Sampling procedures for inspection by attributes - Part 4: Procedures for assessment of declared quality levels
Introduction to standards:
The sampling plan and assessment procedures specified in this part of GB/T 2828 are used to assess whether the quality level of a population (batch or process, etc.) meets a certain claimed quality level. This part of GB/T 2828 can be used for various forms of quality verification. This part replaces GB/T 14437-1997 "Procedures and sampling tables for single-pass supervision sampling inspection by attributes" and GB/T 14162-1994 "Procedures and sampling tables for supervision sampling inspection by attributes". Compared with GB/T 14437-1997, the changes in the technical contents of this part mainly include: - Modification of the standard format in accordance with the requirements of GB/T 1.1-2000 "Guidelines for standardization work Part 1: Structure and writing rules of standards"; - Changing the supervision quality level (p0) to the claimed quality level (DQL); - Adding the discussion of DQL = 0; - Adding the content that can be used for the number of unqualified products per hundred units as the quality indicator; - Adding the operating characteristic curve diagrams and tables of the sampling plan; - Adding the definitions of retest, re-inspection, re-check and verification sampling inspection efficacy; - Special emphasis is placed on the fact that this standard can be applied to operations, work-in-progress, inventory, maintenance operations, data or records, and management procedures in addition to final products, parts and raw materials. GB/T 2828.4-2008 Sampling inspection procedures for attributes Part 4: Assessment procedures for claimed quality levels GB/T2828.4-2008 standard download decompression password: www.bzxz.net
The sampling plans and assessment procedures specified in this part of GB/T2828 are used to assess whether the quality level of a population (batch or process, etc.) meets a claimed quality level. These sampling plans are designed to control the risk of rejecting a qualified population (batch) to 5%, and the risk of not rejecting a non-conforming population (batch) whose actual quality level is LQR times the claimed quality level (see Chapter 4) to 10%.
This part of GB/T2828 provides sampling plans with four discrimination levels.
Unlike the procedures in other parts of GB/T2828, the procedures in this part cannot be used for acceptance sampling of batches. This part of GB/T2828 can be used for various forms of quality verification. This includes situations where the inspection results of samples are used to indicate whether a verification population meets a certain claimed quality level.
This part of GB/T 2828 applies to situations where it is permitted to draw a random sample consisting of a number of unit products from the verification population.
The sampling plan provided in this part of GB/T 2828 can be used for (but not limited to) the inspection of the following products, for example:
---final product;
---parts and raw materials;
---operation;
---work in process;
---inventory;
---maintenance operation;
---data or records;
---management procedures.
This part is mainly used for situations where the number or rate of nonconforming products in the verification population is determined when the unit products to be inspected are divided into conforming products and nonconforming products.
For the case of the number of nonconforming products or the number of nonconforming products per 100 units, this part can also be used with slight changes. The necessary changes are:
---replace the number of nonconforming products with the number of nonconforming products;
---replace the percentage of nonconforming products (or the number of nonconforming products per 100 units) with the number of nonconforming products per 100 units.
In this case, the values given in Tables 2 to 9 are only approximate values.
The values given in Tables 2 to 9 of this part are calculated using the binomial distribution. When the total number of verification is less than 250, or the ratio of the total number of verification to the sample size is greater than 10, the sampling plan retrieved from this part is approximate. The sampling plan can be determined according to the method specified in GB/T 15482.
GB/T2828 "Sampling procedures for inspection by attributes" currently includes the following parts. Its structure and corresponding international standards and national standards to be replaced are:
--- Part 1: Sampling plan for batch inspection based on acceptance quality limit (AQL) (ISO 2859-1:1999, IDT; replaces GB/T2828-1987)
--- Part 2: Sampling plan for isolated batch inspection based on limiting quality (LQ) (ISO 2859-2:1985, NEQ; replaces GB/T15239-1994)
--- Part 3: Skip batch sampling procedure (ISO2859-3:2005, IDT; replaces GB/T13263-1991)
---Part 4: Procedure for assessment of claimed quality level (ISO 2859-4:2002, MOD; replaces GB/T 14437-1997 and GB/T 14162-1993)
---Part 5: System of sequential sampling plans for batch inspection based on acceptance quality limit [AQL] (corresponding to ISO 2859-5:2005)
---Part 10: Introduction to sampling systems by attributes (corresponding to ISO 2859-10:2006)
---Part 11: Procedure for assessment of claimed quality level of small populations (replaces GB/T 15482-1995)
This part is Part 4 of GB/T 2828.
