Single sampling procedures and tables for inspection having desired operating characteristics by variables for percent nonconforming
Some standard content:
National Standard of the People's Republic of China
Metric Standard for Percent Nonconforming Product
Single Sampling Inspection Procedures and Tables
Single Sampling Practices and Tables for Inspection Having Desired Operating Characteristic by Variables for Percent Nonconforming UDC 311. 213. 2
:658. 562.012,7
GB805387
This standard specifies the metrological standard for single sampling inspection procedures and implementation methods with the batch nonconforming product rate as the quality indicator. It is applicable to batch inspections where the product quality characteristics are expressed in measurement values and obey or approximately obey the normal distribution. This standard stipulates that the producer's risk ±=0.05: the user's risk 8=0.101 Terminology and Symbols
1.1 Terminology
1.1.1 Unit product
The basic unit for dividing products for the purpose of implementing sampling inspection. 1.1.2 Inspection batch (abbreviation: batch)
Batch of unit products collected for sampling inspection. 1.1.3 Metrological quality characteristics
Quality characteristics of the inspected unit products that can be measured with a continuous scale. 1.1.4 Metrological sampling inspection
A process of randomly selecting a certain number of unit products from a batch according to a given sampling plan. Obtaining their quality characteristic values by measurement, testing or other methods. Comparing them with the quality requirements and judging whether the batch of products can be accepted. 1.1.5 Single sampling inspection
The process of sampling inspection in which only one sample is selected from the batch and the quality characteristic value of the sample is used to judge whether the batch of products can be accepted. 1.1.6 Random inspection characteristic curve (0C curve)
For a given sampling plan, a curve that represents the functional relationship between the batch quality level and its expected probability of acceptance. 1. 1.7 Acceptable quality
In sampling inspection, it corresponds to a certain, higher probability of acceptance, and is considered to be a satisfactory batch quality level. 1. 1. 8 Limit quality
In sampling inspection, it corresponds to a certain, lower probability of acceptance, and is considered to be a batch quality level that cannot be worse. 1.1. 9 Producer's risk
On the sampling characteristic curve, it corresponds to the probability of rejection at the acceptable quality. That is, the risk rate of a batch with acceptable quality being misjudged as an unqualified batch.
1.1.10 User's risk
On the sampling characteristic curve, it corresponds to the probability of acceptance at the limit quality point, that is, the risk rate of a batch with limit quality being misjudged as a qualified batch. 1.1.11 Standard sampling inspection
In order to protect the interests of both production and use, the producer's risk and the user's risk are fixed to specific values for sampling inspection. 1.1.12 Sampling plan
Approved by the National Bureau of Standards on July 8, 1987
Implementation on April 1, 1988
GB 8053.- 87
Stipulations on the sample size for sampling inspection and the judgment rules for determining whether a batch is acceptable. 1.1.13 Types of sampling inspection
When determining the sampling plan, the different situations of the statistical characteristic values of the batch quality are used to divide the sampling inspection types into two types: known batch standard deviation and unknown batch standard deviation.
1.1.14 "^\ method
When the batch standard deviation is known, the method of using the sample mean and batch standard deviation to judge whether the batch is acceptable. 1.1.15.\ method
When the batch standard deviation is unknown, the method of using the sample mean and sample standard deviation to judge whether the batch is acceptable. 1. 1. 1. 16 Quality statistics
A function consisting of specification limits, sample mean and batch standard deviation (or sample standard deviation) is used to determine whether a batch is acceptable. There are two types of quality statistics: 1. Specification limit and lower specification limit:
1.1.17 Acceptance constant
A constant used to determine whether a batch is accepted or not. This standard gives the minimum value of the upper specification limit quality statistics and (or) the lower specification limit quality statistics for an acceptable batch.
1.1.18 Inspection method
Inspection The inspection method is determined by the required quality specification limits. This standard has two methods: upper specification limit, lower specification limit and bilateral specification limit.
1.1.19 Specification limit
The limit value for determining whether a unit product is qualified. 1.1.20 Upper specification limit
Specifies the maximum limit value of the quality characteristics allowed for qualified products. 1.1.21 Lower specification limit
Specifies the minimum limit value of the quality characteristics allowed for qualified products. 1.1.22 Bilateral specification limit
Specifies the minimum limit value of the quality characteristics allowed for qualified products. |tt||The situation where both the upper specification limit and the lower specification limit are specified at the same time. 1.2 Symbols
u, upper specification limit
L: 'lower specification limit.
: Batch defective rate.
