Analyzing Car-Hacking Dataset to Understand Cyberattacks and Intrusion Detection in Autonomous Vehicles

As autonomous and connected vehicles grow in popularity, securing in-vehicle networks has become a critical operations research (OR) challenge. This study applies OR principles to cybersecurity by framing intrusion detection as a multi-criteria decision optimization problem. We analyze the Car-Hacking dataset to assess its effectiveness in supporting security countermeasures across various attack vectors in Controller Area Network (CAN) systems. Using Decision Tree, Random Forest, Gradient Boosting, and Convolutional Neural Networks, we compare classification models as decision optimization tools. The evaluation highlights how OR techniques balance detection sensitivity and specificity in optimizing vehicle security. Our contribution includes a quantitative framework for assessing security datasets and shows how structured decision-making enhances adaptive defenses in transportation systems. The findings emphasize the importance of representative datasets for developing robust security models capable of addressing complex vehicular network threats.