دانلود رایگان مقاله عدم قطعیت مدل سازی در تحلیل پذیری چند ضلعی تصادفی

Abstract

This paper considers problem contexts in which decision makers are unable or unwilling to assess tradeoff information precisely. A simulation experiment is used to assess (a) how closely a rank order of alternatives based on partial information and stochastic multicriteria acceptability analysis (SMAA) can approximate results obtained using full-information multi-attribute utility theory (MAUT) with multiplicative utility, and (b) which characteristics of the decision problem influence the accuracy of this approximation. We find that fairly good accuracy can be achieved with limited preference information, and is highest if either quantiles and probability distributions are used to represent uncertainty.

5. Discussion

Since SMAA always selects the alternative with the largest acceptability index, the probability that the SMAA model selects the same alternative as MAUT is equivalent to computing the probability that the decision maker's true preferences lie inside the largest favourable weight space. This is precisely what is computed by the largest rank-1 acceptability index. We therefore examine distributional properties of the random variable B1 ðkÞ denoting the k-th largest rank-1 acceptability index, and in particular the expected value of the largest rank-1 acceptability index E½B1 ð1Þ. Unfortunately deriving general analytical results for B1 ðkÞ is difficult, so that we first consider an artificially simple case where exact results can be obtained, and then extrapolate from these using heuristic arguments.