Some standard content:
IrA 03.120.IC
Guidance on statistical techniques for GB/T 19001-2000 (ISO/TR 19001-2000) 100:7-2003.IDT)
Published on September 5, 2005
General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China Administration of Standardization of the People's Republic of China
GB/Z19027—2005/ISO/TR10017; Pre-2003
Normative references
Identification of potential needs for statistical techniques
Description of statistical techniques for identification
Descriptive statistics
Experimental design (E)
Design and test
Test panel Analysis
Process capability analysis
Regression analysis
Valuability analysis
Sampling
Statistical process control (SPC) chart
Statistical deviation
Time series analysis
Participating literature
B/%.19027—2005/1S0/TR10017.2003 This technical document is equivalent to the technical document of the Lei Chang, using IS0)/TR10017:2003(50SCC2.200U navigation technical guide 3, this guiding technical work (H: account standard confirmed In this regard, the technical document B/T12001-2003 is based on the GB/T120327-2001 GB/T1921-1994 unified technical guide (1: J50/TF). This technical document is based on the GB/T12001-2003 family annotation changes of the 2000 version of GB/T12030, and compared with the GB/T1203027-2001 standard, this technical document has made further improvements on the concepts of statistical data such as the design of the design, the test plan, and the sample, and has given more application examples. This technical document is composed of all This technical guideline document is the work of China National Institute of Standardization for Quality Management and Quality Assurance (SACT11). The drafters of this technical guideline document are: China National Institute of Standardization, China New Era Quality Assurance Certification Center, China Quality Association. The drafters of this technical guideline document are: Yan Wu, Cheng Kuiyong, Yu Zhencong, Cao Chun, Xiao Xiong, Duan Yihan. This technical guideline document is for reference only. The relevant technical guideline document is issued by the Ministry of Standardization Administration of China. GB/Z 19027—2005/IS0/TA 10017:2003 Introduction This technical guideline document is intended to assist organizations in establishing and maintaining a quality management system in accordance with the requirements of CB/[SCC]-2010 and the use of statistical techniques. In fact, variation is observed in all operations and results, even in apparently stable conditions, and statistical techniques are the only ones that are useful. Variation can be observed in quantifiable characteristics of products and processes, and can be observed in all stages of the product life cycle, from product development to customer service and ultimately to the final product. Statistics can help to understand, describe, analyze, and interpret the variation, and even to use available data to better understand the nature, causes, and causes of variation, thereby helping to solve the problems that may arise from these variations. Statistical techniques can help to better organize the available data to make decisions, and can be used to continuously improve the quality of products and services to satisfy customers. Statistical techniques are applicable to a wide range of activities, such as market research, design, development, production, verification, installation and service. The purpose of this guidance document is to guide and assist in the selection of statistical techniques suitable for the needs of the organization. The criteria for determining statistical technical requirements and the appropriateness of the selected statistical techniques remain the sole responsibility of the organization. This guidance document refers to other standards applicable to the statistical techniques described in GB/T 19001-2000, in particular GB/T 19004-2000.
1GA/7.19027-2035/TSO/TR10017.2003GB/T 19001-2000. This normative technical document provides guidelines for the selection of statistical techniques. These requirements are of great significance to the establishment and implementation of the statistical requirements. , maintain and deliver compliance (G13/TSC: When the quality management system meets the new requirements, it can be achieved by finding the requirements of G/T19001 and then identifying and expressing the appropriate statistical techniques. The statistical techniques listed in this technical document are not comprehensive or exhaustive. Organizations should not exclude the use of other techniques that may be beneficial to them. Statistics are their basis), and the two technical documents do not intend to require the use of these statistical techniques, nor do they make decisions on how to apply these common design techniques. This technical document does not intend to scan the document for compliance with the requirements of GB/T19001 or certification. It also does not intend to carry out mandatory inspections to determine whether the requirements are met. The rationale for using statistical techniques is that their rapid use should be effective in improving the quality management system.
