A Framework For Real-Time Root Cause Analysis In Connected Vehicle Iot Data Streams Using Aiops
Naresh Kalimuthu · International Journal of Emerging Trends in Computer Science and Information Technology · 2025
The development of connected vehicles is generating unprecedented volumes of IoT data, which in turn impacts the functionalities of modern transportation systems. This puts pressure on conventional IT Operations Management (ITOM) frameworks. In this paper, we propose a novel, multi-layered architecture that integrates AIOps (Artificial Intelligence for IT Operations) for real-time fault prediction and autonomous Root Cause Analysis (RCA) within the context of a connected vehicle ecosystem. The architecture integrates vehicle onboard diagnostics, edge computing, and cloud computing to manage efficient data and workload analytics. It contains a hybrid predictive engine that applies lightweight statistical models for low-latency anomaly detection and advanced cloud deep learning models for recognizing complex failure patterns. For diagnostics, we propose a graph-based RCA engine that dynamically models the V2X system's interrelationships to determine the rapid and precise origin of failures. We address the challenges of latency, data scalability, and model explainability, proposing solutions for each. This study aims to propose an operational intelligence framework for connected mobility solutions