technology or QR code scanning equipment, and record operators, equipment status and production progress data during the execution of production work orders; Data storage module: used to classify and store the collected equipment operation data, material flow data, production work order data and quality inspection data based on a distributed database; supports historical data archiving, multi-dimensional indexing and real-time data caching; Data processing module: used to clean, normalize, standardize and integrate the collected data; through data feature extraction and model input data generation, it supports the analysis needs of equipment failure prediction and production plan optimization; Production optimization model module: including feature extraction module, prediction module and decision-making module, which predicts the equipment operation status and production efficiency based on neural network technology, and generates comprehensive suggestions for task priority adjustment, equipment task allocation, material scheduling and process optimization; Planning management module: used to dynamically adjust the production plan according to the equipment operation status, material inventory information and work order execution progress, provide real-time feedback and optimize task allocation; generate a plan adjustment plan when equipment failure or material shortage is detected; Quality inspection management module: used to generate quality inspection plans and record quality inspection results; by analyzing quality inspection data, generate quality inspection reports and process improvement suggestions, and achieve closed-loop control of quality inspection management and process optimization; Data analysis and decision support module: used to integrate equipment status, production progress, material flow and quality inspection data, generate comprehensive reports on production efficiency, equipment utilization, quality inspection results and personnel skills, and provide optimization support for production bottleneck analysis, equipment replacement decisions and personnel training needs.

[0014] The present invention has the following advantages: Based on machine learning models and neural network technology, the present invention analyzes the historical operation data and real-time parameters of the equipment to accurately predict the future failure probability and failure type of the equipment; when the equipment failure risk exceeds the preset threshold, an early warning is triggered and maintenance or replacement suggestions are generated, which significantly reduces the risk of downtime caused by equipment failure in the production process and realizes intelligent equipment management and predictive maintenance.

[0015] The present invention dynamically adjusts the production plan according to the equipment operation status prediction results and material inventory information, including task reallocation, equipment task switching and production priority adjustment, thereby improving the flexibility and responsiveness of production scheduling. Especially in emergencies such as equipment failure or material shortage, it can quickly generate optimized production plans to ensure the continuity of the production process.

[0016] The present invention generates quality inspection plans in real time, records quality inspection data and analyzes defect types, generates targeted process optimization suggestions, and feeds the optimization results back to the production process, forming a closed-loop mechanism from quality inspection to process improvement, which significantly improves product quality and production efficiency.

[0017] The present invention integrates equipment status, production progress, material flow and quality inspection data, and uses a multi-dimensional correlation analysis algorithm to generate comprehensive reports, including production efficiency, equipment utilization, quality inspection results and personnel skill evaluation, to provide managers with comprehensive optimization decision support. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art description are briefly introduced below. Obviously, the drawings in the following description are only schematic diagrams of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on the provided drawings without creative work. Figure 1 This is a schematic diagram of the structure of a vehicle frame production information management system used in an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, rather than all 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.

[0020] Embodiment 1, a frame production information management method, comprising the following steps: Step S1, equipment and production data collection: collect equipment operating parameters in real time through sensors in the perception layer; collect material flow information through RFID technology, and record operators, equipment status and production progress data during the execution of production work orders; The data collection includes: Equipment operation parameter collection: Through the perception layer, various sensors, including temperature sensors, vibration sensors, current sensors and energy consumption monitors, are arranged on the frame production equipment to obtain the equipment's operating status data in real time; the collected operating parameters are stored in the form of time series data, and are preliminarily processed through the edge computing module to remove abnormal fluctuations; the collection frequency is adjusted according to the equipment's operating conditions to ensure the accuracy and real-time nature of the data; Material flow information collection: On the frame production line, each batch of materials is bound with a unique RFID tag. The RFID reading devices arranged at key nodes (warehouses, processing stations and finished product warehouses) record the material flow information, including the material entry time, exit time, flow path and current process status; at the same time, QR code label scanning is supported as a supplementary technology to ensure the comprehensiveness of material tracking; Production work order execution data collection: During the execution of the production work order, the operator information is collected through the operator identity authentication system and bound to the production task; combined with the equipment operating parameters and the production work order plan, the equipment start and stop time, task completion time, production quantity and workstation status are recorded; the production progress data is updated in real time through the production monitoring terminal and uploaded to the data processing layer for subsequent plan adjustments and efficiency analysis.

