Development of the scientific and methodological framework for the intellectual assessment of parameters in decision support systems

The object of the research is decision support systems. The subject of the research is the process of evaluating parameters in decision support systems. The study proposes: a methodology for the intellectual assessment of parameters in decision support systems and a method for multi-criteria evaluation of hierarchical systems. The originality of the research lies in the use of additional advanced procedures that allow: – verification of the topology and parameters of decision support systems, taking into account the degree of uncertainty in the input data regarding the known information, achieved through the use of an enhanced penguin swarm algorithm. This reduces the time required for the initial configuration of the evaluation methodology during its setup; – initial selection of individuals for configuring the evolving artificial neural network, carried out using an enhanced genetic algorithm, which reduces solution search time and increases the reliability of the obtained results; – exploration of solution spaces for the problem of parameter evaluation in decision support systems, which are described by atypical functions, using the enhanced penguin swarm algorithm; – configuration of the weights of the evolving artificial neural network improves the accuracy of parameter evaluation in decision support systems; – utilization of additional mechanisms for correcting the parameters of the evolving artificial neural network is applied by a procedure for modifying the membership function; – reliability of parameter evaluation in decision support systems is increased by parallel evaluation through multiple assessment methods; – use of hybrid parameter evaluation for decision support systems allows for correct operation in the absence of conditions of stationarity, homogeneity, normality, and independence. An example of the application of the proposed scientific and methodological framework for evaluating parameters in decision support systems showed a 25% increase in evaluation reliability, achieved by utilizing additional procedures while maintaining the required operational efficiency.