What Is Statistical Process Control – Application Of SPC In Process Quality Control

What Is Statistical Process Control

Statistical process control (SPC) technology is a modern process quality management technology integrating production technology and scientific management. In the production process of enterprises, the control of product quality has gone through the following seven steps:
① investigate and understand the final quality of products.
② Analyze individual factors that significantly affect the quality.
③ Control the changes of these factors by manual adjustment.
④ Automatically control these influencing factors.
⑤ The law of continuous change of production effect is obtained by statistical method.
⑥ Comprehensively investigate and understand other secondary factors affecting quality.
⑦ Considering all factors, a complete theoretical model is established to control the whole production process and obtain 100% high-quality and qualified products.

SPC applies statistical principles, implements online trend control, realizes quantitative quality management, provides quality analysis tools and platforms, and finds out the end causes by factor analysis, providing a modern scientific method for solving deep-seated problems, so as to improve the level of fine manufacturing.

The implementation of SPC system by a company can be summarized in the following four aspects:
1. Need of SPC System Application
In the production process of enterprises, the fluctuation of product processing size is inevitable. It is caused by the fluctuation of basic factors such as people, machines, materials, methods and environment. There are two kinds of fluctuation: normal fluctuation and abnormal fluctuation. Normal fluctuations are caused by accidental (unavoidable) factors, which have little impact on product quality, and are difficult to eliminate technically and economically. Abnormal fluctuations are caused by systematic reasons, which have a great impact on product quality, but can be avoided and eliminated by taking measures. The purpose of process control is to eliminate and avoid abnormal fluctuations and keep the process in a normal fluctuation state.

Statistical process control is a process control tool with the help of mathematical statistics. It analyzes and evaluates the production process, timely finds signs of systematic factors according to feedback information, and takes measures to eliminate their effects, so as to maintain the process in a controlled state only affected by random factors, so as to achieve the purpose of quality control. When the process is only affected by random factors, the process is under statistical control. When there are systematic factors in the process, the process is in a statistical runaway state. Due to the statistical regularity of process fluctuations, when the process is controlled, the process characteristics generally follow a stable random distribution. When out of control, the process distribution will change. SPC uses the statistical regularity of process fluctuations to analyze and control the process. Therefore, it emphasizes that the process runs under control and capability, so that the product and service quality can meet the customer’s requirements stably.

The process of implementing SPC is generally divided into two steps: first, use SPC tools to analyze the process, such as drawing control charts for analysis, etc., and take necessary measures according to the analysis results: it may be necessary to eliminate systematic factors in the process, or it may be necessary to involve the management to reduce random fluctuations in the process to meet the requirements of process capability. The second step is to use the control chart to monitor the process.

Control chart is the most important tool in SPC. At present, the traditional control chart based on Shewhart principle is widely used in practical work, but the control chart is not limited to this. In recent years, some advanced control tools have been gradually developed, such as EWMA and CUSUM control charts for monitoring small fluctuations, proportional control charts and target control charts for controlling small batch and multi variety production processes; Control chart for controlling multiple quality characteristics.

SPC is very suitable for repetitive production processes. It can help us make a reliable assessment of the process. Determine the statistical control limit of the process, and judge whether the process is out of control and capable. Provide an early warning system for the process, and monitor the process in time to prevent the occurrence of waste products. Reducing the dependence on routine inspection, regular observation and systematic measurement methods replace a large number of inspection and verification work.

2. Need For Data Collection
SPC system supports multiple data acquisition modes, including offline acquisition, online acquisition and automatic acquisition. The preparation of the quality characteristic data collection plan fully takes into account the following contents: product batch size, difficulty in obtaining data, sampling cost, frequency of data collection, selection of appropriate statistical tools (such as the type of control chart), time sequence of data collection, etc, For this reason, we have selected the following two data collection modes:

① Offline acquisition
– Enter data at one time in the data entry interface provided by the system.
– Import the data in the fixed EXCEL format provided by the system through the collected and sorted quality data.

