DISTINGUISHING BETWEEN CAUSES OF FAILURE USING PROBABILITY MAPS

摘要

In accordance with the classical reliability theory, structural failure is viewed as an excess of demand over supply - a situation when the load is greater than the structure's resistance to it. Since magnitudes of both load and resistance are uncertain, these are treated as random variables. Uncertainty is expressed in terms of the coefficient of variation. The ratio between mean resistance and mean load- the central safety factor - is used to characterise surplus of resistance over load. Coefficients of variation, central safety factor and assumption of normality in distribution of both load and resistance are used in construction of so-called probability maps. The map shows probability of failure for various values of central safety factor and uncertainties about load and resistance. A map is usually constructed for a certain coefficient of variation of load Once constructed, the map is typically used to read off probability ratings for alternatives - failure modes - under study, which enables rationally possible causes of failure and decision on the most likely one.