Premium
Risk Criteria for Nuclear Power Plants: A Pragmatic Proposal 1
Author(s) -
Starr Chauncey
Publication year - 1981
Publication title -
risk analysis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.972
H-Index - 130
eISSN - 1539-6924
pISSN - 0272-4332
DOI - 10.1111/j.1539-6924.1981.tb01406.x
Subject(s) - probabilistic logic , upper and lower bounds , margin (machine learning) , nuclear power , nuclear power plant , probabilistic risk assessment , statistics , risk assessment , risk analysis (engineering) , mathematics , computer science , medicine , computer security , biology , physics , machine learning , mathematical analysis , ecology , nuclear physics
Criteria are proposed for both an acceptable upper bound of nuclear power plant risk and a lower bound as a design target. Recognizing that the public risk associated with a power plant can be estimated only by probabilistic analysis of the design features, the spread between the lower design target and the upper bound provides a margin for uncertainty in th probabilistic estimate. The combination of a low probabilistic design target and this margin provides a reasonable expectation that the overall performance will be in the domain of an acceptable risk level. Because the exposure to potential risk is chiefly in the locality of the nuclear station, it is also proposed that compensatory benefits should be provided locally and that these be included as a cost of operation. It is suggested that the upper bound be set at a risk level equivalent to those risks of routine living which are normally accepted, i.e., about 10 ‐4 deaths per year per person (100 deaths/yr/million). The proposed lower design target is 10 ‐8 (0.1 deaths/yr/million), about one‐hundredth of the minimal risk from the natural hazards all people are exposed to.