Optimized Security Measure Selection: Leveraging MILP Solvers To Balance Risk And Cost

In this paper, we introduce an innovative opti- mization framework for selecting security measures, based on a granular, attribute-driven approach inspired by the SABSA methodology. Our model distinguishes between four types of measures: independent series, independent parallel, redundant antagonistic, and redundant correlated. The selection problem for security measures is inherently combinatorial and involves nonlinear constraints. To overcome this nonlinearity, we develop tailored linearization techniques, transforming the problem into a mixed-integer linear program- ming (MILP) formulation. The results obtained from practical case studies demonstrate that our method effectively identifies optimal or near-optimal solutions while ensuring that the overall residual risk remains below a specified threshold. Finally, the simulation component of our model is fully automated using Python’s DOcplex library, which optimally implements an MILP solver.