EVALUATION OF PROBABILITY POINT ESTIMATE METHODS

In modern engineering designs and analyses, computer models are freque ntly used. Due to the presence of uncertainties associated with the mo del inputs and parameters, which are treated as random variables, anal ysis is feasible if the methods employed no not require excessive comp utations yet produce reasonably accurate results. Point estimate metho ds are such schemes that are potentially capable of achieving the goal s. Assuming normal distributions to the random variables, three point estimate methods (Rosenblueth's, Harr's, and a modified Hart's method) were evaluated in this paper for different numbers of random variable s and different model types. Results of this evaluation indicated that the proposed modified Harr's method yielded comparable, if not better , performance than the other two methods. Also, performance evaluation indicated that additional statistical information, other than the com monly used first two moments, should be incorporated if available, to enhance the accuracy, of uncertainty analysis.