Some standard content:
National Standard of the People's Republic of China
Guide for sampling Inspection
Guide for sampling Inspection1Subject Content and Scope of Application
GB/T1339392
This standard is a guide for selecting sampling systems, sampling plans and sampling schemes. It provides some guiding tables (Table 1, Table 2a, Table 3) to facilitate users to choose a suitable sampling system, sampling plan and sampling scheme. This standard is applicable to the inspection of batches of products (such as materials, parts, components, and complete systems). It can also be used for the inspection of continuously submitted work-in-progress products.
For the determination of the acceptable quality level (AQL), limit quality (LQ), risk quality and (inhibition sampling) inspection level, as well as the comparison of sampling inspection methods, please see Appendix A:
2 Reference standards
GB 2828
Batch inspection attribute sampling procedure and sampling table (suitable for inspection of consecutive batches) GH 2829
GB3358
GB 6378
GB 8051
GB 8052
CB 8053
GB 8054
Periodic inspection counting sampling procedures and sampling tables (applicable to the inspection of the stability of the production process) Statistical terms and symbols
Metric sampling inspection of non-conforming product rate and diagrams (applicable to the inspection of continuous velvet batches) Discussion on sequential sampling inspection procedures and tables
Single-level and multi-level counting continuous sampling inspection procedures and tables Non-conforming product rate standard type - single throw sampling inspection procedures and tables Average measurement standard type - single sampling inspection procedures and tables GB/T13262 Count standard type of non-conforming product rate Secondary sampling inspection procedures and sampling tables GB/T13263 Skip batch counting sampling inspection procedures (R/T13264 Non-conforming product rate small batch counting sampling inspection procedures and sampling tables 3 Market conditions and production characteristics that affect selection 3.1 Market Conditions
In order to select an appropriate sampling system, sampling plans and sampling schemes shall take into account the following conditions: (a) Production is expected to be continuous and the consignee can influence the supplier to improve product quality. (b) After the product is accepted, the supplier remains responsible. (c) Acceptance of a small number of nonconforming products may cause significant losses or serious dangers. (d) Due to the single source of products, the rejection of a batch may cause factory closure or economic losses. (e) The cost and technology of inspecting a unit of product are high. (e) Inspection is destructive.
The opposite situation is as follows:
(\) The consignee cannot influence the supplier to improve quality: (b,) Once the product is accepted: the supplier is no longer responsible. wm. Free download of customer standard industry information CB/T 13393-92
(c) Nonconforming products have no economic and other impacts, and/or are easily identified and eliminated during assembly. d,) Equivalent products are available from many sources and are easily obtained. (e) The cost of testing a unit of product is not high. (f) The test is non-destructive.
3.2 Product characteristics
In addition to market conditions, product characteristics and the supplier's production conditions also affect the selection of sampling systems, sampling plans and sampling schemes. These include:
() The supplier's product quality has always been good. (h) The random sample is selected so that each unit of product has a chance to be sampled. (i) Testing is fast and inexpensive.
(j) It is necessary to know the shape, location and width of the distribution of product characteristics. (k) The product characteristics are normally distributed (or can be converted to a normal distribution). The opposite situation is shown in Table 2h5
Tables 1, 2a and 2h summarize the sampling standards and schemes that can be used under different testing platforms, conditions that affect selection, market conditions and production characteristics. Table 3 includes some notes on the use of Tables 2a and 2bh. Table 1 Indicators for selecting the appropriate acceptance system, plan and program Inspection site
Dangerous goods
Final inspection
Quality history of unknown products
Average delivery quality upper limit
Small batches
Quality of received products is good
Periodic inspection
Work in process submitted successively
Conditions affecting the selection of the program
FLrererefic.
Available sampling programs
100 inspection
(Counting)AQL program
Here AQI, for program
|tt||(Note number) Standard type plan
【Basis of measurement) Standard type treatment
(Counting) AQL piece plan
(Measurement) AQL piece plan
Batch skipping plan
(Counting) Standard type plan
(Counting) Standard type piece plan
AQI. plan
(Counting) AQL plan
(Counting) Standard type plan
Batch skipping plan
(Counting) RQ. plan
Continuous sampling plan
Related sampling standard or plan
Heart 32828
5 6378
GT 8051.GR/T13364
GH 8053,GB 8054
GI2828
GB 6378
H/T13263
CB 8C31.GB/T 13262
CB 8054
Select type
GB 2828
GB/T 13264
GB/T 13263
GH2829
GB 8052
W.GBT 13393--92
Table 2a Guidance for selecting a sampling and acceptance system, plan or scheme based on market conditions and production characteristics Market conditions or production characteristics
(8) The consignee can improve product quality by counteracting the supplier (h) After the product is accepted, the supplier is still responsible () Accepting a small number of unqualified products may cause significant losses (a) Accepting a batch of unqualified products may cause economic losses and factory closure (e) The cost of inspecting products is high
(i) The inspection tape is not easily damaged
(g) The product quality has always been good
) It is easy to draw a random sample
(i) The inspection speed is fast
(i) It is necessary to know the shape, location and width of the distribution of product characteristics (k) The distribution of product characteristics is normal
Sampling standards or schemes that can be used
GB 2828, GB/T 13263, GB 6378,GB 8051CB2828,CB/T13263.C36378.CR8031GB 8051,CB/T 13264.CB/T 13262ACHQI. Scheme
GB 6378
GB 637R
GR/T 13263
GB2828's - or - scheme, G138051(FH 2828's - or - scheme, GT 895;GB6378
GR 6378,GB 8053,GB R064
Guidelines for selecting a sampling acceptance system, plan or program based on market conditions and production characteristics Market conditions or production characteristics
However, the sampling standard practice plan
《）The consignee cannot influence the quality of the supplied product through feedback, and the certified batch or batches of B8051, GB/13264, (ie 8053, batch
（b）Once the product is accepted, the supplier is no longer responsible (L) The acceptance of insufficient products will not cause significant sharpness () There are many sources of equivalent products
（t）The cost of testing a unit of product is not high (E,) The test is not destructive
（g.）There is no history of product quality on the supplier, or the history is not good (hi)) Random sampling is difficult or expensive (i,) The test takes a long time or is expensive
（i）The shape, location and width of the distribution of product characteristics are not important (1) The distribution of product characteristics is unknown or non-normal GH N054GB/T 13262
GH 8051+GB/T 13264.GR 8053,
CB 8054,GB/T 13262
GB 2828,GB 6378,GB 8051,
GB/T13263
Specially consider other factors
Consider other factors