Note 1: Technical information "statistical techniques" and "calculation methods" need to be replaced. Note: According to the definition of LR/19uoU2.0-product, the "product" in this guidance technical document is applicable to services, hardware, and cooking wine products or their related products. Www.bzxZ.net
2 Normative references
The following documents are cited in this guidance technical document and are used as references in this guidance technical document: All references to previous versions, including all amendments (excluding errors), are not applicable to this guidance technical document. However, the parties concerned should study whether the latest versions of these documents can be used according to the guidance technical document. 5. Reference documents without a reference date: The latest versions are used in this guidance technical document. G/T190n-26G3 Quality Management System Requirements
The difference between historical data and blood effect
Home identification light band state
No mistake in distinguishing the common death
This identification plan: previous demand
Determine the demand for tea
Permitted price customers full of hundred
No distinguished demand
This visual energy demand
Unidentified demand
Unidentified demand
Unidentified demand
Unidentified demand
Long-term recognition demand
No new demand
The demand for disease fear and evaluation of the city's verification data
The demand for the page decline and reflection
Statistical technology
See this table 7.2.2 Article
Death table 3.2.1 Element improvement
description connection: unique
indicative design; selection
description careful statistics, also vertical ability division: sampling; EPCE
descriptive notes
abstract statistics: pollution sequence
description situation design: SPr
CB:T=-0020(cost
product realization
1.1 output realization will be the first example
1. The above has a beautiful past
7. 2.1 The relevant requirements are indeed completed
7.2.2 The requirements related to the output are reviewed
consider the internal communication
design and installation
7, 3, 7
Design and development planning
Design and development personnel
Stage planning and development planning
Design and development approval
Design and development risk certification
3.6 Design and development personnel
Control of new development and change
7.4 Procurement
3.4.1 Procurement requirements
7.4.2 Procurement specifications
7.4.3 Verification of purchased products
Production and service provision
? ！ Product and service supply approval
Production and service supply approval
7.5.3 Identification and tracking
7. 5. 4 Yingke Property
Capsule 1 (continued)
Quickly determine the demand for effectiveness
Unidentified demand
Unrecognized mountain demand
Unrecognized demand
GB/219027-2905/IS>/TR10017.2003 Statistical technology
Detailed price period to meet the sales demand of the established insurance requirements
Unrecognized demand
Unrecognized demand
Unrecognized demand
Firmly plan to meet the needs of both people and things Unrecognized demand
To prove that the product meets the quality requirements of the designer according to the ratio of product potential and the current needs of the method
Environmental audit, supervision and verification of the benefits of social design more strictly due to the impact of the final measurement analysis, through social energy analysis
analysis sample statistics homogeneous square
performance flow design frequency design retreat control
column: quick simulation: time series analysis
with frequency design test design must be tested: example analysis: same analysis, reliability
quality analysis: single fuel time slice analysis
this property is sparse design; micro-experimental design, limited design test annual analysis through type analysis: regression analysis: reliability analysis oil single: simulation
adaptability design reduction and improvement design: promote the inspection source with performance analysis·oil depot simulation report
guarantee the production account The compliance of the internal flexibility is required, and the design and design of the parts are tested; the example check analysis: the vertical requirements of the requirements of the Communist Party of China provide the group with the requirements of the production of the other requirements of the attack on the technology of the product, to ensure that the requirements of the monitoring and control of the core business are determined, and the requirements of the monitoring and control soldiers are detected. The slip requirements of the reasonable amount of the reasonable requirements are verified, and the requirements can be verified. The internal requirements of the financial consulting property are divided into the process of strength, and the external analysis can be often collected; the sampling is related to the lack of improvement: the test design: the planning and control analysis: the return analysis: the sample shooting tt||Description of the connection system training 9 inspection and testing of the floating plate
process capacity Dan machine shellability folding extraction
oil description flow: folding, small inspection
return folding, reliability folding F
: time folding
