9.1 WHEN AND HOW TO APPLY

In accept-reject testing two types of risks are considered:

1. Consumer's risk - This is the probability that the customer (the buyer) accepts equipment with a reliability lower than the minimum acceptable, or R _{L 1}, and is designated by ?. Hence

   ? = probability that equipment with a reliability of R ? R _{L 1} passes the acceptance test,

   or

   P( AcceptR ? R _{L 1}) ? ?,

   where ? is usually chosen to be 1%, 5% or 10%. By imposing such a limitation on ? the customer is protected against accepting equipment with lower than acceptable reliability, in the long run.

2. Producer's risk - This is the probability that the manufacturer produces good equipment with reliability as high as, or better than R _{U 1}, the reliability goal to design to, and yet the acceptance test rejects the equipment, and is designated by ?. Hence

   ? = probability that equipment with a designed-in reliability of R ? R _{U 1} fails the acceptance test,

   or

   P( RejectR ? R _{U 1}) ? ?,

   where ? is usually chosen to be 1%, 5% or 10%. ? and ? may be chosen to be equal or different.

The acceptance test consists of running N units for the specified time (mission duration) and accepting the equipment if a or fewer units out of