Abstract:
The article considers the application of a method for taking decisions under uncertainty, which is based on the expert estimates. It differs from the well-known methods in comprehensibility, simplicity of use and universality. The following practical problems were resolved: the need for data processing of different types, the need to obtain result when not coordinated expert answers, the need to substantiate of proximity of expert opinions, the need for processing of incomplete data, the undesirability of convolution of values of all indicators into a indicator of efficiency, the need to provide the results of expertise in an easily interpretable form. The search for the best solution is based on the construction of a probabilistic space of all possible opinions of experts. Alternative solutions are represented as areas of the consistent expert estimates. The procedure for imposing actual restrictions on decisions is described. A tool for comparing solutions is proposed.
Keywords:decison making techniques, expert estimates methods, forecasting, morphological analysis of data.