This part of GB/T2828 is modified to adopt the international standard ISO2859-4:2002 "Sampling procedures for inspection by attributes Part 4: Procedures for assessment of claimed quality levels". Compared with ISO 2859-4:2002, the technical content changes of this part mainly include:
--- Added the sampling plan of L=0 in the main table of sampling plans in Table 1;
--- Added the sampling plans of n=8, n=5, and n=3 in LQR level I of the main table of sampling plans in Table 1, and added the sampling plan of n=8 in LQR level II;
--- Added the limit quality ratio (LQR) of the sampling plan of L=0 and the probability table of the verification population that is wrongly judged as unqualified in random inspection, that is, Table 2;
--- Added the probability table of the sampling plan of L=0 for different values of quality ratio for random inspection unqualified, that is, Table 6;
--- Added the probability curve of the sampling plan of L=0 for different values of quality ratio for random inspection unqualified in Figure 1;
--- Added the discussion of DQL=0 in Appendix A.
This Part replaces GB/T 14437-1997 “Procedures and Sampling Tables for Single-shot Supervision Inspection by Attributes” and GB/T 14162-1994 “Procedures and Sampling Tables for Supervision Inspection by Attributes”. Compared with GB/T14437-1997, the changes in technical contents of this part mainly include:
--- The standard format has been modified according to the requirements of GB/T1.1-2000 "Guidelines for Standardization Part 1: Structure and Writing Rules of Standards";
--- The supervision quality level (p0) has been changed to the claimed quality level (DQL);
--- The discussion of DQL=0 has been added;
--- The content that can be used to take the number of unqualified products per hundred units as the quality indicator has been added;
--- The operating characteristic curve diagram and table of the sampling plan have been added;
--- The definitions of retest, recheck, recheck and verification sampling inspection efficacy have been added;
--- It is particularly emphasized that this standard can be applied to operations, work-in-progress, inventory, maintenance operations, data or records, and management procedures in addition to final products, parts and raw materials.
Appendix A of this part is a normative appendix, and Appendix B and Appendix C are both informative appendices.
This part is proposed by the China National Institute of Standardization. || tt
||This part is under the jurisdiction of the National Technical Committee for Standardization of Statistical Method Application.
The drafting units of this part are: China National Institute of Standardization, the Second Academy of China Aerospace Science and Industry Corporation, Guangdong Provincial Administration for Industry and Commerce, Institute of Mathematics and Systems Science of the Chinese Academy of Sciences, Guangzhou Product Quality Supervision and Inspection Institute, and Ordnance Engineering College of the Chinese People's Liberation Army.
The main drafters of this part are: Yu Zhenfan, Zhao Guanghua, Chen Yehuai, Yin Qin, Ding Wenxing, Deng Suixing, Dang Hua, Ma Yilin, and Zhang Yuzhu.
The
previous versions of the standards replaced by this standard are:
---GB/T14437-1997;
---GB/T14162-1994.
Foreword I
Introduction III
1 Scope 1
2 Normative references 1
3 Terms, definitions and symbols 2
3.1 Terms and definitions 2
3.2 Symbols and abbreviations 3
4 Principle 3
5 Claimed quality level 4
6 Sampling plan 5
6.1 LQR (limiting quality ratio) level 5
6.2 Selection of sampling plan 8
7 Procedure for implementing sampling inspection 8
7.1 Determine the inspection population 8
7.2 Determine the technical performance, quality characteristics and requirements of the unit product 8
7.3 Determine the classification of nonconforming products 8
7.4
7.5 Specifying the claimed quality level8
7.6 Searching the sampling plan8
7.7 Drawing samples8
7.8 Inspecting samples9
7.9 Dealing with nonconforming products9
7.10 Decision criteria9
7.11 Rechecking9
8 Additional information9
8.1 Approximate probability curve for sampling nonconformity9
8.2 Table indicating discrimination ability10
Annex A (Normative) Case where the claimed quality level DQL is equal to 013
Annex B (Informative) Examples of using this standard14
Annex C (Informative) Notes on the modifications to ISO 2859-4:200216
References 17
Some standard content:
ICS 03.120.30
National Standard of the People's Republic of China
GB/T2828.4—2008
Replaces GB/T14437-1997, GB/T14162:1993 Sampling procedures for inspection by attributes--Part 4: Procedures for assessment of declared quality levels(ISO2859-4:2002,MOD)
2008-07-28 Issued
General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China Standardization Administration of China
2009-01-01 Implementation
CB/T2828.4-2008
1 Scope
Normative references
Terms, definitions and symbols
3.1 Terms and definitions
3.2 Symbols and abbreviations
Claimed quality levels
Sampling plan
6.1 LQR (limiting quality ratio) levels
6.2 Selection of sampling plan
7 Procedure for implementing sampling inspection
Determination of the inspection population
Determination of the unit product Technical performance, quality assurance and requirementsDetermination of classification of nonconforming products
Specify the claimed quality level
Specify the LQR level
Search sampling plan
Sampling
Inspection sample
Disposal of nonconforming products
Judgment criteria
Review
Additional information
Approximate probability curve of sampling failure8.2 Table indicating discrimination abilityWww.bzxZ.net
Appendix A (normative)
Appendix B (informative)
Case where the claimed quality level DQI. is equal to 0
Example of using this standard……….