method sedimentation: process cavity folding: return folding
dynamic: time folding
Portuguese note statistics||sampling
GB/Z19027—2005/IS0/TR10017:2003151—33 service clause
product protection
7.Monitoring of salt slurry quantity control
E weighing, analyzing dynamic collection and transportation
8.2 Visual and ideal quantity of goods
18. 2.1 Compliance
8, 2, 2 Internal control
Recent releases and quality
Monitoring and measurement of products
8. Control of nonconforming products
9, Stimulus analysis
6.5 Improvement
6.5.1 Continue to improve
List|||Continue
Make new demands
Monitor the new design technology
Monitor the quality of the product: non-addressable analysis; some requirements
Ensure the monitoring and measurement process and the requirements of the equipment and the commission
To obtain the effectiveness of the results of previous tests
To obtain the information related to the customer
Most of the self-audit programs and reports include the requirements of audit data
Monitor The quality control department should check and measure the characteristics of the products to verify that the requirements of the products are met, and ensure that the delivered products will not be insured. The quality control department should re-verify the products that have been corrected to increase the assurance that they meet the requirements. The quality control department should analyze the following aspects to evaluate the effectiveness of the quality guarantee method and estimate the possible changes in the requirements: a) customer satisfaction b) conformity with product requirements tt||The call and return trend of the core process
Use the following aspects of the light effect loss, the sound system can be collected:
Design and development
Production knowledge supply
Monitoring and flux control
Floor plan: time slice fast
Indicator continuous measurement and analysis process can be analyzed
Analysis back sampling SPC use; final calculation
Difference method time series analysis
Statistical flow design will; measurement back door analysis: sampling: certificate calculation abandoned light method: time column: analysis|| tt||Describe the traditional one: sample
Describe the statistical juice: ladder
Describe the test design ratio to promote the control to
Quantity analysis through the type of clinical method to draw SFC
Chart time series analysis
Describe the statistical: test design, absorb the test plan quantity analysis|through the analysis|regression branch|reliability analysis I draw details, SPC letter time series analysis
Abstract continuity design: draw:
See this table 8.2.1
See the car issue 8.2.1 clause
See this document 9. 2.3 Number of items
See this variable 9.2.3 Number of items
See this table 7.4.1 Clause
See, not table 7.3.3.7.3.5,7.3.6 See not table 7.4.1.7.4.3 Clause
See not table 7, 5.1,7.5.2,7, 5, 5. For the entry, see this journal, 4R/19(31-2090 clause 8.5.7 Correction: 6.5.3 Printing resistance Description of identified statistical techniques 4.1 General Table 1 (reading) Requirements for quantitative data TH/x19027-2005/1S0/7R10017.7003 Institute of Design Technology
Division and different meaningful data to help understand the needs of the four
Division and different and different in the format related
teaching data to help understand its requirements
Specification of descriptive statistics rat risk design generally this test
The new: National; God;
Around; Time series analysis
Abstract discussion or design design test
Process of SPC
Time series analysis
The following statistical techniques have some common needs, we have made a discussion in Table 1: Descriptive summary:
·Experimental design:
Design control test!
Capacity analysis:
Process capability analysis:
Expansion analysis:
Reliability analysis:
Sampling;
Simulation:
Statistical process control (SPC) chart:
Design plan;
Time series analysis.
Of the various statistical techniques listed above, the most important one is descriptive statistics (including graphical methods), which is an important component of many statistical techniques.
As mentioned above, the statistical techniques mentioned above are generally known and widely used, and their application has been widely recognized by users.
The selection of statistical techniques depends on the specific application situation and the application method. Brief descriptions of the other statistical techniques mentioned above are given in 2.2 to 4.13. These descriptions have been prepared to enable the professional reader to evaluate the potential applicability and benefits of using these statistical techniques in the implementation of quality management systems. The actual application of these statistical techniques will require more guidance and expertise than can be provided in this technical guidance document. A wealth of information on statistical techniques is available from the public, such as textbooks, periodicals, industry manuals and other sources of information, which may be helpful in the effective use of statistical techniques by an organization. However, it is beyond the scope of this technical guidance document to list these sources. It is the responsibility of each organization to find such information.
