Introduction to structural reliability

Although rare, events such as ruptures of pipes, tanks or enclosures can have extremely serious consequences. Until recently, we protected ourselves against such events by taking margins, or safety factors, which we accumulated over all influential parameters, using a purely deterministic methodology. However, this approach does not provide a clear picture of the risks involved. In structural reliability analysis, influential parameters are considered as random variables and, based on a physical failure equation whose variables are probabilized, the probability of failure is calculated. Industrial applications are numerous: optimizing maintenance and inspections, calculating residual service life, etc.

Experience feedback and knowledge of degradation kinetics are two of the essential prerequisites for applying structural reliability analysis.

Structural design is based on a regulatory and codified approach, essentially deterministic. Safety coefficients are introduced into the calculations to ensure that large margins are maintained to guarantee structural integrity.

This type of design does not allow us to assess the risk of failure of a structure, or its reliability. It generally gives a deliberately pessimistic margin with regard to the various possible modes of failure, and more often than not leads to unjustified over-dimensioning, and therefore to additional costs. What's more, certain loads can be ignored at the design stage and only discovered during operation. Structures age, material properties may be altered, and operating methods may no longer be the same.

The probabilistic approach to structural reliability is therefore essential. Risk is assessed as a probability, rather than as a binary judgement (design is acceptable or not, operation can be continued or not).

By calculating this probability, we can reduce the risk of failure by organizing maintenance-inspection programs, and extend operating life by optimizing their use.

This article describes a few aspects of structural reliability. After identifying some of the causes of structural failure and the associated preventive measures to avoid failure, we review the main characteristics of passive components. We deliberately take an operator's view, seeking to minimize the risk of failure of a structure and optimize its operation. A brief presentation is given of the reliability-based maintenance optimization methodology (OMF-structures), applied to passive components, one of whose roles is to optimize in-service inspections and maintenance. The fundamentals of structural reliability calculation are briefly explained. Interested readers can refer to the referenced works for further developments and industrial...