Supplier production line quality control method and system based on MES system
By using the unified interface and quality prediction model of the MES system, the quality of supplier production line materials can be screened and analyzed in real time, which solves the problems of lag and information silos in supplier production line quality management, realizes efficient quality risk prediction and rapid response, and reduces management costs.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-15
- Publication Date
- 2026-04-10
AI Technical Summary
Existing supplier production line quality management methods suffer from problems such as delayed post-inspection, information silos and difficulties in traceability, and high management costs. This makes it difficult to recover losses when quality problems are exposed, and results in information delays and high management costs.
By accessing the supplier's MES production data in real time through a unified and standardized interface based on the MES system, material screening and quality analysis are performed. A pre-trained quality prediction model is used to predict potential quality risks. Sampling inspection and root cause analysis are conducted after the materials arrive at the buyer's MES, thereby improving sampling effectiveness and response speed.
By moving the quality management window forward to the supplier's production process, quality risks can be prevented in advance, management costs can be reduced, the transparency and responsiveness of the data traceability process can be improved, and losses can be minimized.
Smart Images
Figure CN121836464A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application relates to the quality control technical field, in particular to a supplier production line quality control method and system based on an MES system. BACKGROUND
[0002] Supplier production line quality management is a key link to ensure the stability of the supplier production process, the conformity of the product to the quality standard, and thus the reliability of the entire supply chain product quality and delivery. It covers planning, monitoring, improvement of the supplier production process, and cooperation with the supplier in quality management, and aims to reduce quality risks and improve the overall competitiveness of the supply chain through systematic management means. At present, the existing supplier production line quality management method has the following problems: 1. Hysteresis of post-inspection: relying on "incoming material sampling inspection" or "discovering problems after going online", the quality problem is exposed, and the loss is difficult to recover.
[0003] 2. Information silos and traceability difficulties: the internal production process of the supplier is a "black box" to the purchaser, and when problems occur, it relies on customer abnormal problem feedback or supplier voluntary reporting, often accompanied by information delay, distortion or concealment.
[0004] 3. Passive response management: more than 60% of the energy of the SQE team is used for abnormal firefighting, and the cost of supply chain quality management is high. SUMMARY
[0005] The main purpose of the embodiments of the application is to provide a supplier production line quality control method and system based on an MES system, which can solve the problems of hysteresis of post-inspection, information traceability difficulties and high management cost.
[0006] To achieve the above purpose, one aspect of the embodiments of the application provides a supplier production line quality control method based on an MES system, comprising the following steps: Based on the pre-constructed standardized interface, production data corresponding to each batch of materials on the supplier MES side is obtained; Before the material arrives at the purchaser MES side, quality analysis is performed on the material according to the production data, and the material is screened according to the quality analysis result to obtain screened material; When the screened material enters the purchaser MES side, the production data is input into a pre-trained quality prediction model for potential quality risk prediction, and a prediction risk level is output; After the screened material arrives at the purchaser MES side, sampling inspection is performed on the screened material according to the prediction risk level and the production data, and a sampling inspection result is obtained; When a quality problem occurs in the material after screening on the production line, a root cause analysis is performed based on the production data and the sampling inspection results to obtain the root cause of the quality problem.
[0007] In some embodiments, obtaining production data corresponding to each batch of materials from the supplier's MES based on a pre-built standardized interface specifically includes: Establish a unified, standardized interface between the supplier's MES and the purchaser's MES. Based on the standardized interface, obtain the production data corresponding to each batch of the material; The production data includes process parameters, inspection data, batch data, and raw equipment data.
[0008] In some embodiments, the production data includes process parameters and inspection data. The step of performing quality analysis on the material based on the production data, and screening the material based on the quality analysis results to obtain screened material, specifically includes: Define the blocking rules; Based on the process parameters and the inspection data, the material is subjected to quality analysis to obtain the quality analysis results; If the quality analysis results meet the interception rules, the corresponding material is intercepted to obtain the screened material, and a quality anomaly alarm is generated. The quality anomaly alarm information is sent to the supplier's MES terminal.
