MODUK - DEF STAN 05-58
Sampling Procedures and Tables for Inspection by Attributes of Isolated Lots
| Organization: | MODUK |
| Publication Date: | 1 January 1986 |
| Status: | inactive |
| Page Count: | 146 |
scope:
The plans and procedures given in BS 6001 are designed for accepting or rejecting lots forming a sequence from continuous uniform production. Provided that all the constraints and actions demanded by BS 6001 are complied with, these procedures ensure that the average outgoing quality is maintained at the AQL. They are not appropriate for determining the acceptability of any lot submitted to satisfy an isolated lot demand, because they take no account of whether that lot is the first or subsequent attempt to fulfil the quality requirement. Use of BS 6001, or similar plans and procedures, inevitably entails the same repeated risk of accepting poor quality at each submission. The effect of successive submissions to meet an isolated lot demand is to make acceptance a virtual certainty after sufficient submissions, even if the quality consistently poor. A specially designed sampling scheme is required to limit the risk of acceptance of a poor quality lot; it must do this whatever the quality of the lots offered, including the worst case of uniformly poor ones.
This Standard outlines a method and gives sampling plans an procedures in which this risk level is limited to a defined small probability over a succession of submissions. The quality associated with this risk level is called the Lower Confidence Limit (LCL) which may be derived by this method from the sampling results obtained. The LCL will never be worse than a specified level, called the LQL, for a matched sampling plan, ie when the sampling plan number equals the submission number. It could be somewhat inferior if the sampling plan and the submission are unmatched.
Document History