Risk and uncertainty in design and management

What is uncertainty and how should it be handled in the sister professions of design and management?

The HSE use the term "risk" and legislation requires that risks be assessed. Little usable guidance is available on how this should be achieved, however, and the published guidance concentrates on the simple, the obvious and the ridiculous, shedding little light on how risk assessment should operate in practice in complex, rapidly moving design or management environments.

The reason may be that, while the HSE work with accidents and their human and legal consequences, they understand neither design nor management. These skills, when rightly understood, involve a common process of problem solving (or critical thinking), which may be summarised as follows:

1. Identify and understand the problem, establishing criteria for a successful solution.
2. Generate many (all?) possible solutions
3. Assess these options against the previously established criteria
4. Select the optimum solution
5. Take action to put he solution into effect.

The real process may be linear or cyclical, involving several iterations with each interim solution revealing more about the problem and providing a better starting point for subsequent attempts.

The types of risk involved follow from this process and are:

1. Envisioning risk - solving the wrong problem
2. Ideation risk - developing poor alternatives
3. Evaluation risk - making the wrong choice
4. Execution risk - failure to implement the decision

Therefore proper control of risk, in both design and management, commences with the adoption of the correct problem solving strategy and continues with assessment of the various types of risk every time a decision is made. Also, whereas the HSE concentrates on health and safety risks as their name suggests, but risks of all types must be considered, that includes technical, financial, contractual, business and environmental hazards (ask BP about several of those).

We next come to a problem of definition; according to most dictionary definitions, risk and uncertainty or synonyms. However, a distinction drawn in the sphere of economics may be helpful here:

Risk is present when future events occur with measurable probability

Uncertainty is present when the likelihood of future events is incalculable or indefinite

(Ref Frank H Knight, Risk Uncertainty and Profit, 1921)

Thus risk assessment strictly refers to cases where probabilities can be quantified. In cases such as this, the risk can also be insured. It is what Donald Rumsfeld might have called "Known unknowns, things we know we don't know" Where uncertainty is the correct term, no quantified risk can be meaningful. Here perhaps are Mr Rumsfeld's "Unknown unknowns, things we don't know we don't know"

Only risks with measurable probabilities can be assessed in a numerical manner (giving an financial costs for the risk, which can be compared to the cost of an improvement). In design and management, this is the exception. Mostly we deal with uncertainty. The common practice is to get around this by assessing, rather arbitrarily, the probability and the consequences as low/medium/high based on the designer's/manager's subjective judgement. This simply obscures the true nature of the problem and puts the moment of forming a judgement off for a few minutes. More direct and helpful is to classify uncertainty as one of the following cases:


R1 Insignificant - Action None

R2 Significant but common -that is addressed by good standard practice
In construction it would be appropriate to rely on standard specifications for this category

R3 Significant and unusual, specific guidance on control of risk/uncertainty is required by operatives executing instructions.
In construction it would be appropriate to add specific instructions/guidance to drawings in addition to issuing full specifications, for example a detailed sequence of work statement might be added to drawings showing alterations to a building's structure.


R4 Unacceptable - Re-plan or redesign in part or in whole


R5 Extreme - Eliminate at source by redesign / abandon or rethink plans

Thus the pretense of a quasi-scientific assessment is dropped and uncertainty handled, as it must be, by judgement as to what is a reasonable response.

In forming this judgement, the manager or designer must educate himself by the study of hazards, accidents, failures, disasters, legal requirements and good practice, so as to be capable of sound assessment of new cases. Furthermore, the attention of several members of a diverse design or management team to risk and uncertainty issues may improve significantly the quality of the judgements formed - two (or more) heads are better than one.

In summary risk and uncertainty are related but not identical terms. The ability to make secure decisions in an environment of uncertainty is a key skill for both designers and managers. The limited approach the HSE brings to this subject, whilst undoubtedly better than the scant attention it previously received, is far from optimal. By making risk and uncertainty considerations integral to their decision making process, both managers and designers can improve their effectiveness, save money and perhaps lives.