Critical views on the application of some methods for evaluating accident probabilities and consequences

Abstract

The mathematical expectation of consequence (also referred to as the detriment) is often used as a measure of the potential impact of a source of risk. Boundary curves such as those used in Farmer diagrams may be used to indicate the non-acceptability of certain foreseen risks, for which the detriment exceeds a preset limit. These curves must not be confused, as sometimes happens, with probability density functions expressing the accident probability within differential intervals of consequence. Another useful way of presenting probabilities and consequences, used in m any safety assessments, is to present the complement of the cumulative distribution, i.e. the probability that the consequence will exceed given values. The integration of this function gives the detriment. At very low probabilities, the detriment (the product of the low probability and the con sequence) will lose its usefulness as a basis for decision-making. It is then no longer a central measure of the distribution and, in addition, the uncertainty of the detriment becomes too large to make it meaningful, even if the probability as such could be estimated by safety assessments with an accurate degree of certainty.