[en] This paper demonstrates a method of taking uncertainty into account in an economic study comparing nuclear and coal generation costs. The study compared similarly sized units of 1200 MW with in-service dates in 1987. A Monte Carlo simulation was employed to generate probability distributions for the costs of each of the alternatives. Also, a difference probability curve was generated that suggests the probability of success or failure in choosing one alternative over another. Results indicate that nuclear power has a lower expected cost than coal, but is more uncertain than a coal project. In spite of this uncertainty, the probability that nuclear power will be more costly than coal was shown to be approximately 20% or less, based on the assumptions made. A sensitivity study was also performed that showed the effect that different assumptions with respect to nuclear accidents have on the results. Even with pessimistic assumptions regarding nuclear accidents, it was shown that the decision maker can be approximately 80% confident that the nuclear project will be more economical than coal. 2 figures, 3 tables