[0009] In some embodiments, the method further includes a step of pre-training the quality prediction model, wherein pre-training the quality prediction model specifically includes: Obtain historical datasets, which include several batches of historical materials, historical production data, historical material pass rates, and historical material defect types. Based on the historical production data, the historical material qualification rate, and the historical material defect type, the historical materials are labeled to obtain a risk level; The historical production data, the historical material qualification rate, and the historical material defect types are preprocessed to obtain a training set. The training set is input into a preset gradient boosting decision tree model for training, and the parameters of the gradient boosting decision tree model are adjusted according to the material sample pass rate and the material sample defect type to obtain the trained quality prediction model.
[0010] In some embodiments, the step of sampling and inspecting the screened material based on the predicted risk level and the production data to obtain sampling and inspection results specifically includes: Determine the predicted risk level corresponding to the screened material, wherein the predicted risk level includes a first risk, a second risk, and a third risk; When the predicted risk level corresponding to the screened material is the second risk or the third risk, the screened material is sampled and inspected according to the production data to obtain the sampling inspection results; The quality prediction model is optimized based on the sampling inspection results.
[0011] In some embodiments, the production data includes process parameters and inspection data, and the root cause analysis based on the production data and the sampling inspection results to obtain the root causes of quality problems specifically includes: Based on a preset virtual collaboration platform, the production data is acquired, including process parameters, inspection data, batch data, and raw equipment data. The process parameters, the test data, and the sampling test results are correlated to obtain the correlation analysis results. Based on the correlation analysis results, a problem analysis is performed on the batch data and the original equipment data to obtain the root cause of the quality problem.
[0012] In some embodiments, the production data includes batch data, and the method further includes: Based on the batch data, the production data is grouped to obtain several subgroups; Calculate the subgroup mean and subgroup range for each subgroup, and then calculate the short-term standard deviation based on the subgroup mean and subgroup range. The population mean is calculated based on the subgroup mean, and the population standard deviation is calculated based on the short-term standard deviation. The short-term process capability index is calculated based on the short-term standard deviation and the population mean, and the long-term process performance index is calculated based on the population standard deviation and the population mean. Determine the target process capability, and generate early warning information based on the short-term process capability index, the long-term process performance index, and the target process capability; The warning information is sent to the supplier's MES terminal.
[0013] To achieve the above objectives, another aspect of this application proposes a supplier production line quality control system based on an MES system, comprising: The first module is used to obtain production data corresponding to each batch of materials from the supplier's MES based on a pre-built standardized interface; The second module is used to perform quality analysis on the materials based on the production data before the materials arrive at the purchasing party's MES terminal, and to screen the materials based on the quality analysis results to obtain screened materials. The third module is used to input the production data into a pre-trained quality prediction model to predict potential quality risks when the screened material enters the purchasing party's MES terminal, and output the predicted risk level. The fourth module is used to perform sampling inspection on the screened material after it arrives at the purchaser's MES terminal, based on the predicted risk level and the production data, and obtain the sampling inspection results. The fifth module is used to perform root cause analysis based on the production data and the sampling inspection results when a quality problem occurs in the material after screening, in order to obtain the root cause of the quality problem.
[0014] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described above.
[0015] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer program product, including a computer program that, when executed by a processor, implements the aforementioned method.
[0016] The embodiments of this application include at least the following beneficial effects: The supplier production line quality control method and system based on the MES system of this application accesses the production data of the supplier's MES terminal in real time through a unified standardized interface. Before the materials arrive at the buyer's MES terminal, the materials are screened to prevent materials that do not meet the quality requirements from being put on the shelf. When the screened materials enter the buyer's MES terminal, the quality of the materials is predicted by a quality prediction model, and the predicted risk level is output. After the screened materials arrive at the buyer's MES terminal, the corresponding screened materials are selected for sampling inspection according to the predicted risk level to replace empirical sampling and improve the sampling effectiveness. When online quality problems occur in the screened materials, root cause analysis is performed based on previous production data to obtain the root cause of the quality problem. This application significantly advances the time window of supplier quality management from the traditional "post-inspection" to the supplier's "production process", which can prevent quality risks in advance, solve the problem of lag in post-inspection, and obtain the corresponding production data for root cause analysis when online quality problems occur. The response speed is fast, the data traceability process is simple and transparent, and the management cost is low. Attached Figure Description
[0017] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the embodiments of this application are described below. It should be understood that the drawings described below are only for the purpose of clearly illustrating some embodiments of the technical solutions in this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0018] Figure 1 A flowchart illustrating the steps of a supplier production line quality control method based on an MES system provided in one embodiment of this application; Figure 2 This application provides a schematic diagram of the structure of a supplier production line quality control system based on an MES system, as one embodiment of the present application. Figure 3 This is a schematic diagram of the hardware structure of an electronic device provided in one embodiment of this application. Detailed Implementation
[0019] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.