② Automatic acquisition
In order to save labor cost and reduce workload, we use SPC system to collect real-time data from other CNC equipment through interface program. The time interval of automatic data collection is set according to the actual production situation. The automatic acquisition process is realized through the following two different acquisition methods:

– CNC equipment operator shall inspect the size, shape and position tolerance and other information of the parts to be inspected according to the requirements of part inspection, and save the files generated by the inspection to the directory designated by the CNC equipment computer. Read the files under the directory specified by the CNC equipment computer through the automatic collection interface program for analysis, and store the parsed data such as parameter number and data value into the quality management information system.
– The CNC program is collected through the edited and solidified quality characteristic parameters on the CNC machine tool, and the operator of the CNC machine tool transmits the part quality data information to the specific server through the network. The received NC program file is parsed by the data acquisition interface application program of the specific server of the data machine tool. The interface program stores the parsed data into the database of the quality management information system through the network, and generates a data log for query.

3. The Need For Process Monitoring

The production characteristics of a company are: large production tasks, it is difficult to monitor the production process uniformly. There are many product quality data, and it is difficult to collect, maintain and retrieve the data; Therefore, it is very necessary to monitor the key processes in the product production process, so as to control the key links in the product quality formation process. Therefore, the monitoring plan for process parameters of multiple key processes, including machining and special processes, has been determined to solve the problem of collecting factory quality data and monitoring process capability indicators such as standard deviation of key processes and CPK.

Process monitoring is a window for process quality monitoring provided by company leaders and quality management personnel. The process monitoring methods mainly include factory monitoring and process monitoring. You can visually see the quality status of each workshop and process, and compare the control charts of related process parameters. The monitoring chart can be used to query and trace the actual quality data, control chart and relevant statistical indicators for exception reporting and correction.

① Define plant wide monitoring
Through the set monitoring scheme, you can view the monitoring parameter layout and scenarios of companies or workshops covered by the scheme. For the set monitoring point, the actual quality status of the position is displayed by the color of the warning light.

Through the SPC monitoring interface, the green indicator light indicates that the product quality parameters produced in the current process are normal, and the red indicator light indicates that the product quality parameters produced in the current process are abnormal. When the red abnormal points are analyzed and closed loop handled, the indicator light will turn blue, indicating that the abnormality has been handled. After clicking the red warning light, the abnormal data of the monitoring point can be displayed. The system automatically links to the statistical analysis and abnormal improvement part, and can call to view the control chart and detailed data analysis. At the same time, it provides a management information platform for the company’s quality managers to understand the quality status of the enterprise’s production site in real time, making the quality status of the production process highly transparent.

② Define process monitoring
Process monitoring is defined to enable simultaneous monitoring of multiple related quality characteristic parameters. Control charts related to processes, products or a group of quality parameters are set on the same monitoring interface for simultaneous viewing, and the change process of different parameters and indicators is compared and analyzed.

4. The Need For Statistical Analysis

Through SPC system, the process analysis, control chart analysis, histogram analysis, trend chart analysis and process capability analysis of machined product parts can be carried out. It can also analyze the quality indicators of batches by product, workshop, process and other dimensions, such as trend chart analysis and comparative analysis by year, month and week. The fluctuation between batches of product parts can be seen intuitively.

In the heat treatment and forming workshop, the process data can be acquired automatically, and the actual process parameter graph can be drawn using statistical analysis. By comparing with the process required parameter graph, it is used to analyze the implementation compliance of key parameters in the heat treatment, forming and processing processes.

These analysis tools are applied in different stages of the SPC system, focusing on different aspects to ensure the realization of the goals of the SPC system. At the same time, the analysis report can be formed and exported through the analysis of data.

① Application in process analysis
By analyzing the process, the system automatically calculates the mean value, standard deviation, process capability index Cp, nonconforming product rate p, etc. of the quality characteristic, draws analysis conclusions, and can derive the sampling analysis report of the quality characteristic.

