Multi-criteria optimization for selecting an information security system for a university’s distributed computing network based on NSGA-III

The growing number of cyber threats and the increasing complexity of university distributed computing networks necessitate a multifactor approach to selecting an information security system (ISS). This paper proposes a multi-criteria model for ISS selection. The model is based on the Nondominated Sorting Genetic Algorithm III (NSGA-III) and addresses conflicting objectives—minimizing implementation and operational costs, reducing residual risk, and maintaining infrastructure performance. The scientific novelty of this work lies in the integration of bilinear quality game outcomes into the optimization process. The game outputs are used as dynamic constraints reflecting the interaction between protection measures and current threats. The developed mathematical framework, incorporating formalized solution encoding and a constraint-handling strategy, demonstrated its effectiveness using data from university networks in Ukraine and Kazakhstan, ensuring a balanced Pareto front of solutions. The limitation of this study is the use of classical NSGA-III without iterative information exchange with the game model. Future research directions include adaptive parameter tuning and dataset expansion.