Qualified quality level when the batch defective rate is the quality indicator. Pa
: : The limit quality level when the batch defective rate is the quality indicator. : Producer risk.
User risk.
Sample size.
Sample The unit product quality characteristic value in represents the unit product quality characteristic value. Main: The average value of the sample quality characteristic value. Referred to as sample mean. t
: Standard deviation of the sample quality characteristic value. Referred to as sample standard deviation, y(-:
GB8053—87
t: Standard deviation of batch quality characteristic value. Referred to as batch standard deviation. : Statistical estimate or empirical estimate of batch standard deviation. Q, quality statistic of upper specification limit.
Qu-“*or 0 -\二"
Qn: quality statistic of the lower specification limit.
α=-or Q=—
! Number of acceptance bands
P acceptance probability of the inspection batch.
For other terms and symbols, see GB 3358-82 "Statistical Terms and Symbols". 2 Sampling inspection procedure
The procedure for implementing this standard is as follows:
a. Select the type of sampling inspection;
b. Determine the sampling inspection method:
Specify the qualified quality and the limit quality;
Determine the sampling plan;
Constitute the batch and draw samples:
Check the batch and the calculation results;
Judge whether the batch can be accepted:
Process the inspection batch.
3 Implementation of sampling inspection
3.1 Selection of sampling inspection type
When the product quality is stable and there is recent quality management or sampling inspection data that can predetermine the batch standard deviation, the "+" method can be used. When there is no recent data or even if there is recent data, but the quality is unstable, the "+" method should be used. The inspection method for whether the product quality is stable can be carried out in accordance with Appendix A of GB8054-87 "Measuring standard type of average value: batch sampling inspection procedures and tables".
When the manufacturer and the user have a long supply period, whether the "+" method or "+" method is used, the sample mean and sample standard deviation must be recorded in the form of a control chart. If in the application of the "+" method, the control chart shows that the sample standard deviation is in a state of statistical control, it is allowed to convert from the "+" method to the "+" method. If in the application of the "+" method, the control chart shows that the sample standard deviation is in a state of statistical control, it is allowed to convert from the "+" method to the "+" method. During the " method", if the control chart shows that the sample standard deviation has lost the statistical control state, it is necessary to immediately switch from the "" method to the "s" method. If the selected chart does not show that the statistical control state has been lost, but it seems that the batch standard deviation has become smaller or larger, the batch standard deviation used should be updated at any time.
The use of control charts shall be in accordance with GB4091.1-83 "General Principles for Conventional Control Charts" and B4091.2--83 "Mean Standard Deviation Control Chart".
3. 2 Determination of Sampling Inspection Methods
This standard has three sampling inspection methods: 1 specification limit, lower specification limit and bilateral specification limit. Those who adopt this standard can choose according to the different specification limits required by the product standard quality.
When using the effective specification limit, the following two conditions must be met before the diagram of this standard can be applied:
EL> 2.8921-e — 0.89 - -
>2u-2p
GB8053—B7
Where 2i--, and ui-2. is the quantile of the standard normal distribution, with the upper probability p2. and 0.2 p. Appendix B gives the quantile values of the priority non-conforming product rate series used in this standard. When the sampling inspection is the "" method, the values in the above conditions shall be agreed upon by the manufacturer and the user based on past experience. 3.3 Provisions on qualified quality and limit quality
Qualified quality and limit quality shall be determined by negotiation between the manufacturer and the user based on the quality requirements in the product standard. 3.4 Determination of sampling end
3.4.1. Method: Determine the sampling plan according to the steps in the table. Specify the quality requirements and determine the sampling plan. 3.4.2. Method: The upper specification limit is determined by the P:P value table. The value is checked by the inspection method. The lower specification limit is changed to the specification limit by the recent 10 to 20 groups (or) control network data of production, or the recent 10 to 20 batches of throwing sample inspection data. Estimate according to the method in Appendix B of "Metric Standard Type Sampling Inspection Procedure and Table of Average Values of B8051-87"
Determine the sampling plan according to the steps listed in the table
Check method
Work content
Upper specification limit
Specify quality base requirements
Confirm the sampling plan
3.5 Batch composition and sample extraction
According to, the value is in Table 2 Detection
Lower specification limit
Both-side specification limit
Unit products must be submitted in batches. The submitted batches may be the same or different from the production batches, sales batches, and transportation batches, but they should be composed of unit products of the same specification model, the same quality grade, and produced from the same material raw materials under the same process conditions. The batch size shall be agreed upon by the manufacturer and the user according to the sales situation and actual production cases. The required samples shall be randomly selected from the entire batch, which can be carried out after the batch is formed or during the batch formation process. 3.6 Inspection and calculation of samples
The samples are inspected one by one according to the test, measurement or other methods specified in the product standards or order contracts and other relevant documents. The inspection results should be recorded completely and accurately. And the mean and standard deviation of the net sample are calculated. GE 8053-87
If there are individual abnormal data that are obviously different from other inspection results of the samples, the technical or physical reasons for the abnormal data should be found first. If the cause cannot be found, with the consent of the user, it can be judged according to GB4883-85 "Statistical processing and interpretation of data, judgment and processing of abnormal values of normal samples". Then, according to the nature of the abnormal data, the manufacturer and the user shall negotiate and determine whether to remove it.
