The Risk and Uncertainty debate…
Posted by Prashant Hegde on June 15, 2007
There is a lot of discussion about the differences between risk and uncertainty. There is no universal agreement on the differences between two terms and the debate is still on!
Quoting Frank Knight – “It will appear that a measurable uncertainty, or “risk” proper, as we shall use the term, is so far different from an unmeasurable one that it is not in effect an uncertainty at all. We … accordingly restrict the term “uncertainty” to cases of the non-quantitative type..”. This essentially means that risk can be quantified in terms of probabilities where as uncertainty is not. According to Knight, there are two types of probabilities – objective and subjective. A probability is called objective probability if any of the following conditions are true -
- When there is symmetry in the information ( for eg: tossing a coin )
- Statistical analysis of experimental data is possible
He considered other probability assessments subjective( or as ‘personal probability’ as some authors prefer to call it). He considered subjective probabilities as uncertainties and not as risks. As some critics noted, there may exist more than one symmetry in a given situation and hence the probabilities might differ from person to person. And, hence the first condition is really subjective. In the real world scenario, we rarely get enough experimental data to carry out statistical analysis on them and deduce probabilities from them. Hence, the the second condition is also very difficult to achieve in practice. Even if we have enough data on a given situation how are we sure that we have complete data and not ignored relevant data. So, we can say most of the times we only have subjective probabilities. So, we can not really make a distinction between risk and uncertainty. Other authors have the view point that calculation of probabilities is inherently subjective and advise that expert opinions should be considered while calculating probabilities. So, this process is more like a Bayesian than a frequentest approach.
Even though there are other theories to distinguish between the two, no theory is able to explain the differences between the two terms convincingly. So, most people use them interchangingly. Many people use the risk in a negative sense( i.e as a threat) and treat uncertainty as both positive ( i.e opportunity ) and negative. PMBOK, for example, uses the term risk in both the positive and negative sense. Design For Six Sigma (DFSS) considers risk in the negative sense. It uses Risk Priority Numbers(RPNs) to quantify risks. System Safety Engineering carries out Hazard analysis to identify risks within the system to be built so that they can be managed safely. Some authors advise using options theory to deal with uncertainties. They advise building flexible systems so that favorable events can be exploited when they arise and the impact of unfavorable events are minimized.