[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.
[0021] Supplier production line quality management is a crucial link in ensuring stable supplier production processes, product compliance with quality standards, and thus guaranteeing product quality and delivery reliability throughout the supply chain. It encompasses multiple aspects, including planning, monitoring, and improving supplier production processes, as well as collaborative quality management with suppliers. Its aim is to reduce quality risks and enhance the overall competitiveness of the supply chain through systematic management methods. Currently, existing supplier production line quality management methods suffer from the following problems: 1. The lag in post-production inspection: Relying on the "incoming material sampling inspection" or "problem discovered after going online" model, by the time quality problems are exposed, it often results in production line stoppage, rework, or customer complaints, and the losses are difficult to recover.
[0022] 2. Information silos and difficulties in traceability: The supplier's internal production process is a "black box" to the purchaser. When problems occur, the purchaser relies on customer feedback on abnormal issues or suppliers to report them proactively. This is often accompanied by information delays, distortions, or cover-ups.
[0023] 3. Reactive response management: The SQE team spends more than 60% of its energy on emergency response, resulting in high supply chain quality management costs.
[0024] In view of this, this application proposes a supplier production line quality control method based on a MES system. It accesses the supplier's MES production data in real time through a unified standardized interface. Before materials arrive at the buyer's MES, they are screened to prevent substandard materials from being shelved. When screened materials enter the buyer's MES, a quality prediction model predicts the material quality and outputs a predicted risk level. After the screened materials arrive at the buyer's MES, sampling inspection is performed on the corresponding screened materials according to the predicted risk level, replacing empirical sampling and improving sampling effectiveness. When online quality problems occur with the screened materials, root cause analysis is performed based on previous production data to obtain the root cause of the quality problem. This application significantly advances the supplier's quality management time window from the traditional "post-inspection" to the supplier's "production process," enabling early prevention of quality risks, solving the lag problem of post-inspection, and obtaining corresponding production data for root cause analysis when online quality problems occur. It offers fast response, simple and transparent data traceability, and low management costs.
[0025] Reference Figure 1 , Figure 1 This is a flowchart illustrating the steps of a supplier production line quality control method based on an MES system according to an embodiment of this application. This application proposes a supplier production line quality control method based on an MES system, which may include, but is not limited to, the following steps S101 to S105: Step S101: Based on the pre-built standardized interface, obtain the production data corresponding to each batch of materials from the supplier's MES terminal; Specifically, this application embodiment adds an interface protocol between the supplier's MES and the purchaser's MES, based on the existing full-process digital management and real-time control of the factory's production process. The purchaser's MES can be the factory or a warehouse.
[0026] As an optional implementation, step S101 can be further divided into the following steps S1011 and S1012: Step S1011: Establish a unified standardized interface between the supplier's MES and the purchaser's MES. Step S1012: Based on the standardized interface, obtain the production data corresponding to each batch of materials; The production data includes process parameters, inspection data, batch data, and raw equipment data.
[0027] In some optional embodiments, a unified interface protocol (such as OPC-UA, MQTT, etc.) is established between the supplier's MES and the purchaser's MES to transmit real-time production data from the supplier's MES. After access, the supplier's part number is mapped to the factory's BOM / master data, and the supplier's production line / process is mapped to the factory's key quality control points. Real-time data streams from the supplier's MES are continuously received via a message queue to obtain real-time production data. The scope of the production data may include: Process parameters: Collect real-time sensor data directly related to the production process, including temperature, pressure, speed, and dimensional measurements, for subsequent analysis of the impact of parameters on quality; Inspection data: Automated inspection results of the supplier's final / critical processes, including test data and corresponding test results for dimensions, appearance, and function; Batch / Serialized Data: Data is summarized by batch, and information such as batch number, production time, and equipment number corresponding to the production data is marked to facilitate the tracing of abnormal batches; Raw equipment data: This includes data such as the operating status of key equipment and alarm information (used for anomaly tracing), which facilitates the purchaser to monitor the production status and output of the production process in real time.