[0021] Step S2, data storage and processing: upload the collected equipment and production data to the data processing layer, store them based on the distributed database, and remove noise and abnormal data through the data cleaning and preprocessing module; The data storage includes: Distributed database storage: The data processing layer stores the collected equipment operation data, material flow data, and production work order data based on the distributed database, supporting efficient access to massive data and fault tolerance under the distributed storage architecture; Data classification and hierarchical storage: The stored data is classified into equipment status data, material flow data, quality inspection data and production work order data, and stored in different data partitions in chronological order; equipment status data includes real-time monitoring data of temperature, vibration, current and energy consumption; material flow data includes information on material inbound, outbound and process flow; quality inspection data includes quality inspection results and defect types; production work order data includes task execution time, operator and equipment allocation information; Historical data archiving and index management: Archiving and storing historical production data, and establishing multi-dimensional indexes to support quick query based on time, equipment number, work order number or material batch number; archived data is used for subsequent trend analysis, process improvement and training of equipment failure prediction models; Real-time data caching: The real-time data generated during the production process is temporarily stored through the cache module to reduce the database access pressure and ensure the rapid processing of real-time data and the efficiency of subsequent data analysis.

[0022] The data processing includes: Data cleaning and exception handling: Clean the collected equipment operating parameters, material flow data and production work order data to filter out missing, duplicate or abnormal data; use statistical rules and anomaly detection algorithms to identify and correct abnormal values ​​to ensure data accuracy and reliability; Data preprocessing: normalize and standardize the cleaned data; slice the time series data into fixed time windows to generate a data format suitable for analysis and prediction; perform structured processing on different types of data and unify the storage format to support subsequent processing; Data integration and association analysis: Integrate equipment status data, material flow data and production work order data, and establish logical associations between data through association rule analysis algorithms; identify potential impact patterns between equipment status and production work order completion efficiency and material flow; Model input data generation: Generate input data sets for production optimization model training based on the integrated data, including feature vectors and labels required for equipment failure prediction model and production optimization model; extract key variables through data feature selection algorithm, remove redundant information, and improve model training efficiency; Real-time data stream processing: Use the stream processing framework to instantly calculate and update the real-time collected data, including equipment status change trends, material inventory status and production progress, to support real-time production plan adjustments and triggering of equipment maintenance warnings.

[0023] The association rule algorithm specifically includes: By analyzing the relationship between equipment operation data, material flow data, production work order data and quality inspection data, key influencing factors in the production process can be identified.

[0024] Specific goals include: Correlation analysis between equipment status and production efficiency: Explore the impact of equipment operating status (vibration frequency, temperature, current fluctuation) on production efficiency.

[0025] Correlation analysis between material flow and work order execution efficiency: Identify the relationship between material flow speed, inventory level and production task completion time.

[0026] Correlation analysis between quality inspection data and process flow: Find out the correlation between specific quality inspection defects (dimensional deviation, poor welding) and process parameters or equipment status.

[0027] Equipment operation data, material flow data, production work order data and quality inspection data are extracted from the data processing layer.

[0028] To facilitate algorithm analysis, continuous data is discretized. For example: "Temperature parameters: low, medium, high; vibration frequency: normal, slightly abnormal, seriously abnormal; production efficiency: high efficiency, normal, low efficiency"; Data encoding: Converting discretized data into transaction record form, for example: “{Equipment temperature = high, vibration frequency = serious abnormality, production efficiency = low efficiency}”; “{Inventory status = insufficient, Task completion time = delayed, Quality inspection defect = poor welding}”; Application process of association rule analysis: Use Apriori algorithm or FP-Growth algorithm to mine association rules.

[0029] Apriori algorithm: gradually generates frequent item sets and screens out rules whose support and confidence meet the requirements.

[0030] FP-Growth algorithm: quickly generates frequent item sets based on FP tree, which is more suitable for large-scale data analysis.

[0031] Parameter settings: Support: reflects the frequency of rule occurrence. For example, the support of "high equipment temperature -> low production efficiency" indicates the proportion of high temperature and low efficiency occurring at the same time.

[0032] Confidence: A measure of the credibility of a rule. For example, the confidence of "if the temperature is high, the production efficiency is low" indicates the probability of low production efficiency under high temperature conditions.

[0033] Lift: A measure of the strength of the rule's association. If lift > 1, it means there is a positive correlation between the two events.

[0034] Rule mining: Input the discretized transaction data, run the association rule algorithm, and obtain the rule set, for example: Rule 1: {Equipment vibration = abnormal} -> {Production efficiency = inefficient}, support = 20%, confidence = 85%, lift = 1.5.

