S-S-04—Sampling plans for the inspection of isolated lots and short series of lots

S-02—Application of specifications for statistical sampling and measurement uncertainty in the verification and reverification of electricity and gas meters


Category: Statistical methods
Specification: S-S-04 (rev. 2)
Document(s): S-S-01, S-S-02, S-S-03, S-02
Issue date: 2009-10-28
Effective date: 2009-10-28
Supersedes: S-S-04 (rev. 1)


Table of contents


1.0 Scope

This specification establishes requirements for sampling inspection of isolated lots and short series of lots of homogeneous measurement devices.

2.0 Authority

This specification is issued under the authority of section 19 of the Electricity and Gas Inspection Regulations.

3.0 Normative references

3.1 ISO 2859-2:1985, Sampling procedures for inspection by attributes - Part 2: Sampling plans indexed by limiting quality (LQ) for isolated lot inspection.

3.2 S-S-02, Measurement Uncertainty and Meter Conformity Evaluation Specifications.

3.3 S-S-03, Prerequisites to the Use of Sampling Inspection.

3.4 S-S-01, Specifications for Random Sampling and Randomization.

3.5 Relevant specification for measurement device verification and reverification.

4.0 Sampling inspection and conformity test requirements

4.1 The requirements of ISO 2859-2 (reference 3.1), as modified by the clarifications and additions described in Annex A, and as extended by the requirements in Annex B for a short series of lots, shall apply, provided that the requirements of S-S-03 (reference 3.3) are satisfied in advance.

4.2 A lot submitted for sampling inspection shall consist of devices of the same make and model, with the same metrological parameters, manufactured or reworked under uniform conditions and within a time frame as specified in the device's specification (reference 3.5). Any additional requirements for lot homogeneity shall be specified in the device's specification.

4.3 A conforming unit for the purpose of the sampling plans shall be one for which the conformity tests have demonstrated that the values of all specified performance quality characteristics are in conformity with their specification requirements in accordance with the requirements of S-S-02 (reference 3.2), and all specified non-performance quality characteristics are in conformity with their device specification requirements (reference 3.5). Any unit failing to meet these criteria shall be classified as a nonconforming unit.

4.4 Quality records shall include the auditing parameters of the random sampling process and all inspection results and calculations, corrected as applicable, used to determine lot acceptability.

5.0 Revision

The purpose of revision 2 is to:

  • amend section 3.0 to update the title of S-S-01.
  • renumber section 5.2 as 4.4.
  • delete sections 5.0 and 5.1, as requirements for accredited meter verifiers to develop and implement policies and procedures are stipulated for accredited meter verifiers in S-A-01, "Criteria for the Accreditation of Organizations to Perform Inspections Pursuant to the Electricity and Gas Inspection Act and the Weights and Measures Act".
  • clarify in section A.2, the classification precedence of a nonconforming unit; also clarify reference to the standard in the definition of a measurement device.
  • add in sections A.3(1), A.4(2) and A.4(3), provision for non-zero acceptance number for nonperformance characteristics.
  • amend note under section A.4(3), to address process requalification recommended practices.
  • add in section A.4(5), requirements covering nonconformity or defects which may have adverse impact on quality. For guidance a note added under this section.
  • amend notes under Tables A.1 and B.1, to indicate Ac = 0 applies to performance characteristics.
  • add new tables A.3 and B.3.
  • add in sections B.4 and B.5.1, reference to table B.3.
  • add in sections B.5.2 and B.5.3, provision for non-zero acceptance number for non-performance characteristics.
  • amended note under section B.5.8, to refer to note under section A.4.(3).

The purpose of revision 1 was:

  • to add "reference 3.1" to sections A.4(1), (2), (3).
  • in section A.4(3)(b), to amend the annex B reference.
  • to renumbered annex B.5.
  • to clarify sample size application in sections B.5.2 and B.5.3.
  • to clarify application of requirements in sections B.5.4, B.5.5 and B.5.6 (and to delete redundant note under section B.5.5).

Annex A (normative) Modifications to ISO 2859-2:1985

This annex lists only the modifications or amendments to the referenced sections of ISO 2859-2:1985. All remaining requirements apply as written in the standard.

A.1 Scope and field of application

Procedure A (for isolated lot inspection) shall be used.

A.2 Terminology

Nonconforming Unit:

A device is a nonconforming unit if any performance or non-performance specification is not met as defined in the device specification (reference 3.5) issued pursuant to section 18 of the Electricity and Gas Inspection Regulations. Classification as a nonconforming unit takes precedence over classification as marginally conforming unit.

