GB/T 2547-2026 Plastics—Sampling method English, Anglais, Englisch, Inglés, えいご
This is a draft translation for reference among interesting stakeholders. The finalized translation (passing through draft translation, self-check, revision and verification) will be delivered upon being ordered.
ICS
CCS
National Standard of the People's Republic of China
GB/T 2547-2026
Plastics - Sampling method
塑料 取样方法
Issue date: 2026-02-27 Implementation date: 2026-09-01
Issued by the General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China
the Standardization Administration of the People's Republic of China
Contents
Foreword
1 Scope
2 Normative references
3 Terms and definitions
4 Sampling procedure
5 Determination of sample size
6 Extraction of samples
7 Sampling
8 Sampling report
Plastics — Methods of sampling
1 Scope
This document specifies the sampling procedure, determination of sample size, extraction of samples, sampling and sampling report for plastics.
This document is applicable to the sampling of homogeneous solid plastics.
2 Normative references
This document has no normative references.
3 Terms and definitions
No terms and definitions are required to be defined in this document.
4 Sampling procedure
The sampling procedure for plastics is as follows:
a) Define the scope of the overall product, such as a delivery batch agreed upon by the buyer and seller, or a production batch from intermittent production, or production material within a certain interval of continuous production, etc.;
b) Determine the historical data of the overall product, and identify the standard deviation or coefficient of variation of the product;
c) Determine the sample size and sampling quantity;
d) Determine the method of sample extraction (random sampling method or systematic sampling method);
e) Determine the sampling method;
f) Complete the sampling according to the determined extraction method and sampling method;
g) Prepare the sampling report.
5 Determination of sample size
5.1 Determination of sample size n
In order to obtain sample data that reflects the true quality of the overall product, an appropriate number of sampling units (i.e., sample size) shall be taken from the population. When historical data of the product are available, the sample size can be calculated by Formula (1):
Formula (1) can be transformed into Formula (2):
The estimated value of the coefficient of variation of the overall product quality (V0V0) is calculated according to Formula (3):
The maximum permissible relative error (ee) is calculated according to Formula (4):
The value of nn shall be determined by rounding up to the nearest integer.
When there is no high requirement for sampling accuracy, V0V0 and ee may be taken as empirical values of 0.1, resulting in Formula (5):
The sample size shall also be adjusted according to the actual product quality. When sufficient data are available and the data are stable, a smaller sample size is permitted. When historical data are limited and high accuracy is required, a larger sample size than that required by Formula (1) and Formula (2) is usually needed. For data with greater variability, such as Charpy impact strength and nominal tensile strain at break, an increase in sample size should be considered to obtain accurate sampling data.
5.2 Determination of σ0σ0 or V0V0
5.2.1 Based on the historical data of the same type of product, the sample standard deviation or coefficient of variation for several batches with equal or similar sample sizes shall be calculated using Formula (6) and Formula (7), respectively.
The standard deviation of the mean shall be calculated according to Formula (8) and Formula (9), with the standard deviation of the mean of the sample standard deviations taken as σ0σ0, and the standard deviation of the mean of the sample coefficients of variation taken as V0V0.
In determining σ0σ0 or V0V0, generally the larger the sample size n′n′ and the larger the number of batches ll, the more accurate the result. However, in practice, if n′n′ is large, the number of batches ll may be smaller; if n′n′ is small, the number of batches ll should be larger. For example, when n′n′ is greater than 20, ll may be 4 to 5. When n′n′ is less than 10, it is preferable that ll be greater than 10.
5.2.2 If no historical data are available, data accumulation shall be carried out in accordance with 5.2.1 to estimate a suitable σ0σ0 or V0V0.
5.3 Determination of maximum permissible error EE or ee
In determining the maximum permissible error EE or ee, an appropriate balance shall be struck between the required accuracy of the estimate and the cost of obtaining such an accurate estimate. When a small change in a quality characteristic affects the product, EE or ee should be set smaller; if there is no significant effect, it is unnecessary to set EE or ee too small, which would incur excessive costs.
5.4 Determination of probability coefficient AA
The probability coefficient may be determined according to the required confidence interval of the results. In industrial production, AA is generally set at 1.96, meaning that the probability that the error between the estimated mean of the overall product quality obtained from the sample and the true overall mean exceeds the maximum permissible error EE or ee is 5 %. Values of AA corresponding to other probabilities may be obtained from Table 1 as required.
5.5 Determination of sample size nn for products with multiple quality characteristics
For plastic products, which usually have several quality characteristics, the required nn for each quality characteristic may be calculated separately, and the maximum value shall be taken as the sample size for the inspection lot. Alternatively, the nn value may be calculated using the maximum coefficient of variation among the key quality characteristics related to the primary use of the product.
Example: A production unit plans to estimate the mean of the melt mass-flow rate of a product population. It is known that each batch has 30 historical data points. Five batches of historical data are selected, and the standard deviations of the melt mass-flow rate are 0.037 g/10 min, 0.016 g/10 min, 0.033 g/10 min, 0.039 g/10 min and 0.024 g/10 min. When the maximum permissible error (EE) is 0.1 g/10 min, calculate the sample size for sampling as follows:
GB/T 2547-2026 Plastics—Sampling method English, Anglais, Englisch, Inglés, えいご
This is a draft translation for reference among interesting stakeholders. The finalized translation (passing through draft translation, self-check, revision and verification) will be delivered upon being ordered.
