Methodology for searching multi-criteria solutions based on digital twins

Abstract

Aim. To justify and propose a methodology for searching multi-criteria Pareto-optimal solutions based on the use of socio-economic models or digital twins of complex dynamical systems.

Tasks. To substantiate the relevance of using predictive methods for making managerial decisions in conditions of rapid changes and the need to use for these purposes recommendation systems using socio-economic models or digital twins. To propose a methodology for searching multi-criteria Pareto-optimal solutions and to evaluate its accuracy on mathematical and socioeconomic models. To formulate recommendations and limitations on the use of this methodology.

Methods. General scientific methods (analysis, synthesis, grouping), mathematical methods of searching for Pareto-optimal solutions and generation of LP -sequences, results of mathematical modelling and numerical experiments were used in the research.

Results. The methodology of searching for multicriteria Pareto-optimal managerial decisions is proposed, proposals and limitations on its application are formulated.

Conclusions. In conditions of rapid changes it is important to be able to make managerial decisions that take into account the influence of many factors and, accordingly, many criteria characterising the socio-economic system, and at the same time it is necessary to make these managerial decisions in accordance with the logic “from the future”. The method of searching for multi-criteria Pareto-optimal solutions based on the use of socio-economic models or digital twins is proposed. The accuracy of this method is investigated on mathematical models of different degrees of nonlinearity with different number of parameters when solving production multi-parametric problems.