ABSTRACT

This chapter begins to open a more detailed argument on the measurement of risk with a set of logical challenges to the normative form of industrial risk measurement, i.e. the combination of probability and impact judgement for discretised risk events. The presence, or absence, of knowledge can be shown to be the true determinant of the utility of such a risk measurement and a mature approach is therefore needed. To reason with probability in particular, the history of the disputed and unresolved science of the relationship between the concept of risk and the measurement of probability shows that great caution is also needed. Valid risk measurement will always be statistically complex. Accepting that probability inference for risk measurement is such a disputed and complex area of science calls into considerable question most of the mainstream approaches that use the normative form. The tendency to reason outside of this requisite science, on the grounds of a lower appetite for complexity, is considered extremely suspect. Using flawed reasoning structures predicated on poor, or absent, knowledge and a misappropriation of statistics – because they seem like a good way to simplify complex scenarios – is not only indefensible, it is potentially destructive of business value.