Appendix C (informative)
Revision of ISO 2859-1:2002
References
GB/T 2828.4-2008
GB/T2828 "Inspection Procedure by Attribute Sampling" currently includes the following parts. Its structure and corresponding international standards and national standards to be replaced are:
Part 1: Sampling plan for batch inspection based on acceptance quality limit (AQL) (ISO2859-1:1999, IDT, replacing GB/T 2828 -1987)
-Part 2: Sampling plan for isolated lot inspection based on limit quality (IQ) (ISO2859-2:1985, NEQ; replaces GB/T15239--1994)
-Part 3: Skip lot sampling procedure (ISO2859-3:2005, 1DT: replaces CB/T13263-1991)-Part 4: Procedure for assessment of claimed quality level (ISO2859-4:2002, MOD: replaces GB/T14437-1997 and GB/T141621993)
- ·Part 5: System of sequential sampling plans for batch inspection based on acceptance quality limit [AQL7] (corresponding to ISO2859-5, 2005) - Part 10: Introduction to sampling system by attributes (corresponding to ISO2859-10:2006) - Part 11: Procedure for assessment of claimed quality level of small populations (replaces GB/T15482-1995) This part is Part 4 of GB/T2828:
This part of GB/T2828 is modified to adopt the international standard ISO2859-4:2002 "Procedures for sampling inspection by attributes Part 4: Procedure for assessment of claimed quality level". Compared with ISO2859-4:2002, the technical content changes of this part mainly include: adding a sampling plan of L=0 in the main table of sampling plans in Table 1; adding sampling plans of n=8, n5, n=3 in LQR level I of the sampling plan king table in Table 1, and adding a sampling plan of n=8 in LQR level I; adding the "limit quality ratio (LQR) and the probability table of the verification population that is erroneously judged as unqualified in random inspection" of the sampling plan of L=0, namely Table 2; adding the "probability table of random inspection unqualified for different values of quality ratio" of the sampling plan of L-0, namely Table 6; adding the "probability curve of random inspection unqualified for different values of quality ratio" of the sampling plan of L=0 in Figure 1, and adding the discussion of DQL=0 in Appendix A. This part replaces GB/T 14137-1997 "Procedures and sampling tables for supervisory sampling inspection by attributes" and GB/T 14162-1994 "Procedures and sampling tables for supervisory sampling inspection by attributes". Compared with GB/I114371997, the changes in technical contents of this part mainly include:
The standard format has been modified according to the requirements of GB/T1.1-2000 Standardization Guidelines Part 1: Structure and Writing Rules of Standards;
- Changed the supervision quality level (efficiency) to the claimed quality level (DQL); - Added the discussion of DQI, -O;
- Added the content that can be used to take the number of unqualified products per hundred units as the quality indicator: Added the operating characteristic function chart and table of sampling plan; Added the definition of retest, recheck, recheck and verification sampling inspection efficacy; Special emphasis is placed on the fact that this standard can be applied to operations, work-in-progress, inventory, maintenance operations, data or records, and management procedures in addition to final products, parts and raw materials. Appendix A of this part is a normative appendix, and Appendix B and Appendix C are both informative appendices. This part is proposed by China National Institute of Standardization. This part is under the jurisdiction of the National Technical Committee for the Application of Statistical Methods. GB/T2828.4—2008. The drafting units of this part are: China National Institute of Standardization, Second Academy of China Aerospace Science and Industry Corporation, Guangdong Provincial Administration for Industry and Commerce, Institute of Mathematics and Systems Science of the Chinese Academy of Sciences, Guangzhou Product Quality Supervision and Inspection Institute, and People's Liberation Army Ordnance Engineering College. The main drafters of this part are: Yu Zhenfan, Zhao Guanghua, Chen Yehuai, Gu Qin, Ding Wenxing, Deng Suixing, Dang Hua, Ma Yilin, and Zhang Yuzhu. The previous versions of the standards replaced by this standard are: -G13/T14137--1997;
-CR/T14162--1994
.. .r-
GB/T2828.4—2008
The scope of GB/T2828.4 is different from Part 1 and Part 3 of GB/T2828. (The system of acceptance sampling procedures specified in Parts 1 to 3 of GB/T 2828 is applicable to a bilateral agreement between two parties (e.g., a producer and a user). Acceptance sampling procedures are used only as practical rules for delivering products after inspecting a sample of the acceptance batch. Therefore, these procedures do not explicitly involve any form of declared quality level.