4.2 Descriptive Statistics
4.2.1 Descriptive Statistics Descriptive statistics refers to the use of quantitative methods to summarize the characteristics of data. Usually, the data characteristics that organizations are concerned about are their distribution (most commonly, the mean) or high-frequency variation (pass-band). The following are the standards and technical notices related to statistical techniques issued by IS and IEC. They are listed only to provide information. They are guiding technical documents and do not require organizations to implement these standards and technical reports. T: B/Z19027—2005/150/TR 10017:2053 Accurately check the distribution of individual characteristics of the data, such as the "degree of gratitude", the information provided by the statistical data can be effectively conveyed through various graphical methods. This method can include a simple diagram:
a trend diagram, also known as a \return reading\>, which is formed by two layers of persistent values of interest over a period of time, and observe how they change with the question.
Dispersion factor, by plotting one variable on the 2nd axis and the other variable on the 1st axis, the relationship between the two variables can be analyzed:
Histogram. Depict the distribution of the values of the property of interest: There are many graphical techniques, they are used to explain and analyze data, and their range can range from simple engineering (and others such as stripe plots and error plots) to more complex search techniques (such as generalization) to the display of multidimensional space and variables.
avoidance techniques are very useful for accurately analyzing data that are not easily found: Graphical methods are widely used in the analysis of relationships between investigating and verifying variables, and in estimating the parameters that describe these relationships. In addition, graphical methods play an important role in summarizing and presenting complex data in an effective way, especially for non-professionals. Descriptive statistics include many of the statistical techniques listed in the 4th Technical Guide. Descriptive statistics should be considered as an essential part of statistical analysis. 4.2.2 Uses of descriptive statistics Descriptive statistics are used to summarize data. It is usually the initial step in the analysis of quantitative data and is often the first step in using other statistical methods. Within certain test limits and confidence levels, the characteristics of the sample data can serve as the basis for the conclusion that the sample is the best. 4.2.3 Benefits Descriptive statistics summarize and represent data in an efficient and relatively simple way, while providing a convenient way to summarize information. In addition, it is a very effective way to present data and convey information. Descriptive statistics are suitable for all situations where statistics are used: they facilitate the analysis and interpretation of data and can provide valuable assistance to policy.
4.2.4 Limitations and caveats
Descriptive statistics provide quantitative measures of sample size (mean and standard deviation). However, these measures are subject to the nature of the quantity and the nature of the method used: unless the underlying statistical assumptions are met, these measures cannot be used to verify the overall characteristics of the sample.
4.2.5 Application examples
Descriptive statistics are applicable to almost all areas where quantitative data can be collected. They can provide information about some other aspects of a product, process or management system, or can be used in management evaluation. The following are some application examples: summarize the relevant characteristics of a product (such as value and temperature);
characterize the service delivery time or response time of a service provider; summarize the customer response time. 4.3 Experimental Design ([XOF]) 4.3.1 Experimental Design GR/Z.19027--2505/1SO/TR10017:2003 Experimental Design is the conduct of research in a planned manner that relies on the evaluation of the results to draw conclusions at a specified confidence level. It usually involves changes in the system being investigated and the evaluation of these changes. The purpose of DE can be to confirm certain characteristics of the system, or to investigate the influence of some characteristics of multiple system components. The test design and the manner in which the test is conducted constitute a test. In such a design, the test conditions and the time used are determined. There are many methods that can be used to analyze the test data, ranging from analytical methods such as differential analysis (AN(VA)) to methods such as probability causality. 4.3.2 Use of experimental design DE can be used to evaluate the performance of a product overdose system. Its purpose is to determine the effectiveness of several systems in a comparative manner. DCF is particularly useful for investigating complex systems that may be affected by large potential events. The test objectives can be to make the test more specific. The characteristics are maximized or optimized, or their variation is minimized. This can also be used to identify the more influential factors in the system, the size of their effects, and the possible independent relationships (i.e., interactions) between them. The results can be used to improve the design and installation of products or processes, or to modify or improve existing systems. The information obtained through design and testing can be used to build an efficiency model. Under certain restrictions (as listed in 4.3.), the model interprets the system characteristics of interest as the coefficients of the influencing factors. Such a model can be used. 4.3.3 Benefits When estimating or confirming the characteristics of interest, it is necessary to ensure that the results obtained are not due to instrumental variation. This applies to evaluations based on efficiency criteria, as well as to comparisons of two or more systems, which allow such evaluations to be made at a specified confidence level.
