阅读量：137
日期：2023-10-23
In the production process, the fluctuation of the processing size of the product is inevitable. It is caused by the fluctuation of basic factors such as people, machines, materials, methods and environment. There are two types of volatility: normal volatility and abnormal volatility. Normal fluctuations are caused by accidental factors (unavoidable factors). It has little impact on product quality, is technically difficult to eliminate, and is not worth eliminating economically. Abnormal fluctuations are caused by systemic causes (abnormal factors). It has a great impact on product quality, but measures can be taken to avoid and eliminate it. The purpose of process control is to eliminate and avoid abnormal fluctuations, so that the process is in a normal state of fluctuation.
Statistical process control (SPC) is a process control tool with the help of mathematical statistics. It analyzes and evaluates the production process, finds the symptoms of systemic factors in time according to the feedback information, and takes measures to eliminate their effects, so that the process is maintained in a controlled state only affected by random factors, in order to achieve the purpose of quality control. It holds that when the process is only affected by random factors, the process is in a statistical control state (referred to as controlled state). When there are systematic factors in the process, the process is in a statistical out-of-control state (referred to as out-of-control state). Because of the statistical regularity of process fluctuation, when the process is controlled, the characteristics of the process generally obey the stable random distribution. When out of control, the process distribution will change. SPC makes use of the statistical regularity of process fluctuation to analyze and control the process. Therefore, it emphasizes the operation of processes in a controlled and capable state, so that products and services can steadily meet customer requirements.
The process of implementing SPC is generally divided into two steps: First, SPC tools are used to analyze the process, such as drawing control charts for analysis; Take necessary actions based on the results of the analysis: it may be necessary to eliminate systemic factors in the process, or it may be necessary to intervene with management to reduce random fluctuations in the process to meet the needs of the process capability. The second step is to monitor the process with a control chart.
The control chart is the most important tool in SPC. At present, the traditional control chart based on Shewhart principle is widely used in practice, 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 multi-variety production processes. Control chart for controlling multiple quality characteristics. (The introduction and application of relevant control charts will be launched in the future)
SPC originated in the 1920s, marked by the control chart invented by Dr. Shewhart in the United States. Since its inception, that is, in the industrial and service industries have been promoted and applied, since the 1950s, SPC has played a vital role in the rise of Japanese product quality. After the 1980s, many large companies in the world have actively promoted the application of SPC within themselves, and have also put forward corresponding requirements for suppliers. In ISO9000 and QS9000 also put forward the application of SPC method in production control requirements, SPC is very suitable for repetitive production process.
It can help us:
1. Make a reliable assessment of the process.
2. Determine the statistical control limits of the process to determine whether the process is out of control and whether the process is capable.
3. Provide an early alarm system for the process to monitor the situation of the process in time to prevent the occurrence of waste.
4. Reduce the dependence on routine inspection, regular observation and systematic measurement methods replace a lot of detection and verification work.
As an important tool for quality improvement, SPC is not only suitable for industrial engineering, but also for all process fields such as service. In the early stages of process quality improvement, SPC can help identify opportunities for improvement, after the completion of the improvement phase, SPC can be used to evaluate the effect of the improvement and maintain the improvement results, and then further improve the work at a new level to achieve a stronger and more stable working capacity.
Everything can be seen as a process. The output of the process is variable due to the change of the input, the influence of various interference factors and the possible interaction between the wave sources. In most cases, this instability of the output will cause us a lot of distress and even loss. What are the factors that significantly influence fluctuations in output? Under what conditions can the output be controlled within the ideal range? Experimental design can help us solve this mystery.
Experimental design is a mathematical theory and method based on probability theory and mathematical statistics to economically and scientifically develop experimental plans for effective statistical analysis of experimental data. The basic idea was put forward by British statistician R.A.Fisher in his field experiment. In the experiment, he found that the environmental conditions were difficult to strictly control, and the random error could not be ignored. Therefore, he proposed that the experimental scheme must be rationally arranged to make the experimental data have a suitable mathematical model, so as to reduce the influence of random error and improve the accuracy and reliability of the experimental conclusions. This is the basic idea of experimental design.