In acceptance sampling, it is considered that there is no clear boundary between the quality levels of acceptable batches and unacceptable batches. For the procedures of Parts 1 to 3 of GB/T 2828, a certain acceptance quality limit agreed by both parties is the worst process average quality level that can be tolerated when a series of consecutive batches are submitted. The transfer rules and sampling plans in GB/T 2828.1 are designed to encourage producers to produce products with a process average quality level that is better than the selected AQI. In order to make the sample size appropriate, the protection provided to prevent the acceptance of individual low-quality (LQ) batches may be better than the protection provided by the sampling plan for the purpose of judging individual batches. The sampling rate should be small (the low-level sampling plan in TSO28594 has a higher β risk). On the contrary, the purpose of designing the procedures in Part 2 of G1/T2828 is to provide good protection against the acceptance of individual inferior batches, but the cost may be a high risk of not accepting batches that are actually considered acceptable by both parties. The procedures in Parts 1 and 3 of GB/T2828 are suitable for acceptance sampling, but not for verifying the claimed quality of a verification population in market evaluation and audit. The main reason is that Parts 1 and 3 of CB/T2828 are retrieved by acceptance quality limits and are only related to the actual purpose of acceptance sampling, so the various risks are balanced. The sampling inspection procedures specified in this part of G1/T2828 are developed for sampling inspections required in formal assessments. When implementing this form of inspection, the responsible department must consider the risk of making incorrect conclusions and take this risk into account in arranging and conducting assessments (or audits, or tests). This part of GB/T2828 is designed.Some rules should be set so that when the actual quality level of the verified population meets the claimed quality level, the risk of judging the random inspection as unqualified is very small. If you also want to make the risk of judging the random inspection as qualified when the actual quality level of the verified population does not meet the claimed quality level also very small, you must have a larger sample size. In order to make the sample size appropriate, it is allowed that when the actual quality level does not actually meet the claimed quality level, the risk of judging the random inspection as qualified is slightly higher. The wording of the evaluation results should reflect the imbalance between the risks of various erroneous conclusions obtained. When the sampling result judges the random inspection as unqualified, there is a high degree of confidence that: "The actual quality level of the verified population is inferior to the claimed quality level." When the sampling results are judged as qualified, it is considered that: "For this limited sample size, it is not found that the actual quality level of the verification population is inferior to the claimed quality level. Therefore, when the sample size is small, the verification department is not responsible for confirming the overall qualification of the verification population that is judged as qualified for random inspection. The sampling plan and assessment procedure specified in this part of CB/T 2828 are used to assess whether the quality level of a population (batch or process, etc.) meets a certain claimed quality level. These sampling plans are designed to control the risk of rejecting a qualified population (batch) to 5%, and the risk of not rejecting an unqualified population (batch) whose actual quality level is LQR times the claimed quality level (see Chapter 4) to 10%. GB/T This part of GB/T 2828 provides sampling plans with four discrimination levels: Unlike the procedures in other parts of GB/T 2828, the procedures in this part cannot be used for batch acceptance sampling. This part of GB/T 2828 can be used for various forms of quality verification. Including the use of sample inspection results to indicate whether a certain verification population meets a certain claimed quality level.
