Are we required to use a statistical sampling system during process verification and for determining conformance of product when manufacturing lots/batches?
(link)
Subpart O--Statistical Techniques
(a) Where appropriate, each manufacturer shall establish and maintain procedures for identifying valid statistical techniques required for establishing, controlling, and verifying the acceptability of process capability and product characteristics.
(b) Sampling plans, when used, shall be written and based on a valid statistical rationale. Each manufacturer shall establish and maintain procedures to ensure that sampling methods are adequate for their intended use and to ensure that when changes occur the sampling plans are reviewed. These activities shall be documented.
(link)
O. Statistical Techniques (Subpart O)
204. Several comments on Sec. 820.250(b) stated that the provision as written seems to require the use of sampling plans, and that every manufacturer does not necessarily use sampling plans. Another comment stated that sampling plans are not often used during reviews of nonconformities, quality audits, or complaints, and that these examples should, therefore, be deleted. Two other comments questioned the meaning of ``regularly reviewed.''
FDA's intent was not to require the use of sampling plans, but to require that where they are used, they should be written and valid. Section 820.250 was revised to make that clear.
Sampling plans are not always required, but any time sampling plans are used [crl: as opposed to 100% inspection], they must be based on a valid statistical rationale.
Further, FDA acknowledges that the most common use of sampling plans is during receiving acceptance, and has deleted the examples. FDA has also clarified the review requirement by stating ``to ensure that when changes occur the sampling plans are reviewed.''
(link)
A review should be made to determine if all sampling plans for inspection and rework are based on an acceptable statistical rationale and technique. (Mil. Std. ANSI/ASQC, etc.)
Acceptance sampling is concerned with making a decision with regard to accepting or not accepting a "lot" based on the result of a sample(s) selected from that lot. A wide range of acceptance sampling plans are available to satisfy specific requirements and applications.
[crl: See appropriate standards, such as ANSI Z1.4 or equivalent.]
When constructing a sampling plan, close attention should be paid to decisions regarding sample size, sampling frequency, sample selection, the basis of sub-grouping, and various other aspects of sampling methodology. [crl: See appropriate standard(s) for statistical sampling systems.]
Sampling requires that the sample be chosen in an unbiased fashion.
[crl: Requires that a random sample can be drawn from the process or lot, ensuring that each unit has the same probability of being selected.]
(may also be found in ANSI Z1.4)
The items selected for the sample shall be drawn from the lot by simple random sampling (see 2.1.5 in ISO 3534-2:1993). However, when the lot consists of sub-lots or strata, identified by some rational criterion, stratified sampling shall be used in such a way that the size of the subsample from each sublot or stratum is proportional to the size of that sublot or stratum.
8.1.2 The use of statistical methods can be beneficial to the organization in a wide range of circumstances, including data collection, analysis and application. These techniques are useful for demonstrating process capability, as well as product conformity to specified requirements. They assist in deciding what data to obtain, and in making the best use of the data to gain a better understanding of customer requirements and expectations.
Statistical methods can also find uses in
8.1.3 Among the statistical methods which can be beneficial for these purposes are the following:
8.1.4 Once the appropriate statistical techniques are chosen, it is important to implement those techniques in such a manner that appropriate data are collected and evaluated, and the results are reported to the relevant departmental functions, so that necessary actions can be taken. The data resulting from the application of statistical techniques can be an effective means of demonstrating conformity to requirements for quality and can be used as quality records. The documentation of such techniques and the records resulting from them might be subject to regulatory requirements.