Marginally Conforming Compressed Specification Limits (CSL):

The type 1 compressed specification limit (CSL) values determined from the device's performance specification limits. The specification limits and CSL values are defined in the device's specification (reference 3.5).

Marginally Conforming Unit:

A device which has no nonconformity but exhibits performance falling outside the interval defined by the lower and upper CSL values (type 1 marginally conforming) or has an MADT value exceeding an MADT limit (type 2 marginally conforming), where classification as type 1 takes precedence over classification as type 2.

Measure of Absolute Deviation from Target (MADT):

A measure of an individual device's conformity to target performance over its designed range of operation, calculated as defined in the device's specification (reference 3.5).

Measurement Device:

For the purposes of the standard, a device, or measurement device is an item and considered a meter.

Responsible Authority:

For the purposes of the standard, Measurement Canada.

A.3 Choice of sampling plan

  1. Subject to item (2) below, sampling plans based on a limiting quality (LQ) index of 2.0% or 3.15% shall be used for the inspection of all quality characteristics.
  2. Subject to the requirements of S-S-03, (reference 3.3), a limiting quality (LQ) index of 8.0% may be used for the inspection of performance characteristics.

A.4 Rules for acceptance and non-acceptance

  1. For the purposes of 4.1 of the standard (reference 3.1), the sample shall be drawn at random without replacement from the lot, using authorized random sampling software developed in accordance with S-S-01 (reference 3.4). The complete list of sample units shall be identified prior to inspection.

  2. For the purposes of 4.2 of the standard (reference 3.1), and subject to A.3(2) above, the decision regarding lot acceptance shall be based on the numbers of marginally conforming and nonconforming units in the sample. A device is considered to be a nonconforming unit if any specification requirement is not fulfilled. Tables A.1 and A.2 define the acceptance criteria for marginally conforming units with respect to the CSL values and the MADT value respectively; subject to A.4(5) below, Table A.3 defines the acceptance criteria for nonperformance nonconforming units.

  3. For the purposes of 4.4 of the standard (reference 3.1), the acceptance numbers (Ac1 for type 1 marginally conforming units, Ac2 for type 2 marginally conforming units and Ac3 for nonperformance nonconforming units) specified for a sampling plan shall apply only to marginally conforming units and nonperformance units; the acceptance number for performance nonconforming units is always 0. In addition under this sub-clause:

    1. an unacceptable lot may be resubmitted for sampling inspection provided that the specified action in item b) of the standard has taken place; re-inspection shall include evaluation of all quality characteristics where the non-acceptance is due to performance characteristics or, for all other types of nonconformities, evaluation of the characteristic(s) causing lot non-acceptance. For inspection occurring under this subclause, the acceptance number for nonconforming units is always 0.
    2. where presentation of additional lots are planned then corrective action, including requalification of the production process where necessary, must have been implemented before production of the next lot. In addition the provisions of Annex B.5.8 below apply.

    Note: Refer to Bulletin S-01: Implementation Guide for the Application of Statistical Sampling Requirements for guidance on the above sections. Organizations should provide appropriate procedures to carry out the corrective action process; including response to events that could lead to cessation of sampling and consequent requalification of the process.

  4. Where the quantities of marginally conforming units in the lot exceed the products of the associated limiting quality (LQ) values and the lot size, recalibrate, dispose of, or replace marginally conforming devices, so that the outgoing quantities of such devices do not exceed the LQ value.

  5. Whenever any nonconformity discovered during sampling inspection is due to noncompliance with the device's notice of pattern approval, or the nonconformity or defect compromises, or may potentially compromise measurement accuracy, meter integrity, or appropriate usage of the meter and it is determined that the cause of the nonconformity or defect may potentially impact additional meters of the same (or similar), makes, types, or models then the lot shall be rejected and sampling inspection shall be suspended pending identification and correction of the cause. The responsible authority shall be promptly notified of such an occurrence and the corrective action proposed before sampling inspection continues.

Note: Regarding nonconformity or defect, the above clause does not apply when the cause of the defect is determined to have occurred as a result of an isolated incident such as mishandling, extreme environmental occurrences, or catastrophic damage. Furthermore, the above clause does not apply to nonconformities which are the result of calibration errors not caused by a malfunction or defect, or the result of errors due to the programming of incorrect data in user-programmable parameters or calibration.

A.5 Double and multiple sampling plans

Double or multiple sampling may be optionally used where the acceptance number for a given single sampling plan exceeds zero. Details of these sampling plans are provided in the tables below.

