McLinn, James A. (1991, ASQC) McLinn and Associates, Loretto, MN
This abstract is an edited version of the author's original.
This paper explores several methods for using the reliability prediction process. The reliability prediction is an important tool during the development cycle of a new product; it yields the first estimate of the level of customer satisfaction that the new product may afford and provides the first indication of the level of exposure of the company to possible warranty claims. The two elements required to make a reliability prediction are failure rates of components and reconciling the field environment with the stresses on the system.
The methods discussed in this paper include: (1) the tyranny of the twos, (2) infant mortality incorporation, (3) a new learning curve approach, (4) the effects of key processes on product reliability, (5) stresses on key parts, (6) the supplier quality connection, (7) design verification or quality in design factors, (8) corrections for low system reliability situations, (9) environmental fine tuning, and (10) correction factors for quality systems in place in manufacturing.
Some of these new methods are appropriate for predicting hardware or software processes, and some methods focus on manufacturing or design systems. Also, some methods give definite recommendations and guidelines, while others depend on "good engineering judgment.
Reliability,Customer satisfaction (CS)