AUTOMOTIVE SYSTEM RELIABILITY AND TECHNOLOGICAL CONVERGENCE: A REVIEW OF SMART POWERTRAIN AND MECHATRONIC DIAGNOSTICS

Abstract

The evolution of automotive systems has accelerated with the integration of intelligent electronics, embedded sensors, and AI-driven software, resulting in increasingly complex vehicles that demand advanced diagnostic and reliability frameworks. This systematic review explores the convergence of technologies that underpin smart diagnostics, focusing on powertrain systems, mechatronic diagnostics, artificial intelligence, digital platforms, and cybersecurity. Guided by the PRISMA 2020 methodology, a total of 112 peer-reviewed articles were rigorously analyzed to identify key advancements, application trends, and emerging challenges in automotive fault diagnostics and system reliability. The findings reveal a paradigm shift from traditional, reactive maintenance approaches to predictive and condition-based strategies enabled by real-time sensor monitoring, machine learning algorithms, and cloud-edge computing architectures. Notably, the implementation of AI techniques—such as convolutional neural networks, support vector machines, and unsupervised learning models—has enabled earlier and more accurate fault classification across critical systems, including powertrains, batteries, and thermal subsystems. The review also highlights the growing adoption of digital twin technologies, which allow virtual modeling of vehicle components for predictive maintenance, system optimization, and remote diagnostics without physical testing. Additionally, the incorporation of cybersecurity frameworks, particularly SAE J3061 and ISO/SAE 21434, is shown to be essential in protecting diagnostic systems against evolving digital threats in connected vehicle environments. Cloud-based and edge-based diagnostic platforms emerged as scalable solutions for managing real-time fault data, supporting over-the-air updates, and ensuring rapid decision-making in distributed vehicle networks. A recurring challenge identified across the literature is the widening skill gap in diagnostics engineering, particularly in the application of AI and systems integration, which hampers effective technology adoption in industry settings. Furthermore, the review underscores the importance of standards compliance and system interoperability to ensure diagnostics consistency across multi-vendor environments. By synthesizing insights from 112 high-impact studies across multiple disciplines, this review offers a comprehensive assessment of current capabilities, limitations, and future directions in smart diagnostics and automotive reliability management. It serves as a critical resource for researchers, engineers, and industry leaders aiming to optimize vehicle performance, safety, and maintainability through intelligent diagnostics technologies.