[0035] Rule 2: {Inventory status = insufficient} -> {Task completion time = delayed}, support = 15%, confidence = 90%, lift = 2.0.

[0036] Application examples: Equipment status optimization: Mining rules: {Equipment temperature = high, vibration frequency = abnormal} -> {Quality inspection defect = poor welding}.

[0037] Analysis results: High temperature and abnormal vibration are the main causes of poor welding.

[0038] Optimization measures: Improve the temperature control capability of the equipment, add vibration monitoring points, and adjust operating parameters in a timely manner.

[0039] Production plan adjustment: Mining rule: {inventory status = insufficient}->{task completion time = delayed}.

[0040] Analysis results: Material shortages significantly extend the time it takes to complete a task.

[0041] Optimization measures: Provide early warning of low inventory status and dynamically adjust material procurement plans.

[0042] Process improvements: Mining rules: {equipment vibration = abnormal}->{production efficiency = inefficient}.

[0043] Analysis results: Abnormal vibration leads to reduced efficiency.

[0044] Optimization measures: Adjust the maintenance cycle of equipment to reduce the occurrence of abnormal vibration.

[0045] Result feedback and iterative optimization: Feedback mechanism, which feeds back analysis results to the application layer, generates optimization suggestions in real time, including adjusting production plans, optimizing process parameters, and formulating maintenance plans. Dynamic update, continuously collect new data, and regularly re-run the association rule analysis algorithm to ensure the real-time and accuracy of the rule set.

[0046] Step S3, equipment failure prediction and optimization suggestions: The historical operation data of the equipment is trained through the production optimization model, and the failure probability and failure type of the equipment in future operation are predicted in combination with the real-time collected operation parameters to obtain the operation status and operation results of the equipment; when the predicted failure risk exceeds the preset threshold, an early warning is triggered, and a comprehensive optimization suggestion based on operation efficiency, maintenance cost and equipment replacement price is generated; The structure of the production optimization model includes: Input layer: receives equipment operating parameters, material flow information, production work order execution data and quality inspection result data, and normalizes the data as input variables, including time series data and multi-dimensional feature vectors; Feature extraction module: extracts features from input time series data through convolutional neural networks to capture potential key features in equipment operation status and production processes; combines fully connected networks to fuse static features (material inventory, equipment status labels); Prediction module: Use recurrent neural network to process the extracted time series features to predict the future equipment operation status, production efficiency and material consumption; use multi-task learning model to simultaneously generate equipment failure probability and production optimization suggestions; Decision-making module: Based on the prediction results, combined with the rule-based optimization algorithm, it generates production plan adjustment suggestions, including equipment task allocation optimization, material procurement plan adjustment, and operator rescheduling suggestions; when the risk of equipment failure is higher than the threshold, it triggers an early warning and generates an equipment replacement plan; Output layer: Convert the output results of the production optimization model into executable production plan adjustment instructions, and send them to relevant equipment and operators in real time through the application layer. At the same time, generate a visual report of optimization suggestions for managers' reference.

[0047] The training of the production optimization model includes: Training data preparation: extract historical equipment operation data, material flow data, production work order data and quality inspection data, and generate feature vectors according to time series; normalize different types of data, and annotate data labels based on actual fault records and production efficiency; Data enhancement: For the time series data of equipment operating parameters, interpolation methods are used to fill missing values, and data enhancement methods are used to improve the generalization ability of the model; the data enhancement methods include smoothing, noise addition, and time window adjustment; Model construction: Build a production optimization model based on a deep learning framework, including feature extraction module, prediction module and decision module; the feature extraction module uses a convolutional neural network to extract key features of the equipment status, the prediction module uses a recurrent neural network to process time series data, and the decision module combines a rule-based optimization algorithm to generate production optimization suggestions; Training process: The supervised learning method is used to train the model, and the model weights are iteratively updated based on the optimized loss function (mean square error loss function); the batch gradient descent algorithm is used to improve the model convergence speed and prediction accuracy; Model verification and optimization: Verify the trained model through the reserved verification set to evaluate the accuracy of the model in equipment failure prediction, production efficiency prediction and optimization suggestion generation; adjust the model hyperparameters (learning rate, number of neurons and network depth) according to the verification results to improve the model performance; Model deployment: Integrate the trained model into the data processing layer and application layer for real-time analysis of equipment operating status, prediction of production efficiency, and generation of optimization suggestions; regularly retrain the model with new data to ensure prediction accuracy and applicability.

