# «SOLID-STATE LIGHTING PRODUCT QUALITY INITIATIVE THIRD EDITION SEPTEMBER 2014 Next Generation Lighting Industry Alliance LED Systems Reliability ...»

A system is a collection of components, subsystems, and assemblies arranged in a specific design in order to achieve desired functions with acceptable performance and reliability. The types of components, their quantities and qualities, and the manner in which they are arranged within the system have a direct effect on the system’s reliability. Often, the relationship between a system and its components is misunderstood or oversimplified.

From a system reliability point of view, the challenge is to master the reliability of its components. Clearly, each system, however complex, can only last as long as its shortest-lived component. Once a product’s failure criteria are established, the reliability may be measured and described in different ways depending on the particular

**situation. Examples are:**

Mean time to failure (MTTF) Number of failures per time unit (failure rate or field call rate) The probability that the item does not fail in a time interval [0, t] (survival probability) The probability that the item is able to function at time t (availability at time t ) If the item is not repaired after failure, the third and fourth bullets above coincide. (For precise mathematical definitions, refer to the textbooks in the References section.) Given the application requirements, one can calculate the reliability performance of the system. It is necessary to investigate the physics of failure in order to understand the failure modes or mechanisms, and obtain test information on the specific components for the design. Verification testing is also needed on a product level to ensure the model is correct, further reinforcing that it is generally only feasible to model a specific product or groups of products of similar design.

In system reliability analysis one will always be working with system models. In practical situations the analyst will have to derive these models, or at least choose from several possible models before an analysis can be performed. To be realistic the models should describe the essential features of the system, but do not necessarily have to be exact in all details. One approach is working with an idealized, simplified model of the system. Traditional handbook-based reliability prediction methods for electronic products include MIL-HDBK

The following informative publications provide information or guidance on reliability studies.

Dodson, Bryan, and Dennis Nolan. Reliability Engineering Handbook. QA Publishing, 1999. ISBN: 0-8247IEC International Standard 60300-3-1:2003, Dependability management – Part 3-1: Application guide – Analysis techniques for dependability – Guide on methodology, ed. 2.0, http://webstore.iec.ch/Webstore/webstore.nsf/ArtNum_PK/47606!openDocument.

Jensen, Finn V. An Introduction to Bayesian Networks, Volume 1. Springer, 1996. ISBN: 0-387-91502-8.

Nelson, Wayne B. Accelerated Testing: Statistical Models, Test Plans, and Data Analyses. John Wiley & Sons,

1990. ISBN: 0-471-52277-5.

Pecht, Michael. Product Reliability, Maintainability, and Supportability Handbook, Second Edition. Taylor & Francis, 2009. ISBN: 978-0-8493-9879-7.

Rausand, Marvin, and and Arnljot Høyland. System Reliability Theory: Models, Statistical Methods, and Applications. Wiley, 2004. ISBN: 0-471-47133-X.

Tobias, Paul A, and David C. Trindade. Applied Reliability. Chapman & Hall, 1994. ISBN: 0-442-00469-9.

32