Consider other factors
GB2828G8051.GB6378's tightened inspectionG:R 2828's one-time sampling,CB 6378
GB2828's one-time sampling,G乃6378
Count sampling,Sampling by Count (Considering Other Factors) Sampling by Count
atb
d .crfr-jr
G1/T1339392
Table 3 Guidance Notes to Tables 2a and 2h
All sampling standards can be used (see 2). The selection of the appropriate plan also depends on other factors. Therefore, for a longer process, where the quality of individual batches is more important, a standard sampling plan of counting (or measurement) can be used. To make the sampling plan have appropriate discrimination (the discrimination ratio should be lower) (see A6.3), some fields (such as whitening inspection) can adopt 100% inspection and require a single sampling plan.All sampling standards can be used (see 2). The selection of the appropriate plan also depends on other factors. Therefore, a longer process is required. The identification ratio of the sampling plan can be appropriately larger, and the sample size can be appropriately smaller. No additional comments
If it is a batch of violations, use the transfer rule and start with strict inspection. When the quality is obtained, it will be transferred to normal inspection. When sampling is difficult, it is best to use a single sampling. Two or five samplings can be used in conjunction with the sub-sample selection rules. Sequential sampling can adopt the tail-cutting method
Use a small sample with a higher risk. However, in the case of long-term quality control, all counting sampling standards can be used. The selection of the appropriate plan also depends on other factors. If the inspection is quantitative, the quantitative value can be converted into a counting value. If information about the distribution can be obtained in the long term, quantitative sampling can be used. In the case of initial delivery, there is no information about the distribution, and the use of measurement sampling may increase the risk. Note: When the supplier is still responsible for the batch that is judged to be accepted after sampling inspection (see 3.1 (b)), GB2828GH8051 or GB/T13263 can be used in the case of counting inspection: GB6378 can be used in the case of measurement inspection (see Table 2a). If the supplier is no longer responsible after the product is accepted, in order to ensure the quality of the batch, GBB051GB/T13264, GB/T132621 can be used in the case of counting inspection; GB138053 or GB8051 can be used in the case of counting inspection (see Table 2b). The selection of the appropriate plan also depends on other factors (see Table 3). ) In the case of a single source of products, where rejection of a batch may result in economic losses or factory closure (see 3.1 (d)), if the inspection cost is not high and there is no destructiveness (see 3.1 (+), 5.1f, >), the A)QI scheme (see Table 2a (d)) can be used to average out the upper quality limit (ACQI is a quality parameter. When the product is rejected, all of them shall be inspected and the defective products in the batch shall be replaced by qualified products. If the source of equivalent products is rare (see 3.1 (j)), after considering other factors, select the appropriate scheme; W.A1 Negotiation procedure for selecting sampling plan
GB/T 73393-92
Appendix A
(Supplement)
The supplier and the consignee shall select a reasonable sampling and acceptance plan based on the product quality requirements, delivery and acceptance regulations through consultation. The negotiation procedure is as follows:
. a. Clarify the criteria for whether the unit product meets the requirements and what are qualified products and unqualified products. b. Clarify whether to ensure the unit product is qualified or the batch is qualified. If it is necessary to ensure that each unit product is qualified, the sampling acceptance cannot be adopted. c. The selection of the sampling plan should be based on the risk factors acceptable to both the supplier and the consignee. At the same time, considering other factors such as the average sample size, the supplier should understand the probability of being accepted when the batch quantity meets the requirements; the consignee should understand the probability of the batch quality being inferior to a certain specified quality. The probability of being accepted at the time,
. Clarify the sampling plan and the criteria for whether the batch is qualified or not. A2 Defective product rate inspection and defective number inspection per 100 units of products In order to decide whether to use defective product rate inspection or defective number inspection per 100 units of products, the following factors should be considered: a. The defective product rate inspection is concerned with whether there are defective units on the unit products. If there is one defective unit product, multiple defective units accumulated will be treated as qualified products: 5. The number inspection per 100 units of products requires counting each defective unit on the inspected unit. For example, there may be countless defective units on a unit product (such as defects or pinholes on a casting), and the number of defective units per 100 units should be used. An inspection mechanism is used (defective product rate pick-up whether It is better to use one method (number of nonconformities per 100 units) than to change from one method to another frequently. Record keeping is important for improving product quality. From the point of view of record keeping, number of nonconformities per 100 units is better, as such records contain all the information related to nonconformities. If the nonconformity rate is used, some nonconformities may be missed. e: Unless the product quality is excellent or the conformity of each characteristic is excellent, the more types of nonconformities that may result in acceptance when the number of nonconformities per 100 units is used, the greater the chance that the batch will be rejected. A3 Risks to producers and users
The risks brought about by sampling acceptance can be calculated. Since the samples constitute only a small part of the batch, sampling acceptance may bring risks to both the producer and the user. The probability that a good batch may be rejected and make such an error is called the producer's risk (expressed by PR); conversely, a bad batch may be accepted, and the probability of making such an error is called the user's risk (expressed by (R). When the batch (or process) quality is known by the 0C curve of the sample plan, the probability of the batch (or process) being accepted can be read out. For a certain specified good quality, the producer requires acceptance with a high probability. This good quality is called the producer's risk quality (expressed by PRQ), and the corresponding rejection probability is the producer's risk; on the other hand, for a certain specified bad quality, the user requires acceptance with a low probability. This bad quality is the user's risk quality (expressed by CR), and the corresponding acceptance probability is the user's risk (see Figures A1 and A2). W. Various US standard industry materials free download cattle production and risk
User risk
User risk
GB/T13393—
4 sound force wind point (PRT)
State production risk
Quality ()
Use risk point Ck
Use k risk
Quality (G)
Batch (property management) quality
By The curve determined by the producer's risk quality (PRQ) and the user's risk quality (CRQ) is the risk drop point RP
User's risk point (CRP)
Acceptance quality
Level (AQ1.)
