What Is Wrong with Criterion FN-Lines for Judging the Tolerability of Risk?

This paper is concerned with the intellectual framework in which judgments are made about the tolerability of so-called societal risk. The current practical approach is based on the position of the FN-curves representing the risks from hazardous systems in relation to criterion FN-lines. The objections to FN-criteria are that they can give unreasonable conclusions and that they are inconsistent. Statistical decision theory suggests an alternative and preferable rule of minimising the expected disutility, that is average harm, from accidents.