This part of CB/T 2828 is applicable to situations where random samples consisting of some unit products are allowed to be drawn from the verification population. The sampling plan provided in this part of GB/T 2828 can be used for (but not limited to) the inspection of the following products, such as: final products;
—·Parts and raw materials; || tt||Operation:
—work in process,
stock;
—maintenance operation;
Data or record:
management procedure.
This standard is mainly used for checking the number of nonconforming products or the percentage of nonconforming products in the population when the units to be inspected are divided into conforming products and nonconforming products.
This standard can also be used with slight modifications for the number of nonconforming products or the number of nonconforming products per hundred units. The necessary modifications are:--replace "number of nonconforming products" with "number of nonconforming products" and replace "percentage of nonconforming products (or number of nonconforming products per hundred units)" with "number of nonconforming products per white unit". In this case, the table values given in Tables 2 to 9 are only approximate values. The table values given in Tables 2 to 9 of this standard are calculated using the binomial distribution. When the total number of verification is less than 250, or the ratio of the total number of verification to the sample size is greater than 10, the sampling plan retrieved from this part is approximate. The sampling plan can be determined according to the method specified in GB/T15482.
2 Normative references
The clauses in the following documents become clauses of this part through reference to this part of GB/T2828. For all referenced documents with an undated date, all subsequent amendments (excluding errata) or revisions are not applicable to this part. However, parties to an agreement based on this part are encouraged to study whether the latest versions of these documents can be used. For all undated referenced documents, the latest versions apply to this part. GB/T321 Priority number and priority number system
GI3/T2828.1---2003 Sampling procedures for inspection by attributes Part 1: Sampling plan for batch inspection based on acceptance quality limit (AQL) (ISO 2859-1:1999.IDI)
GB/T2828.22008 Sampling procedures for inspection by attributes Part 2: Sampling plan for isolated batch inspection based on limit quality (LQ) 1
GB/T 2828.4—2008
Document (ISO2859-2:1985,NEQ)
G3/T2828.3—2008 Sampling inspection procedures by attributes Part 3: Skip lot sampling procedures (ISO2859-3:2005,IDT)GB/I15482 Product quality supervision Small population sampling inspection procedures and sampling tablesCB/T16306 Assessment procedures for re-inspection and re-testing of claimed quality levelsGB/I19000-2000 Quality management system fundamentals and terminology (ISO9000:2000,IDT)T)I S03534-1:2006 Statistical vocabulary and symbols Part 1: General statistical terms and terms used in probability IS035342:2006 Statistical vocabulary and symbols Part 2: Applied statistics 3 Terms, definitions and symbols
GB/T2828.1—2003,1S03534-1:2006, IS03534-2:2006 and GB/T190002000 The terms, definitions and symbols determined by the following terms, definitions and symbols apply to the wood part of GB/T2828. For ease of reference, some terms are directly quoted from the above standards.
3.1 Terms and definitions
Declared quality level The upper limit of the percentage of nonconforming products allowed in the inspection population. 3.1.2
Nonconformity
Not meeting the specified requirements.
According to the importance of the unit product quality characteristics or the severity of the non-conformity of the quality characteristics, nonconformities can be divided into: Class A nonconformities, Class B nonconformities and Class C nonconformities.
Limiting umber of nonconforming itemsThe maximum number of nonconforming items allowed in the sample of the audit population under study based on the claimed quality level. 3. 1. 4
Audit populationaoditpopulatian
The total number of unit products under audit. 3. 1. 5
Audit populationconformityThe actual percentage of nonconforming items in the audit population is less than or equal to the claimed quality level. 3. 1. 6
Audit populationnnnconformityThe actual percentage of nonconforming items in the audit population is greater than the claimed quality level. 3. 1. 7
sampling inspection passed
sampling inspection passed
The number of qualified products d included in the sample is less than or equal to the limit number of unqualified products L3. 1. 8
Fsampling inspection failed
sampling inspection failed
The number of unqualified products included in the sample is greater than the limit number of unqualified products L. 3. 1. 9
repeat test
Perform repeatability or reproducibility test on the original sample. 2
repeat inspection
Take samples from the original inspection population for inspection again to determine whether the inspection population is qualified. 3. 1. 11
repeat test or inspection
Re-inspection and re-testing are collectively referred to as re-inspection.