Limiting quality
Batch (process) quality
OC curve A4AQL, PRQ, LQ Selection of CRQ values A4.1 AQL and PRQ || tt || AQ1. and PRQ above are both permissible quality indicators specified for sampling inspection. The difference is that FRQ indicates the quality of a batch, while AQL indicates the quality level of a process or series of batches. When referring to AQI (or PRQ), the following considerations can be considered: The supplier's batch average quality (or batch quality) is required to be no worse than AQL (or PRQ). And this quality is achievable: For the receiving force. AQI. (or PRQ) is a reasonable quality, which is the compromise between meeting the requirements and being able to afford it. If the required product is easy to buy and the unqualified product can be replaced by a qualified product, the AQL value should be appropriately larger. On the contrary, if an unqualified product may cause the failure of an important component on the equipment and cannot be replaced by a qualified product, a smaller AQL (or PRQ) should be used. GB/T 1339392
If the whole piece includes 100 equally important parts, and the failure of any one of the parts will cause the whole piece to fail, the defective rate X of the whole piece and the defective rate of the parts have the following relationship: X=1--r)
If the percentage defective rate of the whole piece is X, and the percentage defective rate of the parts is, then 1001100)
dEstimate the process average by analyzing past data, and define this estimated value or a set of values slightly smaller than it as AQL. When the AQL is given, if the lower end of the OC curve of the selected sampling plan cannot meet the requirements of the user, a more stringent AQLn
"Don't always select the AQL first and then make other choices. If necessary, you can use the "reverse deduction" method to select a single plan based on other criteria, and then reverse the desired AQL value through the sampling plan table. When using the reverse method, you can consider (a certain important point on the line with a low probability of acceptance, or consider other economic criteria g) to set a known satisfactory quality level (or quality rating) of similar products as AQL (or PRQ). h: Temporarily specify an AQL (or IRQ). Adjust based on long-term usage and experience: i. Build a cost model and select AQ! or PRQ) to minimize your costs A4. 2 1Q and CRQ
1Q and (RQ) are for the purpose of axis inspection, and the remaining is limited to: the method of specifying LQ and CRQ for quality levels with low probability of acceptance is the same as specifying AQL and PRQ A5 Determination of inspection level
Initial inspection level (11.) The inspection quantity index of the sampling plan links the sample size to the batch. Specify {I. The following factors can be considered:
a After specifying AQ1. (or PRQ), the inspection level is determined by considering LQ (and CRQ). For example, in GB2828, first find the normal inspection (OC curve) that is closest to meeting the requirements of AQI. and 1.Q, and the AQL and judgment array corresponding to the OC curve, then find the corresponding sample size code in the normal inspection sampling plan table, and finally find the sample size code table that matches this sample.The inspection level corresponding to the size code and the given batch range. If there is no inspection level in the table corresponding to the code and batch range, the AQL and Q value should be reconsidered, or this practice should be abandoned. For economic reasons or the destructive nature of the test, a low inspection level, such as S1 in GB2828, may have to be used. In this case, the discrimination of the sampling plan is reduced. However, if there are inspection records for a series of consecutive batches, the cumulative results of the samples may indicate that the inspection level is acceptable.
C. GB8053 and GB8054 cards are used for isolated batch inspections, and the inspection level is not used to retrieve the sampling plan. In this case, the manufacturer and the user should pay attention to LQ.
A6 Comparison of sampling inspection methods
A6.1 Count sampling inspection and count-based sampling inspection If the observed values follow a normal distribution, variable sampling inspection has certain advantages over count sampling inspection. In order to provide some quality protection for the producer (or user), the former requires a smaller sample and can provide more information about product quality; the latter is not limited by the distribution shape. Free download of various US standard industry materials.
CB/T13393-92
When the inspection cost is very high or the inspection is perishable, the use of measurement sampling inspection is essentially superior to the counting sampling inspection. If there are many inspection indicators, it is also possible to consider using counting sampling inspection for most of the inspection indicators and using variable sampling inspection for one important indicator. The following table gives the sample size corresponding to several sample size codes of GB 2828 and GB 6378 when using inspection level II and normal inspection sampling plan: Sample size code GB2828 A6.2: Secondary, multiple and sequential sampling plan CB6378 (a unknown) GB 6378 (a unknown) Second, we know:
not more than 3
not more than 14
days more than 33
not more than 95
For the sampling plan with a small number of acceptances, we can judge a one-time, two-time or multiple sampling plan so that their (C curves are very close. Therefore, it is not enough to consider the C curve only to compare the one-time, two-time and multiple sampling plans. Consider the following characteristics comprehensively. H. Simple single-time sampling plan is the easiest to explain and use. The elimination sampling plan may extract the first argument, but it is troublesome to manage. The multiple sampling plan is more complicated than the sequential sampling plan. The variation of the test quantity is accompanied by the detailed plan. The size of the sample is fixed, so the test quantity can be predicted. The test defects of multiple, multiple and sequential sampling schemes vary with the test results of the previous sample; however, for any given access quality, the average number of samples can be calculated. The average number of samples is the least when the input quality is particularly good or particularly high: c. Difficulty in sampling Sometimes it is convenient to draw a second sample. Two samples are no more troublesome than drawing a sample with a smaller combined sample size. However, sometimes sampling is time-consuming. After drawing the first sample, it is difficult to shuffle the rest of the batch to draw the third sample. In this case, it is best to use single sampling.