② Analysis and Application of Control Chart
The control chart can be used to analyze and judge whether the production process is stable, find abnormal conditions in the production process in time and prevent the occurrence of nonconformities.

– Control Chart Analysis: According to the selection criteria such as quality characteristics and date range, the system can automatically call the control chart and auxiliary information corresponding to the quality characteristics. Because the change modes of the points on the control chart are different, the reasons for the variation are also different. According to the change modes of the parameter points on the control chart, the reasons for the abnormality can be analyzed from the aspects of 5M1E (i.e. personnel, equipment, materials, methods, measurement, and environment).

– Histogram Analysis: Histogram analysis is a statistical chart reflecting the distribution state and fluctuation law of product quality data. Its main purpose is to judge the stability of the process, infer the satisfaction of the process quality specifications, analyze the impact of different factors on the quality, and calculate the process capability.

– Platonic Analysis: Plato is a simple and effective tool for solving quality problems. Rank various potential regional variation sources according to their impact on the total variation, and focus on a few key causes while ignoring most unimportant ones.
– Scatter Diagram Analysis: In the actual production process, there are many process parameters and product quality indicators with complex relationships, which are both interrelated and mutually restricted, and may or may not have strong correlation. The scatter diagram can reveal the relationship between them intuitively and effectively. By drawing a scatter chart, the complex data between parameters and quality indicators become points on the coordinate map, and their correlation is clearly displayed.

– Trend Chart Analysis: The fluctuation of quality indicators can be observed through the trend chart, and targeted measures can be taken in time by analyzing the indicators.

– Process Capability Analysis: Process capability refers to the actual processing capability of the process to stably produce qualified products under the normal state of 5M1E. The process capability depends on the precision of machinery and equipment, materials, processes, process equipment, work quality of workers and other technical conditions. The process capability index is represented by Cp and Cpk.

Process capability indicators are applied to:
a) Select an economical and reasonable process plan.
b) Coordinate the relationship between processes.
c) Verify the process quality assurance capability.

5. Need For Abnormal Improvement
The abnormal information detected by the SPC system needs to be maintained and handled in a timely manner, including the query, processing, classification and statistics of the abnormal information and the formation of the improvement status report.

For major abnormal phenomena in process monitoring, it can be handled in three stages: cause analysis, measure formulation and effect verification according to PDCA cycle mode.

① Abnormal information
Abnormal information is the data source for analyzing the causes of abnormalities in product quality improvement, so it is particularly important to effectively manage these data. SPC system can display all abnormal information according to the occurrence time and processing status of the problem, and set specific exception handling sheet, which can facilitate relevant management personnel to query the abnormal information of relevant quality characteristics as required.

② Exception handling
The SPC system can track and handle major quality abnormality information, and complete the handling of abnormal problems through cause analysis, measure correction and effect evaluation.

③ Processing status reminder method
In order to effectively remind and timely deal with the abnormal information in process monitoring, the easy judgment of colors in visual management is used to clearly mark the processing status of abnormal information through three colors: green, yellow and red: red indicates that the feedback is not processed, yellow indicates that the feedback is overdue, and green indicates that the feedback is processed according to normal progress.

④ Quality improvement effect evaluation
The SPC system uses the comparative analysis method of the data before and after the improvement to evaluate the effect. The main comparative factors are the change of the process capability index, the trend change of the conventional quality index, and the change of the number of process exceptions. Due to the non reproducibility of many processes, to some extent, it is necessary to artificially quantify the criteria for evaluating measures. The measures confirmed to be effective can be included in the standard, that is, the control standard for modifying indicators and parameters. The system will control according to the new standard. At this time, the process has reached a steady state, that is, the analysis stage of SPC system can be transferred to the control stage according to this standard.

As an important tool for quality improvement, SPC is not only applicable to military manufacturing, but also to all process areas such as service. At the initial stage of process quality improvement, SPC can help identify opportunities for improvement. After the improvement stage is completed, SPC can be used to evaluate the effect of improvement and maintain the improvement results, and then further carry out quality improvement at a new level to achieve a stronger and more stable process capability.