The process of experimental design can be divided into two parts: the design of experimental scheme and the data analysis of experimental results. The design of experimental scheme includes determining experimental index, selecting factor, determining factor level, establishing mathematical model of experimental index and designing experimental scheme. There are many kinds of experimental design methods, but in order to improve the accuracy and reliability of the experiment, three basic principles must be followed: the principle of randomization, the principle of repetition and the principle of local control. The data analysis of experimental results is the application of linear algebra, probability theory and mathematical statistics and other mathematical tools to analyze and process the experimental data, including fitting the model, testing the model, calculating the experimental statistics and explaining the experiment process.
In practical applications, experimental design can solve the following problems:
1. Arrange experiments scientifically and reasonably, so as to reduce the number of experiments, shorten the experiment period, and improve economic benefits.
2. Find out the main factors affecting the output from the many influencing factors.
3. Analyze the impact of interaction between influencing factors.
4. Analyze the influence of experimental errors to improve experimental accuracy.
5. Find out the optimal parameter combination, and through the analysis and comparison of the experimental results, find out the direction of further experiments to achieve the optimal scheme.
6. Predict the output value of the best solution.
In the 1930s and 1940s, Britain, the United States, the Soviet Union and other countries carried out further research on the experimental design method, and gradually extended it to the field of industrial production, and it has been applied in metallurgy, construction, textile, machinery, medicine and other industries. During World War II, the United Kingdom and the United States adopted experimental design method in industrial experiments and achieved remarkable results. After the war, Japan introduced it as one of the management techniques from the United Kingdom and the United States, which played a role in promoting its economic recovery. Today, experimental design has become a common technique for Japanese business people, engineers, researchers, and managers. In the 1950s, Dr. Kenichi Taguchi proposed the SNR experimental design for reference to the experimental design method, and gradually developed the Taguchi method based on the mass loss function and cubic design. In the 1980s, Taguchi's method entered the United States and gained widespread attention. Nowadays, the application field of experimental design technology has broken through the traditional industrial process improvement and product design, and has widely penetrated into the application of commercial layout, commercial display, advertising design and product packaging.
The experimental design was studied and popularized in China in the 1960s, and Taguchi method was introduced in the 1980s, and certain results were achieved. But experimental design, as a powerful weapon of quality improvement, has not yet exerted its full power.
When designing and manufacturing products, there are usually three lines of defense to control defects: avoiding or eliminating the cause of the failure, identifying or detecting the failure in advance, and reducing the impact and consequences of the failure. FMEA is an effective tool to help us nip defects in the bud from the first line of defense.
FMEA is an important method for reliability design. It is actually a combination of FMA(Failure Mode Analysis) and FEA(Failure Impact Analysis). It evaluates and analyzes possible risks in order to eliminate them or reduce them to an acceptable level on the basis of existing technology. Timeliness is one of the most important factors in the successful implementation of FMEA, it is an "act before", not an "act after". For maximum benefit, FMEA must be performed before failure modes are incorporated into the product.
FMEA is actually a set of serialized activities, the process of which includes: identifying potential failure modes in the product/process; According to the corresponding evaluation system, the risk quantitative evaluation is carried out for the identified potential failure modes. List the cause/mechanism of the failure and look for prevention or improvement measures. Because product failures can be related to design, manufacturing process, use, contractor/supplier, and service, FMEA is subdivided into four categories: Design FMEA, Process FMEA, Use FMEA, and service FMEA. Design FMEA and process FMEA are most commonly used.
The design FMEA(also d-FMEA) shall begin at or before the formation of a design concept, and shall be continuously modified in a timely manner as the design changes or other information becomes available during all stages of product development, and shall end before the drawing is completed. The object of evaluation and analysis is the final product and each system, subsystem and component associated with it. It should be noted that the d-FMEA should also ensure that the manufacturing or assembly can achieve the design intent while reflecting the design intent. Therefore, although d-FMEA does not rely on process control to overcome the defects in the design, it can consider the technical/objective limitations of the manufacturing/assembly process, thus providing a good basis for process control.
Conducting d-FMEA helps to:
1. Balance between design requirements and design schemes.
2. Initial design of manufacturing and assembly requirements.
3. Increase the likelihood of considering potential failure modes and their impact on systems and products in the design/development process.
4. Provide more information for the formulation of comprehensive and effective design test plans and development projects.
5. Establish a set of priority control systems to improve design and development testing.
6. Provide reference for future analysis and study of site conditions, evaluation of design changes and development of more advanced designs.