d. The duration of the test If the test duration is long and there is If you can test many samples at the same time, a single sampling plan is better. If you can only test one or two samples at a time, you can consider using a secondary, multiple or sequential sampling plan. Types of defects If there are many types of product defects, secondary and multiple sampling will become complicated. For example, all characteristics of the product are required to be tested on the first sample, and some of them may also need to be tested on the second sample, the first sample, etc. At this time, labor and inspection equipment can be used effectively. Generally speaking, complex inspections require simple sampling plans, and more complex sampling plans can be used for simple inspections, which can achieve good results. For example: In order to inspect the preservation quality of meat cans, the following The cans are placed in special environmental conditions for three weeks. Three sampling plans are considered: a single sampling plan of 80 cans, a sub-sampling plan of each sample including 50 cans, and a five-time sampling plan of each sample containing 20 cans. If a sub-sampling plan is used, the test results can be obtained in three weeks; using a secondary sampling plan, the results may be obtained in three weeks, or it may take six weeks: using a sub-sampling plan, it may take three and a half months to test, and it is even more unrealistic if a sequential test is used. Therefore, it is appropriate to choose a bundle sampling plan.
Example 2: In order to conduct a destructive test, the test station can obtain all the products in the batch, but the test equipment can only accommodate one product at a time. Since the main cost is the destruction of the test product, it is hoped that the destruction can be reduced as much as possible while meeting the requirements for sampling characteristics. Only one product can be tested each time, so multiple or sequential sampling can save time and worry. 60
GB/T 1339392
Drinking and five-time sampling scheme
P/ AQL
-drinking,... The average sample size curve of the two-time and five-time sampling schemes is given in GB2828. For the sequential sampling scheme, the average sampling number table of the truncated sequential scheme is given in GB8051. Figure 4 shows the average sample size curve of the one-time and five-time sampling schemes equivalent to the one-time sampling scheme with a sample size of 200, an acceptance number of 3, and an acceptance number of 4.
Figure A4 Average sample size curves for the two-time and five-time sampling schemes equivalent to the one-time scheme (code L, AQL = 0.65% + percentage of defective products submitted) As can be seen from Figure A1, when the quality of the submitted inspection batch is the best or the quality is poor, the average sample size can be saved by 50% or more. The number of unit products to be inspected in the one-time and two-time sampling schemes has an upper limit, while the sequential sampling scheme and the non-introduction of the truncation rule generally have no such limit.
Two-time, five-time and sequential sampling provide opportunities for significant savings in sample size, but due to the difficulties in use and management, five-time and sequential sampling are rarely used. With the improvement of management level and the degree of automated production, five-time and sequential sampling have gradually been widely used.
A6.3 The role of OC curve
Each sampling scheme has a corresponding UC curve. In order to compare different sampling schemes, their OC curves can be compared, or the C curve CB/T1339392
CRQ/PRQ is a numerical indicator of the discrimination ratio. The larger this value is, the flatter the C curve is, and the lower the discrimination of the sampling scheme is; the smaller this number is, the steeper the curve is, and the higher the discrimination of the sampling scheme is. This indicator can be used to select and compare sampling schemes A7 The benefits of adopting national standards for sampling acceptance The main driving force for adopting sampling acceptance is economy, which not only ensures that the received batch has a certain quality, but also significantly saves inspection workload and inspection costs.
Sampling acceptance mainly targets risk. Both the producer and the user hope to minimize their own risks. This requires the rational selection of sampling schemes, Control each other's risks. It is beneficial to both parties to stipulate some statistically complete acceptance rules in the product standards and appropriately use the national standards for sampling acceptance.
Non-standardized sampling methods (such as white ratio sampling, etc.) often lack theoretical basis. The national standards for sampling acceptance have reliable theoretical basis. The use of national standards for sampling acceptance can save time for objective discussions, which is particularly important in international trade. Additional notes:
This standard was jointly proposed by the Ministry of Machinery and Electronics Industry and the National Technical Committee for the Application of Statistics and Standardization. The standard was jointly proposed by the Electronic Standardization Research Institute of the Ministry of Machinery and Electronics Industry and the Institute of Systems Science of the Chinese Academy of Sciences. The main drafters of this standard are Shang Yulin, Ma Yilin and Yan Qingzi.Free download of various US standard industry materials.
CB/T13393-92
When the inspection cost is very high or the inspection is subject to corruption, the use of measurement sampling inspection is essentially superior to the counting sampling inspection. If there are many inspection indicators, it is also possible to consider using counting sampling inspection for most of the inspection indicators and using measurement sampling inspection for one important indicator.
