Due to various influences, product lifetime is a stochastic rather than a deterministic value. Therefore, it is commonly described by means of statistic measures, illustrated for example by the widely known model of the stress strength interference [1]. A convergence of loading capacity and occurring stresses over time, e. g. due to degradation effects, leads to an interference, and thus a product failure. The resulting probability of when a failure occurs, the expected time period without failures respectively, is described as reliability. Reliability is consequently defined as “the probability that a product does not fail under given functional und environmental conditions during a defined period of time” [1].

A variety of qualitative and quantitative methods is available to assess the expected reliability of products and components [1] as well as for the acquisition of trustworthy data required for a meaningful stochastic description of the product lifetime [2, 3]. Complemented by the calculation of confidence intervals, which accommodates the uncertainty of how good the used samples reflect the true population parameters [4], these approaches are an essential part of Reliability Engineering in industrial practice.

However, in addition to the importance of a systematic reliability assessment, various authors also emphasize the relevance of design decisions for the assurance of the required product performance over time in case of wear, corrosion, etc. [1, 5, 6]. Elsayed [5] for example clarifies that reliability indicators “may be used as a measure of the system’s success in providing its function properly during its design Iife” indicating that reliability largely depends on the quality of designs [7].

The overall objective of Reliability Engineering consequently is the comprehensive analysis, assurance and improvement of the expected product reliability by means of suitable qualitative and quantitative methods. Including a variety of tasks, such as the systematic description of failure causes, the identification of relevant components, the assessment of failure probabilities extended by a calculation of confidence intervals for the drawn conclusions, Reliability Engineering is a highly complex but, at the same time, extremely important challenge for quality assurance purposes in industrial development projects.

Whereas Meeker [2] and Nelsen [3] offer an overview of suitable data acquisition approaches, further information on the calculation of confidence intervals can be found in O’Connor, Kleyner [4].

[1] Bertsche, B., Lechner, G.: Reliability in Automotive and Mechanical Engineering. Berlin, Heidelberg, New York: Springer-Verlag 2007.
[2] Meeker, W., Escobar, L.: Statistical Methods for Reliability Data. New Jersey: John Wiley & Sons, Inc. 1998.
[3] Nelson, W.: Applied Life Data Analysis. New Jersey: John Wiley & Sons, Inc. 2004.
[4] O’Connor, P., Kleyner, A.: Practical Reliability Engineering. New Jersey: John Wiley & Sons, Inc. 2012.
[5] Elsayed, A.: Reliabiltiy Engineering. New Jersey: John Wiley & Sons, Inc. 2012.
[6] Taguchi, G. et al.: Taguchi’s Quality Engineering Handbook. New Jersey: John Wiley & Sons, Inc. 2005.
[7] Yang, G.: Life Cycle Reliability Engineering. New Jersey: John Wiley & Sons, Inc. 2007.