00344 ¨901c63c87335b6f7bf31c8dbbac4a5

348

Prairie & Zimmer

Ä…uality. The purpose of the plans presented in this paper is the early detection of degredation in product reliability from the requirement.

In the next sections the plans are described and their characteristics developed and presented. Tables and figures are given that can be used for plan selection and implementation. An example application of the plans is demonstrated.

Description of Plans

Plan Features

A.    At the beginning of production, tests are performed at a 100% ratÄ™ until c consecutive successes are observed. The c value of the 100% testing is related to the reliability requirement.

B.    After the initial test is passed, reliability is monitored throughout production at reduced sampling rates, Level 1 and Level 2, and the risks associated with sampling are Ä…uantified.

C.    Units are not grouped into blocks.

D.    Guidance for testing additional samples is provided in the event of failure.

E.    Statistical process control is employed to complement the acceptance testing procedures and to reduce the risk of accepting a defective product.

It must be emphasized that the approach described here is appropriate for continuous production only; it is incompatible with lot production and should not be used for that purpose. An outline of the flow of the plans and their acceptance procedures is given in FigurÄ™ 1.

Determination of the Number of Consecutiye Units After a production has been designated for acceptance sampling, the sampling ratÄ™ shall be 100% until c consecutive units have been tested without failures. The c number depends, in part, on the reliability reÄ…uirement.

Values of c are shown in Table 1 for various values of the reliability requirements, R. The value P at the right of the table is the chance of getting