Relationship between risk assessment and reliability requirements

Many accidents have occurred and have assisted the development of many industries. Now some are called disasters, examples are given. We cannot avoid all accidents and need to consider some degree of improbability. Accepting limitations in knowledge and its use, the problem is how best to anticipate and reduce the risk of accident. Research is required to provide background basic knowledge and to establish some data base. Many assessments of systems, or pressure vessel behavior or emergency cooling provisions proceed on a step by step basis: the logical process variously called event or fault trees. After some review of current objections to risk/consequent assessment the paper considers the current topic of risk acceptability and leads into nonacceptability as a more favored route. The paper concludes with some discussion of individual and large scale societal risks in industrial activities