After the abnormal data are removed, the corresponding number of unit products should be randomly selected from the inspection batch again, supplemented to the sample size required by the sampling plan, and then judged whether the inspection batch is accepted or not. 3.7 Judgment on whether a batch can be accepted
3. 7. 1“\ Judgment rule
8. When the upper specification limit is given,
if ≥batch acceptance, <batch rejection.
b. When the lower specification limit is given,
if ≥batch acceptance, batch rejection.
C. When the double-side specification limit is given,
if both batch acceptance or batch rejection.
3. 7. 2“. \ Judgment rule
Specific. When the upper specification limit is given:
if batch acceptance: batch rejection.
When the lower specification limit is given,
if Q is batch acceptance: <batch rejection.
When the double-side specification limit is given,
if both batch acceptance < or
3.8 Batch processing
batch rejection.
All batches judged as accepted. The user shall accept the whole batch, and the batches judged as rejected. The producer shall not submit them for inspection again without any treatment, and shall handle them in accordance with the provisions of the pre-signed contract. 4 Examples of determining sampling plans
4.1\"Method
4.1.1 Given specification limits
Example: Suppose that a certain product is qualified when the unit product quality characteristic value does not exceed 200. Given α=6, specification=1.00, P8.00, determine the sampling plan that meets the above requirements. Determination steps:
a. Given t =200, p =1. 00. 7. =8. 00, g—6b. According to the number in the intersection position of the horizontal row of P1.00 and the vertical column of P28.00 found in Table 1: 1
GB8053-87
n -10, + -1. 81
c. The sampling plan is [10, 1.81]. Take 10 units of products from the batch, check and get the mean value of the selected product. The judgment rule is 200-
Qu=1.81 batch acceptance: .<1.81 batch rejection. 4.1.2 Given the following specification limit
Example: It is known that the standard deviation of the product characteristic value = 1. The product standard stipulates that the unit product characteristic value is not less than 500 when it is a qualified product. It is also stipulated that the purchase price is 1.00, P=10.00. Find the sampling plan. Determine the steps:
Given L - 500, P -- 1. 00, 10. 00, =1 temperature
According to the known conditions, the number of the intersection of the horizontal row of P, = 1.00 and the vertical column of P10.00 in Table 1 is: b.
# --- 8,x.1.74
c. The sampling plan is (8, 1.747. 8 units of products are selected from the batch. After inspection, the sample average is obtained. The judgment rule is:
If Q12: 1.71 batches are accepted and 01.71 batches are rejected. 4.1.3 Given bilateral specification limits
41—years—500
Example: Suppose a product has a quality characteristic value that requires bilateral specification limits, which are =58.!=67, and the defective rate that exceeds the bilateral specification limits is required to be =5.00, P:=16.00, and the standard deviation of the product is -1.3. Try to find the sampling plan. Determine the steps
, given L = 58, U = 67: = 5. 00: P. = 16. 00,\ -1. 3b. Check whether the above conditions are suitable for the application of the chart of this standard. Calculate
67 - 58 - 6.923
E-L_in
2k(m-, = 2 × 2.32635 - 4,6352.895,. — 0.89k,, = 2.89 X 1.64485 0.89 × 0.99 = 3.8696.923 >3.869 and 6.923 4.653 meet the conditions of the double-sided specifications in Article 3.2, and the chart of this standard can be used to retrieve the oil sample plan. The root saw has met the conditions. From Table 1, the number of intersections between the horizontal row of P=5.00 and the close column of 2.=18.00 is: n — 20, h = 1. 28
d. The sampling plan is [20, 1.28]. 20 units of products are taken from the batch. The sample mean is obtained after inspection. The judgment rule is:
0. 67 -2
If Q1, 28 and Q2, 1.28. The batch is accepted, 0<1.28 or 0<1.28 The batch is rejected. 4.2 "5" method