Table A.1—Compatible single, double, and multiple sampling plans for type 1 marginally conforming units
Lot Size
N
Option 1: LQ = 3.15% Option 2: LQ = 8.0%
Single Double Multiple Single Double Multiple
n, Ac1, Re1 n, Ac1, Re1 n, Ac1, Re1 n, Ac1, Re1 n, Ac1, Re1 n, Ac1, Re1
Up to 25 use 100% inspection use 100% inspection use 100% inspection 17, 0, 1 use Single use Single
26 to 50 use 100% inspection use 100% inspection use 100% inspection 22, 0, 1 use Single use Single
51 to 90 Footnote 1 44, 0, 1 use Single use Single 24, 0, 1 use Single use Single
91 to 150 55, 0, 1 use Single use Single 26, 0, 1 use Single use Single
151 to 280 65, 0, 1 use Single use Single 28, 0, 1 use Single use Single
281 to 500 80, 0, 1 use Single use Single 32, 0, 1 use Single use Single
501 to 1200 125, 1, 2
  • 80, 0, 2
  • 80, 1, 2
use Double 50, 1, 2
  • 32, 0, 2
  • 32, 1, 2
use Double
1201 to 3200 125, 1, 2
  • 80, 0, 2
  • 80, 1, 2
use Double 80, 3, 4
  • 50, 1, 4
  • 50, 4, 5
  • 20, no Ac1, 3
  • 20, 0, 3
  • 20, 1, 4
  • 20, 2, 5
  • 20, 4, 5
3201 to 10000 200, 3, 4
  • 125, 1, 4
  • 125, 4, 5
  • 50, no Ac1, 3
  • 50, 0, 3
  • 50, 1, 4
  • 50, 2, 5
  • 50, 4, 5
125, 5, 6
  • 80, 2, 5
  • 80, 6, 7
  • 32, no Ac1, 4
  • 32, 1, 5
  • 32, 2, 6
  • 32, 4, 7
  • 32, 6, 7
10001 to 35000 315, 5, 6
  • 200, 2, 5
  • 200, 6, 7
  • 80, no Ac1, 4
  • 80, 1, 5
  • 80, 2, 6
  • 80, 4, 7
  • 80, 6, 7
200, 10, 11
  • 125, 5, 9
  • 125, 12, 13
  • 50, 0, 5
  • 50, 3, 8
  • 50, 6, 10
  • 50, 9, 12
  • 50, 12, 13

Notes:

n

Sample size

Ac1

Acceptance number (maximum permissible type 1 marginally conforming units)

Re1

Rejection number (minimum impermissible type 1 marginally conforming units)

Where the sample size equals or exceeds the lot size, perform 100% inspection of the lot.

Ac = 0 for all nonconforming units.

Footnotes

Footnote 1

Refer to the Guideline document concerning sample size recommendation.

Return to footnote 1 referrer

Table A.2—Compatible single, double, and multiple sampling plans for type 2 marginally conforming units
Lot Size
N
Option 1: LQ = 20% Option 2: LQ = 20%
Single Double Multiple Single Double Multiple
n, Ac2, Re2 n, Ac2, Re2 n, Ac2, Re2 n, Ac2, Re2 n, Ac2, Re2 n, Ac2, Re2
Up to 25 use 100% inspection use 100% inspection use 100% inspection 17, 1, 2 use Single use Single
26 to 50 use 100% inspection use 100% inspection use 100% inspection 22, 2, 3 use Single use Single
51 to 90 44, 6, 7 use Single use Single 24, 2, 3 use Single use Single
91 to 150 55, 7, 8 use Single use Single 26, 2, 3 use Single use Single
151 to 280 65, 9, 10 use Single use Single 28, 2, 3 use Single use Single
281 to 500 80, 11, 12 use Single use Single 32, 3, 4 use Single use Single
501 to 1200 125, 19, 20
  • 80, 11, 18
  • 80, 22, 23
use Double 50, 6, 7
  • 32, 3, 6
  • 32, 7, 8
use Double
1201 to 3200 125, 19, 20
  • 80, 11, 18
  • 80, 22, 23
use Double 80, 11, 12
  • 50, 5, 9
  • 50, 14, 15
  • 20, 1, 6
  • 20, 3, 8
  • 20, 6, 10
  • 20, 10, 13
  • 20, 13, 14
3201 to 10000 200, 32, 33
  • 125, 17, 23
  • 125, 40, 41
  • 50, 5, 15
  • 50, 10, 26
  • 50, 20, 32
  • 50, 31, 36
  • 50, 37, 38
125, 19, 20
  • 80, 11, 18
  • 80, 22, 23
  • 32, 2, 9
  • 32, 7, 14
  • 32, 11, 19
  • 32, 18, 21
  • 32, 22, 23
10001 to 35000 315, 53, 54
  • 200, 28, 37
  • 200, 69, 70
  • 80, 9, 23
  • 80, 17, 41
  • 80, 34, 55
  • 80, 49, 66
  • 80, 67, 68
200, 32, 33
  • 125, 17, 23
  • 125, 40, 41
  • 50, 5, 15
  • 50, 10, 26
  • 50, 20, 32
  • 50, 31, 36
  • 50, 37, 38