[0048] Step S4, production plan optimization: according to the equipment operation status prediction results and material inventory information, dynamically adjust the production plan in the case of failure risk or insufficient inventory; by analyzing the equipment maintenance needs, reallocate tasks to other equipment, and adjust the production schedule based on the production demand priority; The production plan optimization includes: Production resource status monitoring: Real-time acquisition of equipment operating status, material inventory levels, and production work order execution status. Analyze equipment availability and material sufficiency through the data processing layer to provide data support for production plan adjustments. Task priority adjustment: Dynamically sort production tasks based on production demand priority and order delivery time; Dynamic equipment task allocation: When a device predicts a risk of failure or enters maintenance status, the system automatically allocates its tasks to other available devices; combining equipment performance data and task requirements to optimize equipment task allocation to improve production efficiency; Material scheduling optimization: Dynamically generate material procurement plans or internal transfer plans by analyzing material inventory data and production task requirements; adjust production sequences to avoid downtime when inventory is insufficient or logistics is delayed; Personnel scheduling management: intelligently match tasks and personnel according to the complexity of production tasks and the skill level of operators; when there is a shortage of personnel, generate temporary deployment suggestions to ensure the normal progress of production tasks; Real-time production adjustment and feedback: Combine equipment status monitoring with production order execution progress to dynamically adjust production plans and provide real-time feedback to production line operators; display the latest production plans and key adjustment points through a visual interface.

[0049] Step S5, quality inspection management and process optimization: generate a quality inspection plan based on the production plan, including the frequency of quality inspection tasks, the setting of quality inspection points and the allocation of personnel; record the quality inspection results in real time, analyze the quality inspection data to generate a quality report, and optimize the process flow based on the quality inspection defect type; The quality inspection management and process optimization include: Quality inspection plan generation: Generate a quality inspection plan based on the production plan and process flow, clarify the frequency, time, quality inspection points and corresponding operators of quality inspection tasks; the quality inspection plan includes daily quality inspection and periodic quality inspection tasks, covering key processes and equipment; Quality inspection data collection and recording: The quality inspection result data, including product qualification rate, defective product type and distribution, are collected in real time through the quality inspection terminal equipment; the operator, quality inspection time and quality inspection process of each quality inspection are recorded, and the data is uploaded to the data processing layer for storage and analysis; Quality inspection data analysis: Use the data processing layer to analyze historical quality inspection data and identify production links and defect patterns with high non-conformity rates; use statistical analysis and machine learning algorithms to locate weak links in the process flow and provide data support for process optimization; Generation of process improvement suggestions: Generate process improvement suggestions based on the results of quality inspection data analysis, including adjusting process parameters, adding quality inspection points, optimizing process sequence, etc.; combine equipment status data and production plans to ensure the feasibility of process optimization suggestions; Real-time feedback and closed-loop improvement: The quality inspection data analysis results are fed back to the production line in real time, and closed-loop control of quality inspection management and process optimization is achieved by adjusting process parameters or reminding operators to pay attention to key issues; Quality inspection report generation: Generate periodic quality inspection reports, including product qualification rate statistics, defective product type distribution and process improvement records, to provide decision support for quality management.

[0050] Step S6, comprehensive data analysis and report generation: Integrate equipment status, production progress, material flow and quality inspection data through the data processing layer to generate daily, weekly and monthly production efficiency reports; provide production bottleneck analysis based on historical data, equipment replacement suggestions and optimization decision support for personnel skill training needs.

[0051] The comprehensive data analysis includes: Data integration and correlation analysis: Integrate equipment operation status, material flow information, production work order data and quality inspection result data through the data processing layer; use multi-dimensional correlation analysis algorithms to identify the logical relationship between various data modules, including the impact of equipment status on production efficiency and quality inspection results, and the dependence of material flow on work order execution; Production efficiency analysis: Based on historical production data and real-time production progress data, analyze the task completion rate, equipment utilization rate and production bottleneck of the production line; identify the change pattern of production efficiency through trend analysis and provide targeted optimization suggestions; Equipment management analysis: Statistical analysis of equipment failure history records and operating status data is performed to generate equipment operation reliability reports; combined with energy consumption and maintenance cost data, quantitative support is provided for equipment replacement decisions to ensure optimal management of the equipment life cycle; Quality inspection data analysis and improvement: classify and count the quality inspection results, analyze the product qualification rate, defective product distribution and defect type; combine production and process data to generate quality improvement measures and optimize the production process; Personnel skills and efficiency assessment: Based on production work order records and quality inspection results, evaluate the operator's task completion efficiency and quality compliance rate; generate personnel training needs reports and provide skills improvement suggestions based on task matching; Multi-dimensional data report generation: Based on the integrated data analysis results, comprehensive reports including production efficiency, equipment utilization, quality inspection results, personnel skills and material consumption are generated; support daily, weekly and monthly multi-time dimension report generation, and display analysis results through a visual interface.