Process FMEA(also p-FMEA) should be initiated prior to tooling preparation for production, at or before the process feasibility analysis stage, and all manufacturing processes from individual parts to assemblies should be considered. The object of evaluation and analysis is all new parts/processes, changed parts/processes and original parts/processes with changes in application or environment. It should be noted that although p-FMEA does not change the product design to overcome the process defects, it takes into account the product design characteristics parameters related to the planned assembly process in order to maximize the assurance that the product meets the requirements and expectations of the user.
p-FMEA generally includes the following:
1. Identify potential process failure modes associated with the product.
2. Evaluate the potential impact of the fault on users.
3. Identify the causes of potential manufacturing or assembly process failures, identify process control variables that reduce the occurrence of failures or identify failure conditions.
4. Compile the classification table of potential failure modes and establish the optimal selection system of corrective measures.
5. Document the manufacturing or assembly process.
The application of FMEA technology is developing rapidly. In the early 1950s, the United States first used FMEA ideas for the design and analysis of a fighter operating system, and by the mid-1960s, FMEA technology was formally used in the aerospace industry (APOLLO program). In 1976, the U.S. Department of Defense issued a military standard for FMEA, but it was limited to design. In the late 1970s, FMEA technology began to enter the automotive industry and medical equipment fields. In the early 1980s, it entered the microelectronics industry. In the mid-1980s, the automotive industry began to apply process FMEA to validate its manufacturing processes. In 1988, the Federal Aviation Administration issued an advisory requiring the use of FMEA in the design and analysis of all aviation systems. In 1991, ISO-9000 recommended the use of FMEA to improve product and process design. In 1994, FMEA became a certification requirement for QS-9000. At present, FMEA has formed a set of scientific and complete analysis methods in engineering practice.
In today's "user is God", constantly meet the needs of users has become the goal of companies, QFD is a market-oriented, based on user needs of a strong planning method. QFD has been described as "a systematic approach that ensures that the development of product characteristics, features and specifications, as well as the selection and formulation of process equipment, methods and controls, are driven by the requirements of the user or the market."
QFD can be regarded as a process consisting of four stages of design, parts, processes and production, which is completed through a series of matrices and charts, which expand user requirements and related technical requirements from product planning and product design to process planning and shop-floor work according to the principle of "the next process is the user of the previous process".
Design stage: In the design stage, the user requirements are transformed into product design requirements or service requirements through the House of Quality matrix, and the relatively important requirements are identified. The House of Quality matrix is a product planning matrix used to describe user needs, related design requirements, target values, and product competitiveness evaluation. It is called the House of Quality because of its shape like a house.
The House of Quality consists of the following parts:
1. Correlation matrix: indicates the correlation between modes.
2. Method: Design method to achieve user requirements.
3. Expected goals: pre-screening criteria that determine whether a method is quantifiable.
4. User requirements: The characteristics of the product or service determined by the user.
5. Importance score: The user's requirements are quantitatively scored to reflect the relative importance of the project to the user.
6. Relationship Matrix: Describes the degree of relationship between a user's requirement and the method used to achieve that requirement.
7. User competitive evaluation: Measures user perceptions of competitive products or services.
8. Technological competitive evaluation: evaluation of the engineering specifications of the various methods adopted by the company as well as the technical specifications of competitors.
9. Probability factors: Indicate the ease or value of each approach adopted by the company.
10. Absolute number: Weighted score of each design requirement based on the correlation determined in the relationship matrix.
11. Relative numbers: Ranking methods according to absolute numbers.
Parts Stage: In this stage, the requirements for parts are determined according to the significant design requirements identified in the first stage. These requirements are strongly related to user-defined product requirements.
Process design phase: Determine the process according to the requirements for the parts. The processes identified in this phase will best achieve the specific requirements of the user for the product.
Production stage: Form production requirements according to process requirements. This process determines the production method that enables the company to produce.
QFD is a powerful tool used in a wide range of fields, such as design, strategic planning, quality improvement, product and service expansion. It brings us the most direct benefit is to shorten the cycle, reduce the cost, improve the quality. More importantly, it changes the traditional quality management idea, that is, from the late reactive quality control to the early preventive quality control. You will also find that it helps us break down the barriers between departments and make the company a cohesive group. Because the development of QFD is by no means a quality department, development department or manufacturing department can be completed independently, it needs collective wisdom and team spirit.