Notes:

n

Sample size

Ac2

Acceptance number (maximum permissible type 2 marginally conforming units)

Re2

Rejection number (minimum impermissible type 2 marginally conforming units)

Where the sample size equals or exceeds the lot size, perform 100% inspection of the lot.

Table A.3—Compatible single, double, and multiple sampling plans for nonperformance nonconforming units
Lot Size
N
LQ = 2.0%
Single Double Multiple
n, Ac3, Re3 n, Ac3, Re3 n, Ac3, Re3
Up to 25 use 100% inspection use 100% inspection use 100% inspection
26 to 50 use 100% inspection use 100% inspection use 100% inspection
51 to 90 Footnote 2 44, 0, 1 use Single use Single
91 to 150 55, 0, 1 use Single use Single
151 to 280 65, 0, 1 use Single use Single
281 to 500 80, 0, 1 use Single use Single
501 to 1200 125, 1, 2
  • 80, 0, 2
  • 80, 1, 2
use Double
1201 to 3200 Footnote 2 125, 1, 2
  • 80, 0, 2
  • 80, 1, 2
use Double
3201 to 10000 200, 1, 2
  • 125, 0, 2
  • 125, 1, 2
use Double
10001 to 35000 315, 3, 4
  • 200, 1, 4
  • 200, 4, 5
  • 80, no Ac3, 3
  • 80, 0, 3
  • 80, 1, 4
  • 80, 2, 5
  • 80, 4, 5

Notes:

n

Sample size

Ac3

Acceptance number (maximum permissible nonperformance nonconforming units)

Re3

Rejection number (minimum impermissible nonperformance nonconforming units)

Where the sample size equals or exceeds the lot size, perform 100% inspection of the lot.

Ac3 = 0 for nonperformance characteristics identified in sections A.4(3)(a) and A.4(5).

Footnotes

Footnote 2

Refer to the Guideline document concerning sample size recommendation.

Return to footnote 2 referrer

Annex B (normative) Extensions to ISO 2859-2:1985

This annex extends ISO 2859-2:1985 to address sampling inspection of short series of homogeneous lots.

B.1 Scope

This annex specifies sampling by attributes plans for a short series of lots.

B.2 Conditions

The following conditions shall exist before these plans may be used for sampling inspection:

  1. production process quality is stable over the series of lots in the production run;
  2. the lots are submitted to inspection in the order of their production;
  3. the series of lots are homogeneous and no single lot can be generally considered to be different than the rest of the lots in the series;
  4. the lots are relatively similar in size.

B.3 Inspection and choice of sampling plan

The requirements of Annex A section A.3(1) and A.3(2) shall apply.

B.4 Rules for acceptance and non-acceptance

The requirements of Annex A A.4(1), A.4(2), A.4(3), and A.4(5) shall apply, with reference to the sampling plans in Tables B.1, B.2 and B.3.

B.5 Implementation

B.5.1 The user shall select a sampling plan to be used for the series from the option 1 columns of Tables B.1, B.2 and B.3. Optionally, for the inspection of device performance characteristics only, a different sampling plan may be selected and used from the same columns, or from the option 2 columns of the tables provided that annex A section 3(2) has been satisfied.

B.5.2 At the start or re-start of sampling inspection, lot acceptance shall be based on the observed number of marginally conforming units (mCSL for type 1 and mMADT for type 2) and nonconforming units d (dp for performance and dnp for nonperformance) in the sample, using two samples with sizes totaling n1 (where second sample is size n2) and combining those samples for the count of marginally conforming and nonconforming units. The lot shall be accepted if mCSL ≤ Ac1, mMADT ≤ Ac2 , dnp ≤ Ac3 and dp = 0 and non-accepted otherwise.