[0028] Step S102: Before the materials arrive at the purchasing party's MES terminal, perform quality analysis on the materials based on production data, and screen the materials based on the quality analysis results to obtain screened materials. Specifically, before materials are en route or arrive at the buyer's MES system, in order to ensure that the quality of materials entering the buyer's MES system meets the standards, the production data transmitted back from the supplier's MES system is analyzed in real time for quality analysis. Based on the quality analysis results, materials are screened to prevent materials with abnormal conditions from being put into the warehouse and put on the shelves.
[0029] As an optional implementation, the step of performing quality analysis on the materials based on production data and screening the materials based on the quality analysis results to obtain screened materials can be further divided into the following steps S1021 to S1024: Step S1021: Determine the blocking rules; Specifically, firstly, multi-dimensional interception rules are preset. These rules can be set in combination with material characteristics, production process requirements, and historical quality data. For example, for the key dimensional parameters of machined parts, the interception rule can be set as "three consecutive test values exceed the tolerance range"; for the electrical performance of electronic components, the interception rules can be set as "insulation resistance value is 20% lower than the standard value" and "predicted yield is less than 95%"; for the process parameters of polymer materials, the interception rule can be set as "temperature, pressure, speed and other parameters exceed the control range during production"; for the appearance or outer packaging of optical components, electronic products, etc., the interception rules can be set as "the number or area of defects such as scratches, cracks, bubbles, and color differences exceeds the standard" and "damaged packaging, incorrect labeling, and lack of moisture-proof and shock-proof measures".
[0030] Step S1022: Based on process parameters and inspection data, perform quality analysis on the material to obtain the quality analysis results; Step S1023: If the quality analysis results meet the interception rules, the corresponding material is intercepted to obtain the screened material, and a quality anomaly alarm message is generated. Step S1024: Send the quality anomaly alarm information to the supplier's MES terminal.
[0031] In some optional embodiments, process parameters and inspection data transmitted back from the supplier's MES are analyzed in real time, and the MES system automatically compares and intercepts the rules. For example, if the detection value of a critical dimension of a batch of materials falls outside the control limit three times in a row, it is determined that the dimension has a trend of deviation; if the predicted yield is lower than the threshold after calculation based on historical data and current parameters, the batch is marked as high-risk material.
[0032] When the monitoring results show that the quality analysis meets the interception rules, the MES system is automatically triggered to lock the batch of materials. An interception command is sent to the Warehouse Management System (WMS) through the MES system interface to prevent the materials from being put into storage and to avoid non-conforming materials from flowing into the production line. A quality anomaly alarm is generated, which includes the material batch number, supplier name, abnormal parameter name, test value, specification requirements, deviation degree, etc., and is accompanied by the original test data and process parameter curve. The quality anomaly alarm information is pushed to the SQE team and supplier in real time, requiring the supplier to provide an analysis report and temporary containment measures.
[0033] Step S103: When the screened material enters the purchaser's MES terminal, the production data is input into the pre-trained quality prediction model to predict potential quality risks and output the predicted risk level. It should be noted that the embodiments of this application combine historical data and real-time production data. When materials are inspected upon arrival at the factory or put into production, a pre-trained quality prediction model is used to analyze the correlation between key parameters and predict the potential quality risks (such as pass rate and probability of key defect types) of the batch of materials during the inspection upon arrival at the factory or the production line. Based on the prediction results, the batch risk level is automatically classified, which can improve the efficiency and accuracy of risk level classification.
[0034] As an optional implementation, the supplier production line quality control method based on the MES system further includes a step of pre-training a quality prediction model. This step of pre-training the quality prediction model can be further divided into the following steps S1031 to S1034: Step S1031: Obtain historical datasets, which include historical materials from several batches, historical production data, historical material pass rates, and historical material defect types. Step S1032: Based on historical production data, historical material qualification rate, and historical material defect type, mark the historical materials to obtain the risk level; Step S1033: Preprocess historical production data, historical material qualification rate, and historical material defect types to obtain a training set; Step S1034: Input the training set into the preset gradient boosting decision tree model for training, and adjust the parameters of the gradient boosting decision tree model according to the material sample pass rate and material sample defect type to obtain the trained quality prediction model.
