Reliability Analysis in Product Development

Abstract

In 1999 the author met managers of a “certified” company developing a new engine; he saw use of the “Duane Method” for predicting the in-service MTBF by elaborating the test reliability data. The paper shows the relationship between the Reliability R(t), the “mean number of failures” M(t) and the predicted MTBF using both the “Integral Theory of Reliability” and the “Duane Method”; the difference between the two methods enhances their respective real applicability. Later he saw “fuzzy Theory” used to evaluate product Quality: is that scientific? Since the absence of the scientific approach before, during and after the experiments typically results in relatively uninformative output of questionable general validity, in order to make sound decisions, managers have to be aware of the consequences of their decisions. The same is valid for Lecturers at Universities. This means that professors MUST teach, in a scientific way, Quality ideas on Quality. Unfortunately several people with very little competence, knowledge and experience are in charge of teaching. Professors need Quality metamorphosis.