Cyberattacks remain a critical threat to Industrial Control Systems (ICS) and the Internet of Things (IoT), despite their integral role in modern enterprises. Many of these attacks go undetected or unreported, posing significant risks to operational integrity, data security, and public safety. Traditional IT security measures, such as firewalls and anti-malware software, are often insufficient in mitigating these vulnerabilities. While deep learning models have been utilized in prior research for attack detection, they lack optimization for computational efficiency, require substantial computing resources, struggle with non-image cybersecurity data, and are susceptible to overfitting. This research introduces a customized Artificial Neural Network (ANN) enhanced by quantum computing techniques and optimized using Particle Swarm Optimization (PSO). Unlike traditional models, the proposed quantum-enhanced approach mitigates local minima issues, reduces computational complexity, and scales efficiently with increasing data complexity. The optimized model achieves an accuracy of 99.22%. Furthermore, Local Interpretable Model-agnostic Explanations (LIME) enhance model interpretability and assess feature importance. This study establishes a more robust and efficient framework for securing ICS and IoT against cyber threats by integrating explainability techniques, quantum computing, and bioinspired optimization algorithms.