The following table gives the sample size corresponding to the sample size when using inspection level II and normal inspection sampling force case in GB 2828 and GB 6378:
Sample size code
GB2828
A6.2: Secondary, multiple and sequential sampling force case
CB6378a unknown)
GB 6378(a unknown) Second, we know:
not more than 3
not more than 14
days more than 33
not more than 95
For the sampling plan with a small number of acceptances, we can judge a one-time, two-time or multiple sampling plan so that their (C curves are very close. Therefore, it is not enough to consider the C curve only to compare the one-time, two-time and multiple sampling plans. Consider the following characteristics comprehensively. H. Simple single-time sampling plan is the easiest to explain and use. The elimination sampling plan may extract the first argument, but it is troublesome to manage. The multiple sampling plan is more complicated than the sequential sampling plan. The variation of the test quantity is accompanied by the detailed plan. The size of the sample is fixed, so the test quantity can be predicted. The test defects of multiple, multiple and sequential sampling schemes vary with the test results of the previous sample; however, for any given access quality, the average number of samples can be calculated. The average number of samples is the least when the input quality is particularly good or particularly high: c. Difficulty in sampling Sometimes it is convenient to draw a second sample. Two samples are no more troublesome than drawing a sample with a smaller combined sample size. However, sometimes sampling is time-consuming. After drawing the first sample, it is difficult to shuffle the rest of the batch to draw the third sample. In this case, it is best to use single sampling.
d. The duration of the test If the test duration is long and there is If you can test many samples at the same time, a single sampling plan is better. If you can only test one or two samples at a time, you can consider using a secondary, multiple or sequential sampling plan. Types of defects If there are many types of product defects, secondary and multiple sampling will become complicated. For example, all characteristics of the product are required to be tested on the first sample, and some of them may also need to be tested on the second sample, the first sample, etc. At this time, labor and inspection equipment can be used effectively. Generally speaking, complex inspections require simple sampling plans, and more complex sampling plans can be used for simple inspections, which can achieve good results. For example: In order to inspect the preservation quality of meat cans, the following The cans are placed in special environmental conditions for three weeks. Three sampling plans are considered: a single sampling plan of 80 cans, a sub-sampling plan of each sample including 50 cans, and a five-time sampling plan of each sample containing 20 cans. If a sub-sampling plan is used, the test results can be obtained in three weeks; using a secondary sampling plan, the results may be obtained in three weeks, or it may take six weeks: using a sub-sampling plan, it may take three and a half months to test, and it is even more unrealistic if a sequential test is used. Therefore, it is appropriate to choose a bundle sampling plan.
Example 2: In order to conduct a destructive test, the test station can obtain all the products in the batch, but the test equipment can only accommodate one product at a time. Since the main cost is the destruction of the test product, it is hoped that the destruction can be reduced as much as possible while meeting the requirements for sampling characteristics. Only one product can be tested each time, so multiple or sequential sampling can save time and worry. 60
GB/T 1339392
Drinking and five-time sampling scheme
P/ AQL
-drinking,... The average sample size curve of the two-time and five-time sampling schemes is given in GB2828. For the sequential sampling scheme, the average sampling number table of the truncated sequential scheme is given in GB8051. Figure 4 shows the average sample size curve of the one-time and five-time sampling schemes equivalent to the one-time sampling scheme with a sample size of 200, an acceptance number of 3, and an acceptance number of 4.
Figure A4 Average sample size curves for the two-time and five-time sampling schemes equivalent to the one-time scheme (code L, AQL = 0.65% + percentage of defective products submitted) As can be seen from Figure A1, when the quality of the submitted inspection batch is the best or the quality is poor, the average sample size can be saved by 50% or more. The number of unit products to be inspected in the one-time and two-time sampling schemes has an upper limit, while the sequential sampling scheme and the non-introduction of the truncation rule generally have no such limit.
Two-time, five-time and sequential sampling provide opportunities for significant savings in sample size, but due to the difficulties in use and management, five-time and sequential sampling are rarely used. With the improvement of management level and the degree of automated production, five-time and sequential sampling have gradually been widely used.
A6.3 The role of OC curve
Each sampling scheme has a corresponding UC curve. In order to compare different sampling schemes, their OC curves can be compared, or the C curve CB/T1339392
CRQ/PRQ is a numerical indicator of the discrimination ratio. The larger this value is, the flatter the C curve is, and the lower the discrimination of the sampling scheme is; the smaller this number is, the steeper the curve is, and the higher the discrimination of the sampling scheme is. This indicator can be used to select and compare sampling schemes A7 The benefits of adopting national standards for sampling acceptance The main driving force for adopting sampling acceptance is economy, which not only ensures that the received batch has a certain quality, but also significantly saves inspection workload and inspection costs.
Sampling acceptance mainly targets risk. Both the producer and the user hope to minimize their own risks. This requires the rational selection of sampling schemes, Control each other's risks. It is beneficial to both parties to stipulate some statistically complete acceptance rules in the product standards and appropriately use the national standards for sampling acceptance.
Non-standardized sampling methods (such as white ratio sampling, etc.) often lack theoretical basis. The national standards for sampling acceptance have reliable practical basis. The use of national standards for sampling acceptance can save time for objective discussions, which is particularly important in international trade. Additional notes:
This standard was jointly proposed by the Ministry of Machinery and Electronics Industry and the National Technical Committee for the Application of Statistics and Standardization. The standard was jointly proposed by the Electronic Standardization Research Institute of the Ministry of Machinery and Electronics Industry and the Institute of Systems Science of the Chinese Academy of Sciences. The main drafters of this standard are Shang Yulin, Ma Yilin and Yan Qingzi.Free download of various US standard industry materials.
CB/T13393-92
When the inspection cost is very high or the inspection is subject to corruption, the use of measurement sampling inspection is essentially superior to the counting sampling inspection. If there are many inspection indicators, it is also possible to consider using counting sampling inspection for most of the inspection indicators and using measurement sampling inspection for one important indicator.
