Why Risk Managers must be Schizophrenic!

It may sound strange but risk managers operate in two different modes, or frames of mind. From day to day, the mind set must be 'all accidents are avoidable'. When carrying out a risk assessment the mind set changes from that of planning and thinking about 'successes' to considering 'failures'. Does this mean that when accidents do happen that there is a culpable failure to manage risk adequately? Clearly not.

Part of the problem for risk managers is that outcomes from an apparently similar starting set of 'conditions or events' can be widely different. Meteorologists call this 'the butterfly effect'. Chemical engineers like myself (along with statisticians) call it a 'stochastic process'. Dictionary.com defines stochastic as 'of or pertaining to a process involving a randomly determined sequence of observations each of which is considered as a sample of one element from a probability distribution'. As David Spiegelhalter says in his excellent, short Worrier's Guide to Risk', '2. Stuff happens. The overall pattern of events can often be predicted surprisingly well but not the detail. We can make a good guess at the number of car fatalities next year, but not who will be involved.'. Risk managers cannot manage the fine detail of every road journey but can do their best to ensure that vehicles and drivers meet standards which cover the common hazardous situations. It is the same for major hazard plant risk managers although the depth of analysis of the hazardous situations must be much greater, hence the standards much higher, for obvious reasons.

In my experience as a regulator, the failure to be sufficiently schizophrenic is most apparent when looking at the predictive elements of a SEVESO safety report. Often the author has failed to remove their 'rose tinted spectacles' and consider unlikely failure sequences because 'it just won't happen'. It is these unlikely failure sequences which often lead to accidents of the highest severity. If you don't think about unlikely, high severity, accidents, how can you make a sensible judgement about the necessary measures to put in place to manage the risk?

One other aspect of the stochastic nature of event sequences is that accidents never happen quite the same way twice. This casts doubt in my mind on the large sums of money spent on investigating complex accidents in great detail without considering whether the lessons learnt will be of value in the future or will only help to explain the past, the outcomes of which are already made plain in the aftermath of an accident. Would the money being spent on investigating the Buncefield explosion phenomena be better spent on researching more general accident phenomena?

This work is licensed under a Creative Commons Attribution-NoDerivs 2.0 UK: England & Wales License.