AI-DRIVEN CYBERSECURITY, IOT NETWORKING, AND RESILIENCE STRATEGIES FOR INDUSTRIAL CONTROL SYSTEMS: A SYSTEMATIC REVIEW FOR U.S. CRITICAL INFRASTRUCTURE PROTECTION
DOI:
https://doi.org/10.63125/mbyhj941Keywords:
AI-Driven Cybersecurity, Industrial Control Systems, IOT Networking Maturity, Cyber-Resilience, Critical Infrastructure ProtectionAbstract
This study investigates how AI-driven cybersecurity, IoT networking maturity, and resilience strategies jointly enhance the protection of industrial control systems within U.S. critical infrastructure. The problem driving this research is the accelerating convergence of ICS, IoT, and cloud architectures, which expands cyber-attack surfaces while many operators still lack integrated AI-enabled detection and resilience frameworks. The purpose of the study is to evaluate, through quantitative evidence, how AI-based intrusion and anomaly detection, secure IoT networking, and resilience engineering contribute to perceived critical-infrastructure protection effectiveness. Employing a quantitative, cross-sectional, case-based research design, the study collected data from 210 practitioners representing energy, water, transportation, and manufacturing ICS environments. Using multi-item Likert five-point scales, the study measured six key variables: AI-driven cybersecurity capability, IoT networking maturity, ICS resilience strategies, governance maturity, behavioral compliance, and perceived protection effectiveness. Descriptive statistics, Pearson correlations, and multiple regression modeling were used to test the proposed conceptual relationships. Findings show robust adoption of AI-driven security (mean = 3.82) and resilience strategies (mean = 3.74), with all constructs demonstrating high reliability (α = .87–.92). Regression analysis revealed that AI capability strongly predicts resilience (β = 0.38, p < .001) and, together with resilience, significantly improves perceived protection (β = 0.34 and β = 0.37, p < .001). IoT networking maturity also positively affects AI adoption (β = 0.23, p < .001). The study concludes that AI adoption is most effective when embedded in mature IoT security architectures and formal resilience programs. Implications highlight the need for integrated AI–IoT–resilience strategies as core pillars of national critical-infrastructure protection.