The following table gives the sample size corresponding to the sample size when using inspection level II and normal inspection sampling force case in GB 2828 and GB 6378:
Sample size code
GB2828
A6.2: Secondary, multiple and sequential sampling force case
CB6378a unknown)
GB 6378(a unknown) Second, we know:
not more than 3
not more than 14
days more than 33
not more than 95
For the sampling plan with a small number of acceptances, we can judge a one-time, two-time or multiple sampling plan so that their (C curves are very close. Therefore, it is not enough to consider the C curve only to compare the one-time, two-time and multiple sampling plans. Consider the following characteristics comprehensively. H. Simple single-time sampling plan is the easiest to explain and use. The elimination sampling plan may extract the first argument, but it is troublesome to manage. The multiple sampling plan is more complicated than the sequential sampling plan. The variation of the test quantity is accompanied by the detailed plan. The size of the sample is fixed, so the test quantity can be predicted. The test defects of multiple, multiple and sequential sampling schemes vary with the test results of the previous sample; however, for any given access quality, the average number of samples can be calculated. The average number of samples is the least when the input quality is particularly good or particularly high: c. Difficulty in sampling Sometimes it is convenient to draw a second sample. Two samples are no more troublesome than drawing a sample with a smaller combined sample size. However, sometimes sampling is time-consuming. After drawing the first sample, it is difficult to shuffle the rest of the batch to draw the third sample. In this case, it is best to use single sampling.
d. The duration of the test If the test duration is long and there is If you can test many samples at the same time, a single sampling plan is better. If you can only test one or two samples at a time, you can consider using a secondary, multiple or sequential sampling plan. Types of defects If there are many types of product defects, secondary and multiple sampling will become complicated. For example, all characteristics of the product are required to be tested on the first sample, and some of them may also need to be tested on the second sample, the first sample, etc. At this time, labor and inspection equipment can be used effectively. Generally speaking, complex inspections require simple sampling plans, and more complex sampling plans can be used for simple inspections, which can achieve good results. For example: In order to inspect the preservation quality of meat cans, the following The cans are placed in special environmental conditions for three weeks. Three sampling plans are considered: a single sampling plan of 80 cans, a sub-sampling plan of each sample including 50 cans, and a five-time sampling plan of each sample containing 20 cans. If a sub-sampling plan is used, the test results can be obtained in three weeks; using a secondary sampling plan, the results may be obtained in three weeks, or it may take six weeks: using a sub-sampling plan, it may take three and a half months to test, and it is even more unrealistic if a sequential test is used. Therefore, it is appropriate to choose a bundle sampling plan.
Example 2: In order to conduct a destructive test, the test station can obtain all the products in the batch, but the test equipment can only accommodate one product at a time. Since the main cost is the destruction of the test product, it is hoped that the destruction can be reduced as much as possible while meeting the requirements for sampling characteristics. Only one product can be tested each time, so multiple or sequential sampling can save time and worry. 60
GB/T 1339392
Drinking and five-time sampling scheme
P/ AQL
-drinking,... The average sample size curve of the two-time and five-time sampling schemes is given in GB2828. For the sequential sampling scheme, the average sampling number table of the truncated sequential scheme is given in GB8051. Figure 4 shows the average sample size curve of the one-time and five-time sampling schemes equivalent to the one-time sampling scheme with a sample size of 200, an acceptance number of 3, and an acceptance number of 4.
Figure A4 Average sample size curves for the two-time and five-time sampling schemes equivalent to the one-time scheme (code L, AQL = 0.65% + percentage of defective products submitted) As can be seen from Figure A1, when the quality of the submitted inspection batch is the best or the quality is poor, the average sample size can be saved by 50% or more. The number of unit products to be inspected in the one-time and two-time sampling schemes has an upper limit, while the sequential sampling scheme and the non-introduction of the truncation rule generally have no such limit.
Two-time, five-time and sequential sampling provide opportunities for significant savings in sample size, but due to the difficulties in use and management, five-time and sequential sampling are rarely used. With the improvement of management level and the degree of automated production, five-time and sequential sampling have gradually been widely used.
A6.3 The role of OC curve
Each sampling scheme has a corresponding UC curve. In order to compare different sampling schemes, their OC curves can be compared, or the C curve CB/T1339392
CRQ/PRQ is a numerical indicator of the discrimination ratio. The larger this value is, the flatter the C curve is, and the lower the discrimination of the sampling scheme is; the smaller this number is, the steeper the curve is, and the higher the discrimination of the sampling scheme is. This indicator can be used to select and compare sampling schemes A7 The benefits of adopting national standards for sampling acceptance The main driving force for adopting sampling acceptance is economy, which not only ensures that the received batch has a certain quality, but also significantly saves inspection workload and inspection costs.
Sampling acceptance mainly targets risk. Both the producer and the user hope to minimize their own risks. This requires the rational selection of sampling schemes, Control each other's risks. It is beneficial to both parties to stipulate some statistically complete acceptance rules in the product standards and appropriately use the national standards for sampling acceptance.