Quality level quality Ievel
The actual percentage of nonconforming products in the audit population. 3.1.13
Quality ratio quality Tatio
The ratio of the actual quality level of the audit population to the claimed quality level. 3.1, 14
Limiting quality ratio limiting quality ratio GB/T 2828. 4--2008
The value of the quality ratio when the risk of incorrectly judging that the audit population is qualified in the random inspection is limited to a certain smaller value (specified as 10% in this standard). 3.1.15
Limiting quality ratio level limiting quality ratio level limiting quality ratio level.
Power of audit sampling When the actual quality level of the audit population is greater than the claimed quality level DQ1., the probability that the audit population is judged as unqualified in the random inspection. 3.2 Symbols and abbreviations
The symbols and abbreviations used in this part of GB/T 2828 are as follows: DQL Claimed quality level
1 Limit number of nonconforming products
n Sample size
(n;L) Sampling plan
Mass ratio
Limiting mass ratio
Total inspection size
Number of nonconforming products in the inspection population
Number of nonconforming products in the sample
Actual quality level of the inspection population
The probability of judging the inspection population as qualified for random inspection according to the sampling plan when the actual quality level of the inspection population is equal to force. Probability of first type error (risk of misjudgment)
β Probability of second type error (risk of missed judgment)
4 Principle
Any assessment based on sampling will have inherent uncertainty due to the randomness of sampling. The procedures provided in this part of GB/T 2828 only judge the verification population as unqualified when there is sufficient evidence that the actual quality level is inferior to the claimed quality level. These procedures are designed in the following way, that is, when the actual quality level of the verification population is equal to or better than the claimed quality level, the risk of judging the random inspection as unqualified is controlled at about 5%. Therefore, when the actual quality level is inferior to the claimed quality level, there is a risk of judging the random inspection of the verification population as qualified. This risk depends on the value of the quality ratio: the limit quality ratio LQR is introduced to represent the maximum quality ratio that can be tolerated. When the actual quality level is 1.QR times the claimed quality level, the procedures in this part of GB/T 2828 have a 10% risk of judging the random inspection as qualified (equivalent to a probability of 90% of judging the random inspection as unqualified). Details related to the four LQR levels are given in 6.1. The sampling plans provided in this part of GB/T 2828 are indexed by the claimed quality level (DQL) and the limit quality ratio (LQR) levels, as given in Table 1.
Table 1 Sampling plan master table
Percentage of nonconforming products
(Number of nonconforming products per hundred units of product
)
LQR level 0
LQR level 1
LQR level II
Sampling plan based on the claimed quality level DQt. of nonconforming products and the limit quality ratio I.QR level. 2
LQR level III
The arrow → in the table indicates that there is no appropriate sampling plan for the limit quality ratio level, and the sampling plan on the right corresponding to the smaller limit quality ratio is used. The arrow → in the table indicates that there is an appropriate sampling plan for the limit quality ratio level, and the sampling plan on the left corresponding to the larger limit quality ratio is used. 5 Claimed quality level
For mass-produced products, the overall quality requirements for the products shall be specified in the product standards. 3
When the manufacturer accepts the inspection, the DQL value of the quality level claimed by the manufacturer shall not be greater than the overall quality level of the product required in the product standard, provided that the manufacturer is sure of it. When the responsible department proposes the inspection, the DQL value of the claimed quality level specified shall not be less than the overall quality level of the product required in the product standard. The sampling plan provided in this part of GB/T 2828 is retrieved by DQL and I.QR level. The DQL values in Tables 1 to 5 are generally referred to as priority DQL. This series corresponds to the priority AQI value series given in GB/T 2828.1. The content of the priority number is shown in GB/T321. When a DQL is specified for a certain type of non-conformity, it means that the supplier has sufficient reason to believe that the quality level of its product is not worse than the DQI.
6 Sampling plan
6.1 LQR (limiting quality ratio) level
6.1. 1 Level 0
GB/T 2828.4—2008
LQR level O can be used when the sample size is expected to be very small. For the sampling plan with LQR level O, the value range of the limiting quality ratio is from 27.36 to 48.70. For example, if the claimed quality level DQL is 2.5, when the actual quality level of the verification population is 27.36 times the claimed quality level, that is, the actual percentage of defective products is 68.4, then the risk of judging that the verification population is qualified in the random inspection is 10% (see Table 2)
Table 2 Limiting quality ratio (LQR) and the probability of the verification population being wrongly judged as qualified as unqualified in the random inspectionDQL Percentage of defective products
(Number of defective products per 100 units)
\LQR level O plan
Verification population being wrongly judged as qualified
Probability of unqualified in the random inspection a/%
For example, assume that the sampling plan corresponding to the claimed pan level DQL of 1.0 is n=5 and L=0. For this sampling plan, when the actual quality level is 36.91 times the DQI., that is, if the actual percentage of non-conforming products is 36.91, the risk of judging the verification population as qualified is β-10%. On the contrary, if the actual quality level is already the DQL, that is, if the actual percentage of non-conforming products is 1.0, the rate of incorrectly judging the verification population as unqualified is α-4.9%.
