Development of a verification method of estimated indicators for their use as an optimization criterion

Abstract

The optimization criterion is that reference point of the controlled systems, which provides the maximum coherence of the operational process results with the purpose of its owner. Difficulties of solving the tasks, connected with the choice of an adequate optimization criterion are caused by the fact that a large number of indicators intended for use as an optimization criterion have been developed and continue to be developed now. The verification process of the indicator, which can potentially be used as an optimization criterion, is rather difficult. This difficulty is caused by the fact that the identification process is based on an exclusion method. Those indicators that have shown contradictory results on the classes of operations models identified by the rating efficiency should be excluded from the "contenders". During the researches, the hypothesis on the possibility of creating the reference models, allowing the development of direct verification methods wasn't confirmed. However, the comparative research of cybernetic models of operations with different duration for consistency has allowed setting a restriction on the formation rules of models of this class and defining the rules of their equivalent converting. Besides, during the researches, the method of creating a class of simple operation models with the predetermined identification, concerning their rating efficiency, has been developed. This means that the rating creation of these models doesn't require determination of local efficiency criteria. Creation of a new class of reference models expands the possibilities of the estimated indicator verification method. At the same time, the positive effect of expansion of the opportunities of the verification method is shown not so much in an increase in the probability of excluding the inadequate estimated indicator, but in the introduction of a restriction on the rules of creating the classes of reference operation models. Introduction of a restriction prevents the possibility of excluding the adequate estimated indicator from consideration.