Non-standardized sampling methods (such as white ratio sampling, etc.) often lack theoretical basis. The national standards for sampling acceptance have reliable theoretical basis. The use of national standards for sampling acceptance can save time for objective discussions, which is particularly important in international trade. Additional notes:
This standard was jointly proposed by the Ministry of Machinery and Electronics Industry and the National Technical Committee for the Application of Statistics and Standardization. The standard was jointly proposed by the Electronic Standardization Research Institute of the Ministry of Machinery and Electronics Industry and the Institute of Systems Science of the Chinese Academy of Sciences. The main drafters of this standard are Shang Yulin, Ma Yilin and Yan Qingzi.Simplicity Single-shot sampling is easiest to explain and use. Multiple-shot sampling is more complex than sequential sampling. Variation of test quantity The size of the test quantity is fixed for multiple-shot sampling, so the test quantity can be predicted. The test defects of multiple-shot, multiple-shot and sequential sampling vary with the results of the previous sample; but for any given access quality, the average number of samples can be calculated, and the average number of samples is the smallest when the input is particularly good or particularly difficult: c. Difficulty of sampling Sometimes it is convenient to draw a second sample, and two samples are no more difficult than drawing one sample with a smaller combined sample size. However, sometimes sampling is time-consuming and it is difficult to draw the first sample and then shuffle the rest of the batch to draw the third sample. In this case, it is best to use single-shot sampling.
d, Duration of the test If the test duration is long and it is possible to test more samples at the same time, a single sampling plan is better. If only two samples can be tested at a time, consider using a secondary, multiple or sequential sampling plan. , Types of defects If there are many types of product defects, secondary and multiple sampling will become complicated. For example, all characteristics of the product are required to be tested in the first sample, and some of the characteristics may also need to be tested in the second sample, the first sample, etc. At this time, the labor and inspection equipment can be used effectively. Generally speaking, complex inspections require simple sampling plans, and more complex sampling plans can be used for simple inspections to achieve good results.
Example: In order to test the preservation quality of meat cans, the cans are placed in special environmental conditions for three weeks. Consider the selection of three sampling plans: a single sampling plan for sampling 80 cans, a sub-sampling plan for sampling each sample including 50 cans, and a five-time sampling plan for each sample containing 20 cans. If a sub-sampling plan is used, the test results can be obtained within three weeks; using a secondary sampling plan, the results may be obtained within three weeks, or it may take more than one week. Six weeks: If we use multiple sampling, it may take three and a half months to test. If we use sequential testing, it will be even more unrealistic. Therefore, it is appropriate to choose a batch sampling plan. Example 2: In order to conduct a destructive test, the test station can obtain all the products in the batch. However, the test equipment can only accommodate one product at a time. Since the main cost is the destruction of the test product, we hope to reduce the destruction as much as possible while meeting the requirements for sampling characteristics. Only one product can be tested each time. Using multiple or sequential sampling can save time and worry. The number of samples, 60 GB/T 1339392 Drinking and Five Changes P/ AQL
- ... Reach or exceed 50%. The number of unit products to be inspected by the one-time and five-time sampling schemes has an upper limit, while the sequential sampling scheme and the non-introduction of the truncation rule generally have no such limit.
Two-time, five-time and sequential sampling provide opportunities for significant savings in sample size, but due to the difficulties in use and management, five-time and sequential sampling are used less frequently. With the improvement of management level and the degree of automated production, five-time and sequential sampling have gradually been widely used.
A6.3 The role of OC curve
Each sampling scheme has a corresponding UC curve. In order to compare different sampling schemes, their OC curves can be compared, or by C curve CB/T1339392
CRQ/PRQ is a numerical indicator of discrimination ratio. The larger this value is, the flatter the C curve is, and the lower the discrimination of the sampling scheme is; the smaller this number is, the steeper the UC curve is, and the higher the discrimination of the sampling scheme is. This index can be used to select and compare sampling plans. A7 The benefits of adopting national standards for sampling acceptance The main motivation for adopting sampling acceptance is economy, which not only ensures that the received batch has a certain quality, but also significantly saves inspection workload and inspection costs.
The main purpose of sampling acceptance is risk. Both the manufacturer and the user want to minimize their own risks. This requires the rational selection of sampling plans to control their respective risks. It is beneficial to both parties to stipulate some statistically complete acceptance rules in the product standards and appropriately use the national standards for sampling acceptance.
Non-standardized sampling methods (such as white ratio sampling, etc.) often lack theoretical basis. The national standards for sampling acceptance have reliable empirical basis. The adoption of national standards for sampling acceptance can save time for objective discussions, which is particularly important in international trade. Additional remarks:
This standard was jointly proposed by the Ministry of Machinery and Electronics Industry and the National Technical Committee for the Application of Statistics and Standardization. The standard was jointly issued by the Electronic Standardization Institute of the Ministry of Machinery and Electronics Industry and the Institute of Systems Science of the Chinese Academy of Sciences. The main drafters of the standard are Shang Yulin, Ma Yilin and Yan Qingzi.Simplicity Single-shot sampling is easiest to explain and use. Multiple-shot sampling is more complex than sequential sampling. Variation of test quantity The size of the test quantity is fixed for multiple-shot sampling, so the test quantity can be predicted. The test defects of multiple-shot, multiple-shot and sequential sampling vary with the results of the previous sample; but for any given access quality, the average number of samples can be calculated, and the average number of samples is the smallest when the input is particularly good or particularly difficult: c. Difficulty of sampling Sometimes it is convenient to draw a second sample, and two samples are no more difficult than drawing one sample with a smaller combined sample size. However, sometimes sampling is time-consuming and it is difficult to draw the first sample and then shuffle the rest of the batch to draw the third sample. In this case, it is best to use single-shot sampling.
d, Duration of the test If the test duration is long and it is possible to test more samples at the same time, a single sampling plan is better. If only two samples can be tested at a time, consider using a secondary, multiple or sequential sampling plan. , Types of defects If there are many types of product defects, secondary and multiple sampling will become complicated. For example, all characteristics of the product are required to be tested in the first sample, and some of the characteristics may also need to be tested in the second sample, the first sample, etc. At this time, the labor and inspection equipment can be used effectively. Generally speaking, complex inspections require simple sampling plans, and more complex sampling plans can be used for simple inspections to achieve good results.