B.5.3 From the second lot onwards following the start or re-start of sampling inspection, lot acceptance shall be based on the cumulative numbers of marginally conforming units (mCSL for type 1 and mMADT for type 2) and nonconforming units in the current sample (i) using sample size n2 and immediately preceding sample (i − 1) using sample size n2. The lot shall be accepted if mi + m i − 1 ≤ Ac1 (for mCSL), mi + m i − 1 ≤ Ac2 (for mMADT), dnpi + dnp i − 1 ≤ Ac3 (for nonperformance), and dpi + dp i − 1 = 0, and non-accepted otherwise.

B.5.4 If resubmitted for inspection, non-accepted lots shall be presented for inspection under Annex A.4(3): for the case of performance characteristics demonstrating nonconformities or marginal conformities, screening inspection shall be for all of the device's performance characteristics; for the case of non-performance characteristics demonstrating nonconformities, screening inspection shall be for the nonconformities causing non-acceptance.

B.5.5 If, under screening inspection for performance characteristics, the number of marginally conforming units in the lot exceeds the products of the associated limiting quality (LQ) values and the lot size, recalibrate, dispose of, or replace marginally conforming devices, so that the outgoing quantities of such devices do not exceed the LQ value.

B.5.6 Subject to B.5.4 and B.5.7(c), screening inspection shall continue for all of the applicable characteristics of the devices in the next lot in the series.

B.5.7 Sampling inspection may be re-started for a given quality characteristic if the following criteria have been satisfied, as applicable:

  1. no nonconformities have been found in the previously screened lot.
  2. the number of marginal nonconforming units in the previously screened lot do not exceed the products of the associated LQ values and the lot size.
  3. corrective action, including requalification of the production process where necessary, has been implemented before production of the next lot.

B.5.8 In the event that lots are frequently non-accepted under LQ 8.0% then sampling inspection may continue under LQ 3.15%. If that option is not viable, or frequent lot non-acceptance occurs under LQ 3.15%, or other conditions warrant then sampling shall be discontinued and corrective action implemented. The process must be requalified in accordance with S-S-03 (reference 3.3) for sampling to resume under either sampling option.

Note: Refer to S-01, Implementation Guide for the Application of Statistical Sampling Requirements for guidance on the above sections. Refer to note under section A.4(3).

B.5.9 The above rules may be implemented so that they are applied separately to individual device quality characteristics, leading to some characteristics being inspected under sampling inspection whereas others are being 100% inspected.

Table B.1—Compatible sampling plans for type 1 marginally conforming units
Option 1: LQ = 3.15% Option 2: LQ = 8.0%
n1, n2, Ac1, Re1 n1, n2, Ac1, Re1
80, 40, 0, 1 32, 16, 0, 1
125, 63, 1, 2 50, 25, 1, 2
200, 100, 3, 4 80, 40, 3, 4
315, 158, 5, 6 125, 63, 5, 6

Notes:

n

Sample size

Ac1

Acceptance number (maximum permissible type 1 marginally conforming units)

Re1

Rejection number (minimum impermissible type 1 marginally conforming units)

Where the sample size equals or exceeds the lot size, perform 100% inspection of the lot.

Ac = 0 for all performance nonconforming units.

Table B.2—Compatible sampling plans for type 2 marginally conforming units
Option 1: LQ = 20% Option 2: LQ = 20%
n1, n2, Ac2, Re2 n1, n2, Ac2, Re2
80, 40, 11, 12 32, 16, 3, 4
125, 63, 19, 20 50, 25, 6, 7
200, 100, 32, 33 80, 40, 11, 12
315, 158, 53, 54 125, 63, 19, 20

Notes:

n

Sample size

Ac2

Acceptance number (maximum permissible type 2 marginally conforming units)

Re2

Rejection number (minimum impermissible type 2 marginally conforming units)

Where the sample size equals or exceeds the lot size, perform 100% inspection of the lot.

Table B.3—Compatible sampling plans for nonperformance nonconforming units
Option 1 Option 2
n1, n2, Ac3, Re3 n1, n2, Ac3, Re3
80, 40, 0, 1 80, 40, 0, 1
125, 63, 1, 2 Footnote 3 125, 63, 1, 2 Footnote 3
200, 100, 1, 2 200, 100, 1, 2
315, 158, 3, 4 315, 158, 3, 4

Notes:

n

Sample size

Ac3

Acceptance number (maximum permissible nonperformance nonconforming units)

Re3

Rejection number (minimum impermissible nonperformance nonconforming units)

Where the sample size equals or exceeds the lot size, perform 100% inspection of the lot.

Ac3 = 0 for nonperformance characteristics identified in sections A.4(5), B.5.4 and B.5.7(a).

Footnotes

Footnote 3

Refer to the Guideline document concerning sample size recommendation.

Return to footnote 3 referrer