[0052] Example 2, a frame production information management system, see Figure 1 As shown, it includes the following modules: Perception module: used to collect equipment operating parameters through sensors; collect material flow information through RFID technology or QR code scanning equipment, and record operators, equipment status and production progress data during the execution of production work orders; Data storage module: used to classify and store the collected equipment operation data, material flow data, production work order data and quality inspection data based on a distributed database; supports historical data archiving, multi-dimensional indexing and real-time data caching; Data processing module: used to clean, normalize, standardize and integrate the collected data; through data feature extraction and model input data generation, it supports the analysis needs of equipment failure prediction and production plan optimization; Production optimization model module: including feature extraction module, prediction module and decision-making module, which predicts the equipment operation status and production efficiency based on neural network technology, and generates comprehensive suggestions for task priority adjustment, equipment task allocation, material scheduling and process optimization; Planning management module: used to dynamically adjust the production plan according to the equipment operation status, material inventory information and work order execution progress, provide real-time feedback and optimize task allocation; generate a plan adjustment plan when equipment failure or material shortage is detected; Quality inspection management module: used to generate quality inspection plans and record quality inspection results; by analyzing quality inspection data, generate quality inspection reports and process improvement suggestions, and achieve closed-loop control of quality inspection management and process optimization; Data analysis and decision support module: used to integrate equipment status, production progress, material flow and quality inspection data, generate comprehensive reports on production efficiency, equipment utilization, quality inspection results and personnel skills, and provide optimization support for production bottleneck analysis, equipment replacement decisions and personnel training needs.

[0053] Embodiment 3, the frame production information management method and system proposed by the present invention include a perception layer, a data processing layer and an application layer, and each layer specifically implements the following functions in the system: 1. The perception layer is mainly responsible for real-time data collection and transmission and is the basic data source of the system. Its specific implementation is as follows: Sensor network: Various sensors configured on the frame production line, including temperature sensors, vibration sensors, current sensors, energy consumption sensors, etc., are used to collect equipment operating parameters in real time, such as temperature changes, vibration intensity, current fluctuations, and energy consumption levels.

[0054] RFID technology: RFID tags and reading devices are installed at the key nodes of material entry, exit and circulation to record the material's location information, batch number, process status, etc. in real time. For materials without RFID function, QR code scanning technology is used to realize the supplementary collection of material circulation information.

[0055] Data collection equipment: including Industrial Internet of Things (IIoT) gateways, which aggregate data from sensors and RFID devices and transmit them to the data processing layer; support local caching and data upload to ensure real-time and data integrity.

[0056] 2. The data processing layer is responsible for the storage, cleaning, analysis and processing of collected data and is the core data analysis module of the system. Its specific implementation is as follows: Distributed database: It adopts a distributed storage architecture to classify and store equipment operation data, material flow data, work order data and quality inspection data, supports high-concurrency reading and writing and multi-dimensional index query, and meets the storage needs of massive data.

[0057] Data cleaning and preprocessing: Clean and normalize the uploaded data to remove noise data and outliers; slice the time series data according to fixed windows to ensure the standardization of input data.

[0058] Data analysis and model operation: Use neural networks and machine learning algorithms to model and analyze data; Equipment failure prediction model: predicts failure risk and type based on historical equipment operation data and real-time parameters.

[0059] Production optimization: Generate production plan adjustment suggestions based on real-time production data and equipment status.

[0060] Quality inspection optimization: Analyze the causes of defects based on quality inspection results to provide a basis for process optimization.

[0061] Real-time stream processing: Stream process the real-time data of the production line to generate instant equipment status monitoring, material inventory status and production work order progress reports.

[0062] 3. The application layer is responsible for interacting with users and production line operators to provide data visualization and optimization decision support. Its specific implementation is as follows: Production plan optimization: Dynamically adjust the production plan based on the analysis results of the data processing layer, including task reallocation, equipment maintenance schedule and material allocation plan.

[0063] Quality inspection management and process optimization: Provides the function of generating and adjusting quality inspection plans, records quality inspection results in real time, and generates process optimization suggestions based on data analysis results, which are then fed back to the production line.