ICS
CCS
National Standard of the People's Republic of China
GB/T 2547-2026
Plastics - Sampling method
塑料 取样方法
Issue date: 2026-02-27 Implementation date: 2026-09-01
Issued by the General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China
the Standardization Administration of the People's Republic of China
Contents
Foreword
1 Scope
2 Normative references
3 Terms and definitions
4 Sampling procedure
5 Determination of sample size
6 Extraction of samples
7 Sampling
8 Sampling report
Plastics — Methods of sampling
1 Scope
This document specifies the sampling procedure, determination of sample size, extraction of samples, sampling and sampling report for plastics.
This document is applicable to the sampling of homogeneous solid plastics.
2 Normative references
This document has no normative references.
3 Terms and definitions
No terms and definitions are required to be defined in this document.
4 Sampling procedure
The sampling procedure for plastics is as follows:
a) Define the scope of the overall product, such as a delivery batch agreed upon by the buyer and seller, or a production batch from intermittent production, or production material within a certain interval of continuous production, etc.;
b) Determine the historical data of the overall product, and identify the standard deviation or coefficient of variation of the product;
c) Determine the sample size and sampling quantity;
d) Determine the method of sample extraction (random sampling method or systematic sampling method);
e) Determine the sampling method;
f) Complete the sampling according to the determined extraction method and sampling method;
g) Prepare the sampling report.
5 Determination of sample size
5.1 Determination of sample size n
In order to obtain sample data that reflects the true quality of the overall product, an appropriate number of sampling units (i.e., sample size) shall be taken from the population. When historical data of the product are available, the sample size can be calculated by Formula (1):
Formula (1) can be transformed into Formula (2):
The estimated value of the coefficient of variation of the overall product quality (V0V0) is calculated according to Formula (3):
The maximum permissible relative error (ee) is calculated according to Formula (4):
The value of nn shall be determined by rounding up to the nearest integer.
When there is no high requirement for sampling accuracy, V0V0 and ee may be taken as empirical values of 0.1, resulting in Formula (5):
The sample size shall also be adjusted according to the actual product quality. When sufficient data are available and the data are stable, a smaller sample size is permitted. When historical data are limited and high accuracy is required, a larger sample size than that required by Formula (1) and Formula (2) is usually needed. For data with greater variability, such as Charpy impact strength and nominal tensile strain at break, an increase in sample size should be considered to obtain accurate sampling data.
5.2 Determination of σ0σ0 or V0V0
5.2.1 Based on the historical data of the same type of product, the sample standard deviation or coefficient of variation for several batches with equal or similar sample sizes shall be calculated using Formula (6) and Formula (7), respectively.
The standard deviation of the mean shall be calculated according to Formula (8) and Formula (9), with the standard deviation of the mean of the sample standard deviations taken as σ0σ0, and the standard deviation of the mean of the sample coefficients of variation taken as V0V0.
In determining σ0σ0 or V0V0, generally the larger the sample size n′n′ and the larger the number of batches ll, the more accurate the result. However, in practice, if n′n′ is large, the number of batches ll may be smaller; if n′n′ is small, the number of batches ll should be larger. For example, when n′n′ is greater than 20, ll may be 4 to 5. When n′n′ is less than 10, it is preferable that ll be greater than 10.
5.2.2 If no historical data are available, data accumulation shall be carried out in accordance with 5.2.1 to estimate a suitable σ0σ0 or V0V0.
5.3 Determination of maximum permissible error EE or ee
In determining the maximum permissible error EE or ee, an appropriate balance shall be struck between the required accuracy of the estimate and the cost of obtaining such an accurate estimate. When a small change in a quality characteristic affects the product, EE or ee should be set smaller; if there is no significant effect, it is unnecessary to set EE or ee too small, which would incur excessive costs.
5.4 Determination of probability coefficient AA
The probability coefficient may be determined according to the required confidence interval of the results. In industrial production, AA is generally set at 1.96, meaning that the probability that the error between the estimated mean of the overall product quality obtained from the sample and the true overall mean exceeds the maximum permissible error EE or ee is 5 %. Values of AA corresponding to other probabilities may be obtained from Table 1 as required.
5.5 Determination of sample size nn for products with multiple quality characteristics
For plastic products, which usually have several quality characteristics, the required nn for each quality characteristic may be calculated separately, and the maximum value shall be taken as the sample size for the inspection lot. Alternatively, the nn value may be calculated using the maximum coefficient of variation among the key quality characteristics related to the primary use of the product.
Example: A production unit plans to estimate the mean of the melt mass-flow rate of a product population. It is known that each batch has 30 historical data points. Five batches of historical data are selected, and the standard deviations of the melt mass-flow rate are 0.037 g/10 min, 0.016 g/10 min, 0.033 g/10 min, 0.039 g/10 min and 0.024 g/10 min. When the maximum permissible error (EE) is 0.1 g/10 min, calculate the sample size for sampling as follows:
