ABSTRACT
This chapter begins by discussing safety goals, particularly quantitative ones. It is clear that such goals need to specify not only numerical levels to be achieved, but also how the risk is to be estimated and how the comparison is to be made. Technological risks are imposed on society because the benefits which result from the technology outweigh the risks. This is the fundamental social safety goal. An evolutionary approach might aim to set targets which are fairly close to existing practice, using quantities which can be reasonably easily estimated with a moderate degree of uncertainty. The technical community should try to explain its need for quantitative targets as an engineering necessity. The problems of uncertainty could be further reduced if rules are made by the regulatory authority on how the remaining uncertainties should be dealt with. One such evolutionary approach can be inferred from the Health and Safety Executive (HSE) discussion document.
