RISK ANALYSIS

We now turn away from general ideas to detailed comments on risk assessment. However, we are faced with the immediate problem that the applications of risk assessment are wide ranging, with different objectives, terminologies and technical demands. The nature of a quantified risk assessment for a petrochemical plant would be different from that for many environmental risk assessments and very different from the risk analysis involved in the development of a structural design code. Therefore, as this chapter must cover all risk assessment, it is not possible to give a detailed list of specific tasks; in any case, specific applications are dealt with later in the book. Instead, I shall discuss fundamental factors that all risk assessments seem to have in common.

Common Elements of Risk Assessment

There are four underlying ideas common to all risk assessments that need to be discussed in detail. They are:

1. There is a need for tight discipline.

2. The development of a risk assessment is cyclical.

3. The process is a team effort.

4. There are a number of tasks common to most assessments.

The need for tight discipline Naturally, all engineering tasks need to be approached with disciplined thinking. Woolly thinking produces woolly results. Nevertheless, a tight and consistent methodological approach seems particularly necessary for risk assessment problems. It needs to be stressed at all levels.

A major reason for the importance of discipline seems to be that 'risk' is a complex and sophisticated concept. It thus needs a high degree of attention to its meaning and appropriate usage, which is the reason for the emphasis on its philosophical underpinnings earlier in the chapter. At the overall level, a systems approach is needed; that is a clear overview must be kept in focus while at the same time paying attention to detail.

There is an implication here that it is important to get the complete system defined and understood. A tightly disciplined approach is also needed at the detailed level, where in a sense the analyst is playing a game with the problem in hand according to a set of rules. The rules are flexible and can be adapted to the task in hand, but, once set, they should be accepted rigorously. Here again the 'game' is more complex than most engineering activities, thus requiring increased attention to consistency.