[0035] Specifically, first, a historical dataset is acquired. This dataset includes several batches of historical materials, corresponding historical production data (such as parameters like temperature, pressure, and speed), historical material pass rates, and types of defects present in the materials. Next, the historical materials are categorized into risk levels based on the historical production data, material pass rates, and defect types. For example, if a batch of materials experiences significant parameter fluctuations and a low pass rate during production, along with multiple serious defect types, then this batch of materials may be categorized as high-risk. Conversely, if the parameters are stable, the pass rate is high, and the defect types are few or minor, it may be categorized as low-risk.
[0036] Then, to improve data quality and usability, historical production data, material qualification rates, and defect types are preprocessed to eliminate noise, outliers, and missing values. For production data, standardization or normalization is performed to allow data of different dimensions to be compared and analyzed on a unified scale. For material qualification rates, they are converted into a more suitable numerical form for model processing. For defect types, one-hot encoding or other methods can be used to convert them into numerical features. After preprocessing, a well-structured and reliable training set is obtained, preparing the model for training.
[0037] Finally, the preprocessed training set is input into a pre-defined Gradient Boosting Decision Tree (GBDT) model for training. The GBDT model improves the predictive performance of the model by constructing multiple decision trees and weighting them, and can flexibly handle various types of data and features. Therefore, this embodiment uses the GBDT model as the base model for training. During training, the model continuously adjusts its parameters and structure based on the data in the training set to minimize prediction errors. Simultaneously, the model parameters are adjusted according to the pass rate and defect type of the material samples. For example, if the model performs poorly when predicting materials with low pass rates, the learning rate and tree depth are adjusted to improve the model's performance in this case; for different types of defects, the model's focus on relevant features is adjusted to improve the ability to identify specific defects. After multiple iterations of training and parameter adjustments, a well-trained quality prediction model is finally obtained. This quality prediction model is used to predict the potential quality risks of materials during incoming inspection or production, and outputs the predicted risk level (high risk, medium risk, and low risk) for each material.
[0038] Step S104: After the screened material arrives at the purchaser's MES terminal, the screened material is sampled and inspected based on the predicted risk level and production data to obtain the sampling inspection results.
[0039] It should be noted that traditional empirical sampling mainly relies on the long-term accumulated work experience of inspection personnel to conduct sampling work, including simple random sampling, stratified sampling, and cluster sampling. This approach suffers from problems such as excessive subjectivity, low sampling effectiveness (oversampling or undersampling), and difficulty in quantifying risk. Therefore, in this embodiment, after the screened materials arrive at the purchasing party's MES, based on the predicted risk level output by the aforementioned quality prediction model and real-time monitoring production data, materials with higher risk levels are selected for sampling inspection. This can accurately locate problematic materials, prevent oversampling or undersampling, and improve sampling effectiveness.
[0040] As an optional implementation, the step of sampling and inspecting the screened material based on the predicted risk level and production data to obtain the sampling and inspection results can be further divided into the following steps S1041 to S1043: Step S1041: Determine the predicted risk level of the screened material. The predicted risk level includes the first risk, the second risk, and the third risk. Step S1042: When the predicted risk level of the screened material is the second or third risk, the screened material is sampled and inspected according to the production data to obtain the sampling inspection results. Step S1043: Optimize the quality prediction model based on the sampling inspection results.
[0041] Specifically, the predicted risk level of the screened material is first determined, including three levels: low risk (first risk), medium risk (second risk), and high risk (third risk). If the predicted risk level of a batch of screened material is either medium risk (second risk) or high risk (third risk), it indicates that the batch is more likely to have quality problems, and therefore, further sampling inspection of this batch is prioritized. The sampling plan or inspection items for Incoming Quality Control (IQC) are dynamically adjusted through the Quality Management System (QMS) module in the MES system. Furthermore, the obtained sampling inspection results are automatically fed back to the system for calibrating and optimizing the parameters or structure of the quality prediction model.
[0042] It should be noted that through this dynamic closed-loop management approach, the MES system can continuously learn and adapt to the actual situation on the production line, improving the accuracy and reliability of quality prediction. Simultaneously, it can dynamically adjust the inspection plan based on sampling inspection results, ensuring that high-risk materials receive sufficient attention and inspection, thereby effectively reducing the probability of quality problems and improving the overall product quality level.
[0043] Step S105: When a quality problem occurs in the material after screening on the production line, root cause analysis is performed based on production data and sampling inspection results to obtain the root cause of the quality problem.