6. 1. 2 Level 1
When the sample size is small, LQR level I can be used. For the sampling plan of LQR level I, the value range of the limit quality ratio is from 8.01 to 12.96. For example, if the claimed quality level DQL is 0.10, when the actual quality level of the verification population is 12.30 times the claimed quality level, that is, when the actual percentage of non-conforming products is 1.230, the risk of judging the verification population as qualified is 10% (see Table 3).
GB/T 2828. 4—2008
Table 3 Limiting quality ratio (I.QR) and probability of falsely judging the inspection population as qualified for random inspection failure - DQL Percentage of unqualified products
(Number of unqualified products per unit)
-LQR level I plan
Falsely judging the inspection population as qualified
Probability of unqualified random inspection α/%
Example: Assume that the sampling plan corresponding to the claimed quality level DQL of 0.1 is n-315 and L=1. For this sampling plan, when the actual quality level is 12.3 times the DQL, that is, if the actual percentage of unqualified products is 1.23, the risk of judging the inspection population as qualified for random inspection is β-1C added. On the contrary, if the actual quality level is already the DQL, that is, if the actual percentage of non-conforming products is 0.1, the probability of erroneously judging that the inspection population is unqualified is α-4.0%.
6.1.3 Level Pressure
When a larger sample size is allowed, the sampling plan of level II can be used. For the sampling plan of level II, the value range of the limit quality ratio is from 5.38 to 7.07. For example, if the claimed quality level DQL is 0.10, when the actual quality level of the inspection population is 6.64 times the claimed quality level, that is, when the actual percentage of non-conforming products is 0.664, the risk of judging that the inspection population is qualified is 10% (see Table 4).
Table 4 Limiting quality ratio (LQR) and the probability that the verification population is erroneously judged as qualified as unqualified in random inspection DQL Percentage of unqualified products
(Number of unqualified products per 100 units)
-LQR level II plan
Verification population erroneously judged as qualified
Probability of unqualified in random inspection α/
TQL Percentage of unqualified products
(Number of unqualified products per 100 units)
Table 4 (continued)
GB/T 2828.4—2008
Verification population erroneously judged as qualified
Probability of unqualified in random inspection α/%
Example: Assume that the sampling plans corresponding to the claimed quality expectation level IQ1 of 0.1 are n-800 and L-2. For this sampling plan, when the actual quality level is 6.64 times the DQL, that is, if the actual percentage of non-conforming products is 0.664, the risk of judging that the inspection population is qualified is α = 10%. On the contrary, if the actual quality level is already the LXQL, that is, if the actual percentage of non-conforming products is .1, the probability of incorrectly judging that the inspection population is unqualified is α-4.7%.
6.1.4 level sub
level sub is suitable for situations where a relatively small LQR is required at the expense of a larger sample size. For the sampling plan of level II, the value range of the limit quality ratio is from 4.44 to 5.55. For example, if the claimed quality level is 0.10, and the actual quality level of the inspection population is 5.34 times the claimed quality level, that is, the actual percentage of non-conforming products is 0.534, the risk of judging that the inspection population is qualified is β=10% (see Table 5).
Table 5 Limiting quality ratio (LQR) and probability of the inspection population being wrongly judged as qualified being unqualified in random inspection DQL Percentage of unqualified products
(number of unqualified products per hundred units)
-LOR level III plan
Inspection population wrongly judged as qualified
Rate of unqualified random inspection α/%
Example: Assume that the sampling plan corresponding to the claimed quality level DQL of 0.1 is L=1250 and L-3. For this sampling plan, when the actual quality level is 5.34 times the DQL, that is, if the actual percentage of unqualified products is 0.534, the risk of judging that the inspection population is qualified in random inspection is β=10%. On the contrary, if the actual quality level is already the DQ level, that is, if the actual percentage of defective products is 0.1, the probability of incorrectly judging that the overall inspection is unqualified is -3.8%.
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.