ABSTRACT
In 1983 the Royal Society published a survey of risk assessment which was probably the first attempt to cover the many different disciplines involved. All systems have a probability of failure, and engineers and scientists have to work to reduce that probability to an acceptable level. An anticipatory approach based on judgement and experience will be subject to substantial uncertainties due to the inadequacies of data and knowledge. The identification of hazards depends on the knowledge, experience and imagination of the analyst. Checklists of hazards are not an alternative to thinking, but should be regarded more as aidesmemoire. The analyst identifies the events on the next level down that could cause the top event, and determines whether they will happen as an AND function or an OR function. Severity scales can be developed to fit product line of the company. The extremes of a scale are usually fairly easy to determine, but the levels in between may require much consideration.