[0064] Equipment management and early warning: Display equipment operating status and maintenance requirements in real time. When the risk of equipment failure exceeds the threshold, an early warning is triggered to remind maintenance personnel to conduct inspections.

[0065] Comprehensive reports and decision support: Generate daily, weekly and monthly comprehensive production analysis reports for managers, including production efficiency, equipment utilization, quality inspection data statistics and production bottleneck analysis.

[0066] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A frame production information management method, characterized in that: The following steps are involved: Step S1, equipment and production data collection: collect equipment operating parameters in real time through sensors in the perception layer; collect material flow information through RFID technology, and record operators, equipment status and production progress data during the execution of production work orders; Step S2, data storage and processing: upload the collected equipment and production data to the data processing layer, store them based on the distributed database, and remove noise and abnormal data through the data cleaning and preprocessing module; Step S3, equipment failure prediction and optimization suggestions: The historical operation data of the equipment is trained through the production optimization model, and the failure probability and failure type of the equipment in future operation are predicted in combination with the real-time collected operation parameters to obtain the operation status and operation results of the equipment; when the predicted failure risk exceeds the preset threshold, an early warning is triggered, and a comprehensive optimization suggestion based on operation efficiency, maintenance cost and equipment replacement price is generated; Step S4, production plan optimization: according to the equipment operation status prediction results and material inventory information, dynamically adjust the production plan in the event of failure risk or insufficient inventory; Analyze equipment maintenance needs, reallocate tasks to other equipment, and adjust production schedules based on production demand priorities; Step S5, quality inspection management and process optimization: generate a quality inspection plan based on the production plan, including the frequency of quality inspection tasks, the setting of quality inspection points and the allocation of personnel; record the quality inspection results in real time, analyze the quality inspection data to generate a quality report, and optimize the process flow based on the quality inspection defect type; Step S6, comprehensive data analysis and report generation: Integrate equipment status, production progress, material flow and quality inspection data through the data processing layer to generate daily, weekly and monthly production efficiency reports; provide production bottleneck analysis based on historical data, equipment replacement suggestions and optimization decision support for personnel skill training needs.

2. A frame production information management method according to claim 1, characterized in that: The data collection includes: Equipment operation parameter collection: Through the perception layer, various sensors, including temperature sensors, vibration sensors, current sensors, and energy consumption monitors, are arranged on the frame production equipment to obtain the equipment's operating status data in real time; the collected operating parameters are stored in the form of time series data, and are preliminarily processed through the edge computing module to remove abnormal fluctuations; Material flow information collection: On the frame production line, each batch of materials is bound with a unique RFID tag. The RFID reading equipment arranged at key nodes records the material flow information, including the material entry time, exit time, flow path and current process status. At the same time, it supports QR code tag scanning as a supplementary technology to ensure the comprehensiveness of material tracking. Production work order execution data collection: During the execution of the production work order, the operator information is collected through the operator identity authentication system and bound to the production task; combined with the equipment operating parameters and the production work order plan, the equipment start and stop time, task completion time, production quantity and workstation status are recorded; the production progress data is updated in real time through the production monitoring terminal and uploaded to the data processing layer.

3. A frame production information management method according to claim 1, characterized in that: The data storage includes: Distributed database storage: The data processing layer stores the collected equipment operation data, material flow data, and production work order data based on the distributed database, supporting efficient access to massive data and fault tolerance under the distributed storage architecture; Data classification and hierarchical storage: The stored data is classified into equipment status data, material flow data, quality inspection data and production work order data, and stored in different data partitions in chronological order; equipment status data includes real-time monitoring data of temperature, vibration, current and energy consumption; material flow data includes information on material inbound, outbound and process flow; quality inspection data includes quality inspection results and defect types; production work order data includes task execution time, operator and equipment allocation information; Historical data archiving and index management: Archive and store historical production data, and establish multi-dimensional indexes to support quick query based on time, equipment number, work order number or material batch number; archived data is used for subsequent trend analysis, process improvement and equipment failure prediction model training; Real-time data caching: The real-time data generated during the production process is temporarily stored through the cache module to reduce the pressure on database access and ensure the rapid processing of real-time data and the efficiency of subsequent data analysis.

