Architectural Strategies for Automotive Predictive Maintenance: Balancing System Efficiency and Security Through Advanced Risk Analysis

The increasing complexity of modern vehicles calls for advanced predictive maintenance strategies to ensure safety, reliability, and security. While preventive maintenance improves vehicle lifespan, growing connectivity exposes systems to cybersecurity threats that may compromise operational integrity. This paper proposes architectural strategies for automotive predictive maintenance, balancing system performance with safety and cybersecurity. The Holistic Vehicle-Cloud Network Architecture (HVNA) integrates Internet of Things (IoT) devices, cloud computing, and machine learning to predict and mitigate failures in both in-vehicle and over-the-air communications. Cybersecurity assessment follows the International Organization for Standardization/Society of Automotive Engineers (ISO/SAE) 21434 standard, combining Medini Analysis with Threat Analysis and Risk Assessment (TARA) and the Spoofing, Tampering, Repudiation, Information Disclosure, Denial of Service and Elevation of Privilege (STRIDE) model to identify, classify, and mitigate vulnerabilities across system components and communication channels. Results reveal 162 threats in the vehicle architecture and 198 in the external communication architecture, including a notable share of high-level threats (1.8% and 5.6%, respectively). These findings highlight the importance of integrating predictive maintenance with robust cybersecurity measures to ensure resilience in connected and autonomous vehicles.