Research Information System
5th International Conference on Modern and Advanced Research (ICMAR 2026), Konya, Turkey, 7 - 08 May 2026, vol.1, no.1, pp.333-340, (Full Text)
Artificial intelligence (AI) technologies are among the most transformative innovations in modern engineering systems. AI-based optimization approaches offer significant advantages for the decision-making, operational efficiency, adaptive control, and predictive analysis in the critical engineering domains including defense industries, aerospace systems, autonomous vehicles, industrial automation, healthcare infrastructure, and smart manufacturing. However, the growing complexity of AI models has also introduced important concerns on transparency, explainability, reliability, cybersecurity and trustworthiness. The inability to interpret AI decisions can have serious operational, ethical, and security ramifications in mission-critical environments. Hence, explainable and secure AI optimization approaches are emerging as critical components of next-generation engineering infrastructures. The paper provides a general review of secure and explainable AI optimization approaches for critical engineering systems. The paper discusses the significance of explainable artificial intelligence (XAI), reliable optimization techniques, swarm intelligence-based optimization, and autonomous decisionmaking mechanisms for engineering applications. In addition, we discuss issues on AI security, adversarial attacks, ethical risks and system reliability. Some representative examples of critical infrastructures are studied, namely defense industry applications and autonomous engineering systems. Finally, future perspectives including explainable autonomous systems, trustworthy AI, digital twins, federated optimization and human-centered AI systems are evaluated.