Background

When it was born as a separate engineering discipline in 1952, the reliability theory—i.e. the basis for safety and dependability studies—was already the result of a long knowledge acquisition process started in the forties (e.g. design of the FMECA approach). Until now, many related approaches have been developed (e.g. HAZOP, fault trees, Markov or Petri net techniques) to improve the analyses. Based on a probabilistic approach, they prove to be very effective when used in synergy with traditional deterministic approaches (i.e. implementation of rules, regulations and standards) and this chapter explains why, when and how to undertake such reliability engineering analyses. After a brief description of the above topics, the various names given to this new discipline are presented and the etymology of the term risk is analysed in order to highlight the serious semantic drifts observed nowadays and to introduce the acceptation adopted all along the book (combination chance x consequence) which leads to the introduction of risk matrices.