Example: In order to test the preservation quality of meat cans, the cans are placed in special environmental conditions for three weeks. Consider the selection of three sampling plans: a single sampling plan for sampling 80 cans, a sub-sampling plan for sampling each sample including 50 cans, and a five-time sampling plan for each sample containing 20 cans. If a sub-sampling plan is used, the test results can be obtained within three weeks; using a secondary sampling plan, the results may be obtained within three weeks, or it may take more than one week. Six weeks: If we use multiple sampling, it may take three and a half months to test. If we use sequential testing, it will be even more unrealistic. Therefore, it is appropriate to choose a batch sampling plan. Example 2: In order to conduct a destructive test, the test station can obtain all the products in the batch. However, the test equipment can only accommodate one product at a time. Since the main cost is the destruction of the test product, we hope to reduce the destruction as much as possible while meeting the requirements for sampling characteristics. Only one product can be tested each time. Using multiple or sequential sampling can save time and worry. The number of samples, 60 GB/T 1339392 Drinking and Five Changes P/ AQL
- ... Reach or exceed 50%. The number of unit products to be inspected by the one-time and five-time sampling schemes has an upper limit, while the sequential sampling scheme and the non-introduction of the truncation rule generally have no such limit.
Two-time, five-time and sequential sampling provide opportunities for significant savings in sample size, but due to the difficulties in use and management, five-time and sequential sampling are used less frequently. With the improvement of management level and the degree of automated production, five-time and sequential sampling have gradually been widely used.
A6.3 The role of OC curve
Each sampling scheme has a corresponding UC curve. In order to compare different sampling schemes, their OC curves can be compared, or by C curve CB/T1339392
CRQ/PRQ is a numerical indicator of discrimination ratio. The larger this value is, the flatter the C curve is, and the lower the discrimination of the sampling scheme is; the smaller this number is, the steeper the UC curve is, and the higher the discrimination of the sampling scheme is. This index can be used to select and compare sampling plans. A7 The benefits of adopting national standards for sampling acceptance The main motivation for adopting sampling acceptance is economy, which not only ensures that the received batch has a certain quality, but also significantly saves inspection workload and inspection costs.
The main purpose of sampling acceptance is risk. Both the manufacturer and the user want to minimize their own risks. This requires the rational selection of sampling plans to control their respective risks. It is beneficial to both parties to stipulate some statistically complete acceptance rules in the product standards and appropriately use the national standards for sampling acceptance. bzxz.net
Non-standardized sampling methods (such as white ratio sampling, etc.) often lack theoretical basis. The national standards for sampling acceptance have reliable empirical basis. The adoption of national standards for sampling acceptance can save time for objective discussions, which is particularly important in international trade. Additional remarks:
This standard was jointly proposed by the Ministry of Machinery and Electronics Industry and the National Technical Committee for the Application of Statistics and Standardization. The standard was jointly issued by the Electronic Standardization Institute of the Ministry of Machinery and Electronics Industry and the Institute of Systems Science of the Chinese Academy of Sciences. The main drafters of the standard are Shang Yulin, Ma Yilin and Yan Qingzi.The larger this value is, the slower the curve is, and the lower the discrimination of the sampling scheme is; the smaller this number is, the steeper the curve is, and the higher the discrimination of the sampling scheme is. This indicator can be used to select and compare sampling schemes. A7 The benefits of adopting national standards for sampling acceptance The main driving force for adopting sampling acceptance is economy, which not only ensures that the accepted batch has a certain quality, but also significantly saves inspection workload and inspection costs.
Sampling acceptance mainly targets risk. Both the manufacturer and the user hope to minimize their own risks. This requires the rational selection of sampling schemes to control their respective risks. In the product standard, some statistically complete acceptance rules are specified, which are suitable for When using the national standard for sampling acceptance, it is beneficial to both parties.
Non-standardized sampling methods (such as white ratio sampling, etc.) often lack theoretical basis, while the national standard for sampling acceptance has a reliable theoretical basis. The national standard for sampling acceptance can save time for discussion, which is particularly important in international trade. Additional remarks:
This standard was jointly proposed by the Ministry of Machinery and Electronics Industry and the National Technical Committee for the Application of Statistics and Standardization. The standard was jointly proposed by the Electronic Standardization Research Institute of the Ministry of Machinery and Electronics Industry and the Institute of Systems Science of the Chinese Academy of Sciences. The main drafters of the standard are Shang Yulin, Ma Yilin and Yan Qingzi.The larger this value is, the slower the curve is, and the lower the discrimination of the sampling scheme is; the smaller this number is, the steeper the curve is, and the higher the discrimination of the sampling scheme is. This indicator can be used to select and compare sampling schemes. A7 The benefits of adopting national standards for sampling acceptance The main driving force for adopting sampling acceptance is economy, which not only ensures that the accepted batch has a certain quality, but also significantly saves inspection workload and inspection costs.
Sampling acceptance mainly targets risk. Both the manufacturer and the user hope to minimize their own risks. This requires the rational selection of sampling schemes to control their respective risks. In the product standard, some statistically complete acceptance rules are specified, which are suitable for When using the national standard for sampling acceptance, it is beneficial to both parties.
Non-standardized sampling methods (such as white ratio sampling, etc.) often lack theoretical basis, while the national standard for sampling acceptance has a reliable theoretical basis. The national standard for sampling acceptance can save time for discussion, which is particularly important in international trade. Additional remarks:
This standard was jointly proposed by the Ministry of Machinery and Electronics Industry and the National Technical Committee for the Application of Statistics and Standardization. The standard was jointly proposed by the Electronic Standardization Research Institute of the Ministry of Machinery and Electronics Industry and the Institute of Systems Science of the Chinese Academy of Sciences. The main drafters of the standard are Shang Yulin, Ma Yilin and Yan Qingzi.
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.