4.2.1 When the upper specification limit is given
Example: Assuming that the standard deviation in Example 4.1.1 is unknown, sincerely find the sampling plan and determine the steps
Given = 200, P, = 100, 2 = 8.00
b. According to the known conditions, the number of intersections of the horizontal row of p.=1.00 and the vertical column of P.-8.00 in Table 2 is; GA 8053-87
n =28. f =1.83
e. The obtained sampling plan is (28, 1.833 (the sample size is obviously larger than the sampling plan of the method. 28 units of products are selected from the batch, and the sample mean and sample standard steam are obtained after inspection. The judgment rule is: Q
If Q.≥1.83, the batch is accepted, and Q1.83, the batch is rejected. 4.2.2 When the lower specification limit is given
200 --r
Example: Assuming that the difference in the bid in Example 4.1.2 is unknown, try to find its sampling plan. Determination steps:
Given L=500, P1.00, P10.00
According to the known conditions, it can be found from Table 2. The number of intersections between the horizontal row of =1.00 and the vertical column of =10. 00 is; = 21, ± = 1- 76
The sampling plan is [21, 1.76] (the sample size is obviously larger than the sampling plan of the "method": 21 units of products are drawn from the batch, and the sample mean is obtained after inspection: and the sample standard is:. The judgment rule is; Q, = 4-500
If 1.76 batches are accepted: Q1.76 batches are rejected.
4. 2.3 Given the double gauge potential limit time
Example: Assume that the standard deviation in Example 1.1, 3 is unknown, try to find its sampling plan: Determine the steps:
a Given L=58, =67, P=5.00, P=16.00b.In order to check whether the above conditions are suitable, this standard chart is applied. Based on past experience, the standard deviation of this product varies between 0.6 and 1.8, so the maximum standard deviation is used for calculation:
E-F_67-58 - 5.000
2u -0.ap, = 2 x 2.32635 - 4.6352.89w. -r— 0.89# -, = 2.89 X 1.64485 — 0.89 × 0.991463.869
5.D00 = -3.869 and 5.000 4.653 meets the conditions of the two-sided specification limit in Article 3.2. The table retrieval sampling plan of this standard can be used. According to the known tea pieces, the number of the intersection of the horizontal row of product = 5.00 and the vertical column of product = 16.0 in Table 2 is: # 38. =1.29
The sampling plan is [38, 1.29. 38 units of products are taken from the towel. After inspection, the sample value r and the sample standard d are obtained. The judgment rule for the difference s is: If Q1.29 and Q121.29, 5 sampling table and 0c curve, 0 - 67 -, 5—58, batch acceptance u≤1. 29 or 4 1. 29, batch rejection. This standard provides the sampling retrieval table of the "π" method and the "π" method, see Table 1 and Table 3, and gives some results. The "π" method and the "π" method have 0℃ curves of 0.1, 0.2, 0.4, 0.5, 0.8, 1.0, 2.0, 4.0, 5.0.8.0, 10.0, etc. See Figures 1 and 2. Low reference is used. The above DC curve diagram is also applicable to various specification limits. If the sampling plan is not within the above range, its OC curve can be drawn in white according to the method in Appendix A.
GB8053-87
C0*6- I1*2
P*7--18
8- ~ h-1
-1*1-- 106*0
f ~ 112\0
012~- 950
99*0 ~[$f'0
09 0 ~ 9 0
15'0-[8'0
BT*- T'0
ori-err
Trusted 0
GB 6053-87
D950082
00\ 1[ ~ 010*6
0'G ~
T'-OT9*s
9* ~018*zWww.bzxZ.net
D z*2 ~ 018*1
08\1 -0*1
WFI-OEI'E
E1\1 ~ F060
6*0 -112'0
D2\~T9s*0
9 5'0 ~Th'0
5FE*0 -[& 0
7'0 ~ zz\0
F2\0 ~ LI\0
t' ~ TT-0
t'o ~rtro
2'o ~uu0
GB 8053—87
Figure=10.00*\ method sampling characteristic curve
P.=10.00“\ method sampling characteristic curve table value (batch defective product rate%) Scheme
GB 8053—87
Figure 1-2—8.00””” method sampling characteristic curve P,=8.00\ method sampling characteristic curve table value (batch defective product rate%) Sampling scheme
GB 8053—87
Figure 1-3—5.00α” method sampling characteristic curve5.00“ method sampling characteristic curve table value (batch defective product rate%)
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.