4. A frame production information management method according to claim 1, characterized in that: The data processing includes: Data cleaning and exception handling: Clean the collected equipment operating parameters, material flow data and production work order data to filter out missing, duplicate or abnormal data; use statistical rules and anomaly detection algorithms to identify and correct abnormal values ​​to ensure data accuracy and reliability; Data preprocessing: normalize and standardize the cleaned data; slice the time series data into fixed time windows to generate a data format suitable for analysis and prediction; perform structured processing on different types of data and unify the storage format to support subsequent processing; Data integration and association analysis: Integrate equipment status data, material flow data and production work order data, and establish logical associations between data through association rule analysis algorithms; identify potential impact patterns between equipment status and production work order completion efficiency and material flow; Model input data generation: Generate input data sets for production optimization model training based on the integrated data, including feature vectors and labels required for equipment failure prediction model and production optimization model; extract key variables through data feature selection algorithm, remove redundant information, and improve model training efficiency; Real-time data stream processing: Use the stream processing framework to instantly calculate and update the real-time collected data, including equipment status change trends, material inventory status and production progress, to support real-time production plan adjustments and triggering of equipment maintenance warnings.

5. A frame production information management method according to claim 1, characterized in that: The structure of the production optimization model includes: Input layer: receives equipment operating parameters, material flow information, production work order execution data and quality inspection result data, and normalizes the data as input variables, including time series data and multi-dimensional feature vectors; Feature extraction module: extracts features from input time series data through convolutional neural networks to capture potential key features in equipment operation status and production processes; combines fully connected networks to fuse static features; Prediction module: Use recurrent neural network to process the extracted time series features to predict the future equipment operation status, production efficiency and material consumption; use multi-task learning model to simultaneously generate equipment failure probability and production optimization suggestions; Decision-making module: Based on the prediction results, combined with the rule-based optimization algorithm, it generates production plan adjustment suggestions, including equipment task allocation optimization, material procurement plan adjustment, and operator rescheduling suggestions; when the risk of equipment failure is higher than the threshold, it triggers an early warning and generates an equipment replacement plan; Output layer: Convert the output results of the production optimization model into executable production plan adjustment instructions, and send them to relevant equipment and operators in real time through the application layer. At the same time, generate a visual report of optimization suggestions for managers' reference.

6. A frame production information management method according to claim 1, characterized in that: The training of the production optimization model includes: Training data preparation: extract historical equipment operation data, material flow data, production work order data and quality inspection data, and generate feature vectors according to time series; normalize different types of data, and annotate data labels based on actual fault records and production efficiency; Data enhancement: For the time series data of equipment operating parameters, interpolation methods are used to fill missing values, and data enhancement methods are used to improve the generalization ability of the model; Model construction: Build a production optimization model based on a deep learning framework, including feature extraction module, prediction module and decision module; the feature extraction module uses a convolutional neural network to extract key features of the equipment status, the prediction module uses a recurrent neural network to process time series data, and the decision module combines a rule-based optimization algorithm to generate production optimization suggestions; Training process: Use supervised learning methods to train the model, iteratively update the model weights based on the optimized loss function; use batch gradient descent algorithm to improve the model convergence speed and prediction accuracy; Model verification and optimization: Verify the trained model through the reserved verification set to evaluate the accuracy of the model in equipment failure prediction, production efficiency prediction and optimization suggestion generation; adjust the model hyperparameters according to the verification results to improve the model performance; Model deployment: Integrate the trained model into the data processing layer and application layer for real-time analysis of equipment operating status, prediction of production efficiency, and generation of optimization suggestions; regularly retrain the model with new data to ensure prediction accuracy and applicability.

7. A frame production information management method according to claim 1, characterized in that: The production plan optimization includes: Production resource status monitoring: Real-time acquisition of equipment operating status, material inventory levels, and production work order execution status. Analyze equipment availability and material sufficiency through the data processing layer to provide data support for production plan adjustments. Task priority adjustment: Dynamically sort production tasks based on production demand priority and order delivery time; Dynamic equipment task allocation: When a device predicts a risk of failure or enters maintenance status, the system automatically allocates its tasks to other available devices; combining equipment performance data and task requirements to optimize equipment task allocation to improve production efficiency; Material scheduling optimization: Dynamically generate material procurement plans or internal transfer plans by analyzing material inventory data and production task requirements; adjust production sequences to avoid downtime when inventory is insufficient or logistics is delayed; Personnel scheduling management: intelligently match tasks and personnel according to the complexity of production tasks and the skill level of operators; when there is a shortage of personnel, generate temporary deployment suggestions to ensure the normal progress of production tasks; Real-time production adjustment and feedback: Combine equipment status monitoring with production order execution progress to dynamically adjust production plans and provide real-time feedback to production line operators; display the latest production plans and key adjustment points through a visual interface.

