Guidance from the U.S. Department of Energy (DOE) for the performance
of accident analysis in support of environmental impact assessments ca
lls for a graded approach that considers frequencies as well as conseq
uences of accidents, focuses on high-risk scenarios, and avoids boundi
ng analyses that can obfuscate comparisons of alternative actions. Thi
s guidance reflects the fact that at the heart of an environmental imp
act statement is a comparative analysis of alternatives, including the
proposed action; this analysis should address the environmental impac
ts in proportion to their potential significance, avoid addressing ins
ignificant impacts in detail, and focus analysis resources to be as co
st-effective as possible. Accordingly, a probabilistic risk analysis-b
ased methodology to satisfy DOE guidance was developed and implemented
for the DOE Waste Management Programmatic Environmental Impact Statem
ent (WM PEIS). The methods are described and illustrated, and an outli
ne of the computational framework is presented. This methodology, alth
ough developed for the WM PEIS, is, of course, applicable to general s
afety analyses. The implementation of the methods for the WM PEIS is s
ummarized, and the extension of the methods to site-specific applicati
ons is explained.