Nuclear energy policy analysis under uncertainties

Abstract

For the purpose of analyzing the nuclear energy policy under uncertainties, new utility theoretic approaches were applied. The main discoveries of new utility theories are that, firstly, the consequences can affect the perceived probabilities, secondly, the utilities are not fixed but can change, and finally, utilities and probabilities thus should be combined dependently to determine the overall worth of risky option. These conclusions were applied to develop the modified expected utility model and to establish the probabilistic nuclear safety criterion. The modified expected utility model was developed in order to resolve the inconsistencies between the expected utility model and the actual decision behaviors. Based on information theory and Bayesian inference, the modified probabilities were obtained as the stated probabilities times substitutional factors. The model theoretically predicts that the extreme value outcomes are perceived as to be more likely to occur than medium value outcomes. This prediction is consistent with the first finding of new utility theories that the consequences can after the perceived probabilities. And further with this theoretical prediction, the decision behavior of buying lottery ticket, of paying for insurance and of nuclear catastrophic risk aversion can well be explained. Through the numerical application, it is shown that the developed model can well explain the common consequence effect, common ratio effect and reflection effect. The probabilistic nuclear safety criterion for core melt frequency was established: Firstly, the distribution of the public``s safety goal (DPSG) was proposed for representing the public``s group preference under risk. Secondly, a new probabilistic safety criterion (PSC) was established, in which the DPSG was used as a benchmark for evaluating the results of probabilistic safety assessment. Thirdly, a log-normal distribution was proposed as the appropriate DPSG for core melt frequency using the m...