One of the main advantages of TOF is that it is more efficient and economical to analyze the effects of multiple factors on a process than to analyze each factor separately. The ability of TOF to identify the interaction effects between certain factors also allows the organization to gain a deeper understanding of the process. This concern of DO is particularly prominent when dealing with complex processes (such as those involving a large number of influencing factors). When analyzing dependent systems, there may be accidental connections between the individual or multiple factors, and there is a risk of finding more than one causal relationship. The risk of error can be reduced by using appropriate experimental design. 4.3.4 Precautions and considerations
All systems have some level of inherent variation (commonly known as "variability"), which can sometimes affect the results of an investigation.Other potential sources of error include the presence of unknown (or only technically recognized) complex effects in the system or the dependencies between various factors in the system. The risk of these errors can never be reduced by a well-designed test (e.g., by the choice of parameters or by consideration of their effects in the test period). These risks cannot be eliminated and this should be taken into account when drawing theoretical conclusions. Strictly speaking, the test results apply to the salt formation under test and its range of pressures and effects. Therefore, caution must be exercised when extrapolating (or donating) values that fall significantly outside the range of the test series. Although these assumptions may exist (e.g., there may be some significant relationship between the product type and the actual product under study), the correctness of these tests is worth considering. 4.3.5 Application Examples
Product validation experiments are often performed on products or products to verify the effectiveness of a treatment, or to evaluate the relative effectiveness of several types of treatments. Examples of applications where validation should be considered include product validation experiments performed to specified performance standards. These tests are generally used to identify factors that affect the process and thereby control or improve a characteristic of interest, such as process yield, product durability, sound quality, or component quality. Such tests are often encountered in the production of electronic components, automobiles, and chemicals. It is widely used in various fields such as agriculture and information technology, and has great potential applications. GB/Z19027-2005/1S0/TR10017:20034.4 Hypothesis testing
4.4.1 Concept of hypothesis testing
Hypothesis testing is used to determine whether a set of data (including all data from the data) conforms to a given statistical method at a specified risk level. A hypothesis may be an assumption about the estimation of a specific statistical point or model, or it may be about a specific expected value (mean value). The method of hypothesis testing also includes evaluating the data in the form of experimental data to determine whether a given statistical model or method should be rejected
Many of the statistical techniques listed in this guidance document explicitly or implicitly use hypothesis testing. Test: light sampling, SPC chart, experimental design, and analysis.
4.4.2 Use of low-hypothesis test
The use of low-hypothesis test
The use of low-hypothesis test
The use of low-hypothesis test
The use of low-hypothesis test
The use of low-hypothesis test
The use of low-hypothesis test
The use of low-hypothesis test
The use of low-hypothesis test
The use of low-hypothesis test
The use of low-hypothesis test
The use of low-hypothesis test
The use of low-hypothesis test
The use of low-hypothesis test
The use of low-hypothesis test
The use of low-hypothesis test
Hypothesis testing makes a judgment about some parameter of a population with a specified confidence level, so hypothesis testing may help to make decisions based on these data.
