Decision support based on simulation

Abstract

The paper deals with a possible solution to the problem of choosing the best variant of a simulation model from a finite set of alternatives in the process of decision-making support. Evaluation of possible solutions is carried out via methods of multi-criteria analysis of alternatives. The integration of simulation models and models of decision support is set-theoretically formalized. The problem of choosing the rational set of simulation model parameters, considering restrictions imposed by experts, is solved by heuristic optimization method based on the local search strategy. Developed algorithms and software form a separate module in the decision support system (DSS) "Space" that makes it possible to rank alternatives submitted by simulation models. DSS software architecture is developed by means of object-oriented paradigm and is built of C# and T-SQL code. Combined with client-server model via thin-client architecture it gives additional flexibility to the users due to their option preferences of hardware and software. DSS “Space” allows ranking of simulation results using both traditional methods, such as weight and multiplicative function, ideal value and fuzzy and hybrid models of user’s preferences identification.Designed software allows applying it for variable scientific and technical civil and military problems due to its universality.