C-02—Minimum Sample Size for Prepackaged Commodity Inspection

C-02—Minimum Sample Size for Prepackaged Commodity Inspection (PDF, 11 Ko, 2 pages)

Date: 1991-09-30
Bulletin: C–02–E
Category: Commodity
Document(s): W&M Regs, section 52; Sch. III, Part I

1. Prepackaged commodities grouped in lots are normally sampled when inspection for compliance with the Weights and Measures Regulations is performed. The minimum number of packages that must be sampled is indicated in Schedule III, Part I of the same Regulations.

2. A lot cannot be accepted if less than the minimum number of packages has been tested. However, a lot can be rejected when the number of packages that has been tested is less than the minimum, if the number of defective packages exceeds the maximum number permitted under subsections 52(4)(b) and 52(4)(c) of theRegulations.

3. If the minimum number of packages that must be sampled in order to accept a given lot is 32, such a lot could be rejected after testing only two units if their net content was found to be below the declared quantity by more than two times the limit of error. The lot would then fail the T2 criterion defined in subsection 52(4)(c) of the Regulations.

4. In practice, however, the following guidelines should be applied by Legal Metrology inspectors when determining the number of packages that are to be sampled and tested:

4.1 When the inspection may lead to a prosecution, the number of packages which are sampled should always be equal to or greater than the number specified in Schedule III, Part I of the Regulations (e.g. 32 units for a lot of 2,000).

4.2 When the inspection is not likely to lead to a prosecution, a lot should not be seized and detained if less than ten units have been tested. Therefore, even if after testing a few units there is no doubt that the lot fails, additional units should be selected and tested until a total of ten units have been sampled. A sample of ten should suffice to provide the inspector with a general idea of the characteristics of the lot.

5. Whenever a lot is rejected, the sample mean and lot weighted average should be calculated to determine compliance with the average criterion.

6. For additional information, please contact:

Gilles P. Vinet
Vice-President
Program Development Directorate