[0044] It should be noted that existing supplier quality management follows the "2485 principle," with traditional problem response and handling times ranging from 24 to 72 hours. However, in this application's embodiment, after an online quality issue occurs, the production data of the problematic batch produced by the supplier can be directly reviewed through the MES system, accelerating root cause analysis and identifying the root cause of the quality problem. Through real-time monitoring with the supplier's system, this application's embodiment can reduce the response time for handling abnormal issues to ≤2 hours, enabling rapid response to abnormal situations, and the data traceability process is simple and transparent.
[0045] As an optional implementation, the step of conducting root cause analysis based on production data and sampling inspection results to obtain the root cause of the quality problem can be further divided into the following steps S1051 to S1053: Step S1051: Based on the preset virtual collaboration platform, acquire production data, which includes process parameters, inspection data, batch data, and raw equipment data; Step S1052: Perform correlation analysis on process parameters, test data, and sampling test results to obtain correlation analysis results; Step S1053: Based on the correlation analysis results, perform problem analysis on the batch data and original equipment data to obtain the root causes of quality problems.
[0046] Specifically, after a material quality issue occurs on the production line, the MES system generates an electronic quality report with precise timeframes, problem parameters, relevant batches, and original data from the supplier, which is automatically pushed to the corresponding supplier. Within the MES system platform or through a virtual collaboration platform integrated with PLM / SRM, a shared space is provided to view real-time / historical data and exchange problem analysis information. SQE engineers can directly review the complete process parameters and testing data from when the supplier produced the problematic batch through the virtual collaboration platform.
[0047] Furthermore, abnormal fluctuations in process parameters are correlated with anomalies in test data and sampling test results through correlation analysis. For example, if a temperature parameter rises abnormally during the period in which the problem occurs, and the strength test data of semi-finished products produced during that period is found to be low, it can be initially inferred that the temperature anomaly may be one of the reasons for insufficient product strength. Then, based on the results of the correlation analysis, the specific production process is traced back to the original equipment. The operating status of the production equipment and the operation records of the operators are reviewed based on the original equipment data during the period of temperature anomaly. It is checked whether there are any equipment malfunctions or improper maintenance, and whether the operators are operating according to standard operating procedures. For example, if a malfunction is found in the equipment's temperature control system, causing the temperature to be unable to be controlled normally, then the equipment malfunction is likely the direct cause of the problem.
[0048] In addition to the production process, other factors need to be investigated based on batch data. For example, has the raw material supplier changed batches or suppliers? Are the storage conditions of the raw materials meeting the requirements? Are they affected by environmental factors (such as humidity and temperature)? If it is found that the raw material supplier has recently changed batches, and a certain performance indicator of the new batch of raw materials differs from the previous batch, then the raw material quality problem may be the cause of this online quality problem.
[0049] As an optional implementation, the supplier production line quality control method based on the MES system further includes the following steps S106 to S111: Step S106: Based on the batch data, group the production data to obtain several subgroups; Step S107: Calculate the subgroup mean and subgroup range for each subgroup, and then calculate the short-term standard deviation based on the subgroup mean and subgroup range. Specifically, the MES system automatically collects production data uploaded by the supplier's MES system, calculates process capability indices (Cp / Cpk, Pp / Ppk) in real-time or near real-time, and monitors their changing trends. First, the production data is divided into several subgroups according to production time or batch number. Then, the subgroup mean is calculated using the following formula. range of subgroups : ; ; in, Indicates the first Subgroup mean of each subgroup Indicates the first Subgroup range of each subgroup This indicates the sample size of the subgroup.
[0050] Next, based on the subgroup mean Sum of subgroups The short-term standard deviation is calculated using the following formula. : ; in, Indicates short-term standard deviation, This represents the mean of the range. This represents the control chart constant.
[0051] Step S108: Calculate the population mean based on the subgroup mean, and calculate the population standard deviation based on the short-term standard deviation; Specifically, the overall mean is calculated by merging all subgroup data using the following formula. and population standard deviation : ; ; in, This represents the population mean. Indicates the population standard deviation. This indicates the total number of subgroups.
[0052] Step S109: Calculate the short-term process capability index based on the short-term standard deviation and the population mean, and calculate the long-term process performance index based on the population standard deviation and the population mean. Specifically, the short-term process capability index includes the process capability index Cp and the modified process capability index Cpk, which are calculated using the following formula: ; ; in, Indicates the upper limit of the specification. Indicates the lower limit of the specification.
[0053] Long-term process performance indices include the process performance index Pp and the modified process performance index Ppk, which are calculated using the following formula: ; .
[0054] Step S110: Determine the target process capability, and generate early warning information based on the short-term process capability index, the long-term process performance index, and the target process capability; Step S111: Send the warning information to the supplier's MES terminal.
[0055] Specifically, the target process capability can be set according to actual production needs. For example, the target process capability can be set as follows: process capability index Cp ≥ 1.33, modified process capability index Cpk ≥ 1.0, process performance index Pp ≥ 1.33, and modified process performance index Ppk ≥ 1.67. When any of the short-term process capability index or long-term process performance index deviates from the target process capability or the capability decreases, an automatic production warning is issued, alerting the SQE team and suppliers. This allows the purchasing side to monitor the process in real time and promote preventative improvements.
[0056] The above describes the supplier production line quality control method based on the MES system according to the embodiments of this application. It can be recognized that, compared with existing supplier production line quality management methods, the embodiments of this application have the following advantages: First, by combining historical data with real-time production data, a pre-trained quality prediction model can be used to analyze the correlation between key parameters during material inspection upon arrival or production on the line. This model can predict the potential quality risks of the batch of materials during inspection upon arrival or production on the line, and then automatically classify the batch risk level based on the prediction results, thereby improving the efficiency and accuracy of risk level classification.
[0057] Second, after the screened materials arrive at the purchaser's MES, the materials with higher risk levels are selected for sampling inspection based on the predicted risk level output by the quality prediction model. This can accurately locate problematic materials, prevent over-sampling or under-sampling, and improve sampling effectiveness.
[0058] Third, after an online quality issue occurs, the production data of the supplier when producing the problematic batch can be directly viewed through the virtual collaboration platform in the MES system. The response speed is fast, and the data traceability process is simple and transparent.
[0059] Reference Figure 2 This application also provides a supplier production line quality control system based on an MES system, including: The first module is used to obtain production data corresponding to each batch of materials from the supplier's MES based on a pre-built standardized interface; The second module is used to perform quality analysis on the materials based on production data before the materials arrive at the purchasing party's MES terminal, and to screen the materials based on the quality analysis results to obtain the screened materials. The third module is used to input production data into a pre-trained quality prediction model to predict potential quality risks when the screened material enters the buyer's MES terminal, and output the predicted risk level. The fourth module is used to conduct sampling inspections on the screened materials after they arrive at the purchaser's MES terminal, based on the predicted risk level and production data, and to obtain the sampling inspection results. The fifth module is used to perform root cause analysis based on production data and sampling inspection results when quality problems occur in the material after screening, in order to obtain the root cause of the quality problem.
[0060] It is understood that the content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.
[0061] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.
[0062] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.
[0063] Please see Figure 3 , Figure 3The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 1001 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 1002 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 1002 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1002 and is called and executed by the processor 1001 using the methods described in the embodiments of this application. Input / output interface 1003 is used to implement information input and output; The communication interface 1004 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 1005 transmits information between various components of the device (e.g., processor 1001, memory 1002, input / output interface 1003, and communication interface 1004); The processor 1001, memory 1002, input / output interface 1003 and communication interface 1004 are connected to each other within the device via bus 1005.
[0064] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.
[0065] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.
[0066] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.
[0067] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.
[0068] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
[0069] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.
[0070] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.
[0071] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.
[0072] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.
[0073] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0074] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.
[0075] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.
[0076] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0077] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0078] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0079] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.
Claims
1. A supplier production line quality control method based on an MES system, characterized in that, Includes the following steps: Based on a pre-built standardized interface, obtain the production data corresponding to each batch of materials from the supplier's MES terminal; Before the material arrives at the purchasing party's MES terminal, the material is analyzed for quality based on the production data, and the material is screened based on the quality analysis results to obtain screened material; When the screened material enters the purchasing party's MES terminal, the production data is input into a pre-trained quality prediction model to predict potential quality risks and output the predicted risk level. After the screened material arrives at the purchaser's MES terminal, the screened material is sampled and inspected according to the predicted risk level and the production data to obtain the sampling inspection results. When a quality problem occurs in the material after screening on the production line, a root cause analysis is performed based on the production data and the sampling inspection results to obtain the root cause of the quality problem.