8. A frame production information management method according to claim 1, characterized in that: The quality inspection management and process optimization include: Quality inspection plan generation: Generate a quality inspection plan based on the production plan and process flow, clarify the frequency, time, quality inspection points and corresponding operators of quality inspection tasks; the quality inspection plan includes daily quality inspection and periodic quality inspection tasks, covering key processes and equipment; Quality inspection data collection and recording: The quality inspection result data, including product qualification rate, defective product type and distribution, are collected in real time through the quality inspection terminal equipment; the operator, quality inspection time and quality inspection process of each quality inspection are recorded, and the data is uploaded to the data processing layer for storage and analysis; Quality inspection data analysis: Use the data processing layer to analyze historical quality inspection data and identify production links and defect patterns with high non-conformity rates; use statistical analysis and machine learning algorithms to locate weak links in the process flow and provide data support for process optimization; Generation of process improvement suggestions: Generate process improvement suggestions based on the results of quality inspection data analysis, including adjusting process parameters, adding quality inspection points, optimizing process sequence, etc.; combine equipment status data and production plans to ensure the feasibility of process optimization suggestions; Real-time feedback and closed-loop improvement: The quality inspection data analysis results are fed back to the production line in real time, and closed-loop control of quality inspection management and process optimization is achieved by adjusting process parameters or reminding operators to pay attention to key issues; Quality inspection report generation: Generate periodic quality inspection reports, including product qualification rate statistics, defective product type distribution and process improvement records, to provide decision support for quality management.

9. A frame production information management method according to claim 1, characterized in that: The comprehensive data analysis includes: Data integration and correlation analysis: Integrate equipment operation status, material flow information, production work order data and quality inspection result data through the data processing layer; use multi-dimensional correlation analysis algorithms to identify the logical relationship between various data modules, including the impact of equipment status on production efficiency and quality inspection results, and the dependence of material flow on work order execution; Production efficiency analysis: Based on historical production data and real-time production progress data, analyze the task completion rate, equipment utilization rate and production bottleneck of the production line; identify the change pattern of production efficiency through trend analysis and provide targeted optimization suggestions; Equipment management analysis: Statistical analysis of equipment failure history records and operating status data is performed to generate equipment operation reliability reports; combined with energy consumption and maintenance cost data, quantitative support is provided for equipment replacement decisions to ensure optimal management of the equipment life cycle; Quality inspection data analysis and improvement: classify and count the quality inspection results, analyze the product qualification rate, defective product distribution and defect type; combine production and process data to generate quality improvement measures and optimize the production process; Personnel skills and efficiency assessment: Based on production work order records and quality inspection results, evaluate the operator's task completion efficiency and quality compliance rate; generate personnel training needs reports and provide skills improvement suggestions based on task matching; Multi-dimensional data report generation: Based on the integrated data analysis results, comprehensive reports including production efficiency, equipment utilization, quality inspection results, personnel skills and material consumption are generated; support daily, weekly and monthly multi-time dimension report generation, and display analysis results through a visual interface.

10. A frame production information management system, characterized in that: The system applies a frame production information management method as described in any one of claims 1 to 9, and comprises the following modules: Perception module: used to collect equipment operating parameters through sensors; collect material flow information through RFID technology or QR code scanning equipment, and record operators, equipment status and production progress data during the execution of production work orders; Data storage module: used to classify and store the collected equipment operation data, material flow data, production work order data and quality inspection data based on a distributed database; supports historical data archiving, multi-dimensional indexing and real-time data caching; Data processing module: used to clean, normalize, standardize and integrate the collected data; through data feature extraction and model input data generation, it supports the analysis needs of equipment failure prediction and production plan optimization; Production optimization model module: It includes feature extraction module, prediction module and decision-making module, which predicts the equipment operation status and production efficiency based on neural network technology, and generates comprehensive suggestions for task priority adjustment, equipment task allocation, material scheduling and process optimization; Planning management module: used to dynamically adjust the production plan according to the equipment operation status, material inventory information and work order execution progress, provide real-time feedback and optimize task allocation; Generate a plan adjustment plan when equipment failure or material shortage is detected; Quality inspection management module: used to generate quality inspection plans and record quality inspection results; By analyzing quality inspection data, generating quality inspection reports and process improvement suggestions, we can achieve closed-loop control of quality inspection management and process optimization; Data analysis and decision support module: used to integrate equipment status, production progress, material flow and quality inspection data, generate comprehensive reports on production efficiency, equipment utilization, quality inspection results and personnel skills, and provide optimization support for production bottleneck analysis, equipment replacement decisions and personnel training needs.

Images