Hypothesis testing can make judgments about the quality of the distribution of the population and the validity of the data itself. 4.4.4 Limits and precautions
To ensure the validity of the hypothesis test, basic statistical requirements must be met, in particular the sample should be independent and random. However, the sample determines the accuracy of the statistical analysis. In theory, there is no controversy about whether some conclusions can be drawn from a certain set of rules. 4.4.5 Application examples
When a judgment must be made about a certain number or a certain number of items (estimated from a sample), it is necessary to verify the data itself. When evaluating the price, the application of low-set tests is recommended. For example, hypothesis tests can be used in the following ways: to test whether the mean (or standard deviation) of the total product is a given value, such as a standard or a standard; to test whether the means of two or more populations are different when comparing different batches of products; to test whether the total defect rate does not exceed a given value; to test the difference in the defect rate of the whole process; to test whether a certain result of the experiment is a "double value", that is, an extreme value of questionable validity; to test whether some products or process characteristics have changed; to use the sample size required to determine the acceptable service report at a specified confidence level, but the confidence interval of the total product may exist. 4.5 Measurement Analysis
4.5.1 The concept of measurement analysis
Measurement analysis (also known as “measurement uncertainty analysis” or “measurement system analysis”) is a set of methods for evaluating the accuracy of a measurement system under the conditions under which the system is operating. The analysis of its error tolerance can be used to analyze the characteristics of the product. 4.5.2 The use of measurement analysis
It collects the uncertainty of the measurement system and uses it to evaluate the suitability of the measurement system for the expected performance within the specified range. The measurement analysis quantifies the variation from various sources, such as variation from the measurement personnel, variation from the measurement process, or uncertainty.Hypothesis testing can be used in the following ways: to test the mean (or standard deviation) of the total product, such as a standard or a daily standard; to test whether the means of two or more populations are different when comparing different batches of products; to test whether the total defect rate does not exceed a given value; to test the difference in the defect rate of the whole process; to test whether a certain result of the experiment is a "double value", that is, an extreme value of questionable validity; to test whether some products or process characteristics have changed; to use the sample size required to determine the acceptable service report at a specified confidence level, but the confidence interval that may exist is determined. 4.5 Measurement Analysis
4.5.1 The concept of measurement analysis
Measurement analysis (also known as “measurement uncertainty analysis” or “measurement system analysis”) is a set of methods for evaluating the accuracy of a measurement system under the conditions under which the system is operating. The analysis of its error tolerance can be used to analyze the characteristics of the product. 4.5.2 The use of measurement analysis
It collects the uncertainty of the measurement system and uses it to evaluate the suitability of the measurement system for the expected performance within the specified range. The measurement analysis quantifies the variation from various sources, such as variation from the measurement personnel, variation from the measurement process, or uncertainty.Hypothesis testing can be used in the following ways: to test the mean (or standard deviation) of the total product, such as a standard or a daily standard; to test whether the means of two or more populations are different when comparing different batches of products; to test whether the total defect rate does not exceed a given value; to test the difference in the defect rate of the whole process; to test whether a certain result of the experiment is a "double value", that is, an extreme value of questionable validity; to test whether some products or process characteristics have changed; to use the sample size required to determine the acceptable service report at a specified confidence level, but the confidence interval that may exist is determined. 4.5 Measurement Analysis
4.5.1 The concept of measurement analysis
Measurement analysis (also known as “measurement uncertainty analysis” or “measurement system analysis”) is a set of methods for evaluating the accuracy of a measurement system under the conditions under which the system is operating. The analysis of its error tolerance can be used to analyze the characteristics of the product. 4.5.2 The use of measurement analysis
It collects the uncertainty of the measurement system and uses it to evaluate the suitability of the measurement system for the expected performance within the specified range. The measurement analysis quantifies the variation from various sources, such as variation from the measurement personnel, variation from the measurement process, or uncertainty.
Tip: This standard content only shows part of the intercepted content of the complete standard. If you need the complete standard, please go to the top to download the complete standard document for free.