2. The method according to claim 1, characterized in that, The process of obtaining production data for each batch of materials from the supplier's MES system based on a pre-built standardized interface specifically includes: Establish a unified, standardized interface between the supplier's MES and the purchaser's MES. Based on the standardized interface, obtain the production data corresponding to each batch of the material; The production data includes process parameters, inspection data, batch data, and raw equipment data.
3. The method according to claim 1, characterized in that, The production data includes process parameters and inspection data. The step of performing quality analysis on the material based on the production data, and screening the material based on the quality analysis results to obtain screened material, specifically includes: Define the blocking rules; Based on the process parameters and the inspection data, the material is subjected to quality analysis to obtain the quality analysis results; If the quality analysis results meet the interception rules, the corresponding material is intercepted to obtain the screened material, and a quality anomaly alarm is generated. The quality anomaly alarm information is sent to the supplier's MES terminal.
4. The method according to claim 1, characterized in that, The method further includes a step of pre-training the quality prediction model, wherein the pre-training of the quality prediction model specifically includes: Obtain historical datasets, which include several batches of historical materials, historical production data, historical material pass rates, and historical material defect types. Based on the historical production data, the historical material qualification rate, and the historical material defect type, the historical materials are labeled to obtain a risk level; The historical production data, the historical material qualification rate, and the historical material defect types are preprocessed to obtain a training set. The training set is input into a preset gradient boosting decision tree model for training, and the parameters of the gradient boosting decision tree model are adjusted according to the material sample pass rate and the material sample defect type to obtain the trained quality prediction model.
5. The method according to claim 1, characterized in that, The step of sampling and inspecting the screened material based on the predicted risk level and the production data to obtain the sampling and inspection results specifically includes: Determine the predicted risk level corresponding to the screened material, wherein the predicted risk level includes a first risk, a second risk, and a third risk; When the predicted risk level corresponding to the screened material is the second risk or the third risk, the screened material is sampled and inspected according to the production data to obtain the sampling inspection results; The quality prediction model is optimized based on the sampling inspection results.
6. The method according to claim 1, characterized in that, The production data includes process parameters and inspection data. The root cause analysis based on the production data and the sampling inspection results to obtain the root causes of quality problems specifically includes: Based on a preset virtual collaboration platform, the production data is acquired, including process parameters, inspection data, batch data, and raw equipment data. The process parameters, the test data, and the sampling test results are correlated to obtain the correlation analysis results. Based on the correlation analysis results, a problem analysis is performed on the batch data and the original equipment data to obtain the root cause of the quality problem.
7. The method according to any one of claims 1 to 6, characterized in that, The production data includes batch data, and the method further includes: Based on the batch data, the production data is grouped to obtain several subgroups; Calculate the subgroup mean and subgroup range for each subgroup, and then calculate the short-term standard deviation based on the subgroup mean and subgroup range. The population mean is calculated based on the subgroup mean, and the population standard deviation is calculated based on the short-term standard deviation. The short-term process capability index is calculated based on the short-term standard deviation and the population mean, and the long-term process performance index is calculated based on the population standard deviation and the population mean. Determine the target process capability, and generate early warning information based on the short-term process capability index, the long-term process performance index, and the target process capability; The warning information is sent to the supplier's MES terminal.
8. A supplier production line quality control system based on an MES system, characterized in that, include: The first module is used to obtain production data corresponding to each batch of materials from the supplier's MES based on a pre-built standardized interface; The second module is used to perform quality analysis on the materials based on the production data before the materials arrive at the purchasing party's MES terminal, and to screen the materials based on the quality analysis results to obtain screened materials. The third module is used to input the production data into a pre-trained quality prediction model to predict potential quality risks when the screened material enters the purchasing party's MES terminal, and output the predicted risk level. The fourth module is used to perform sampling inspection on the screened material after it arrives at the purchaser's MES terminal, based on the predicted risk level and the production data, and obtain the sampling inspection results. The fifth module is used to perform root cause analysis based on the production data and the sampling inspection results when a quality problem occurs in the material after screening, in order to obtain the root cause of the quality problem.
9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method of any one of claims 1 to 7.
10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.