Content area
Predictive maintenance has rapidly grown in automotive industries with the advancements in artificial intelligence (AI) technologies like machine learning, deep learning, and now generative AI. The amount of data extracted from machines with sensors and other network technologies can be valuable and useful for building advanced solutions in predictive maintenance tasks. This, in turn, helps improve vehicle up-time and reliability. This paper comprehensively reviews the different technologies and methods used for predictive maintenance. A systematic literature review of 94 papers was conducted from renowned databases such as Scopus and Web of Science. The paper reviews various techniques applied for predictive maintenance, highlighting the role of techniques in AI and the importance of explainable AI for predictive analytics. This review examines AI applications in vehicle maintenance strategies and diagnostics to reduce costs, maintenance schedules, remaining useful life predictions, and effective monitoring of health conditions. In addition, publicly available data sets relevant to predictive maintenance tasks are discussed, which play a crucial role in research and model development. The paper also identifies various challenges in predictive maintenance related to data quality, scalability, and integration of AI technology. In addition, emerging research topics within the domain are highlighted with future directions to address these challenges, thus optimizing maintenance strategies in the automotive industry.
Article Highlights
Comprehensive bibliometric and methodology review of machine and deep learning techniques applied in vehicle predictive maintenance.
Applications of emerging research topics in AI, such as explainable AI and generative AI with strategies to revolutionize predictive maintenance in the automotive domain.
AI applications in vehicle predictive maintenance strategies have implications for cost reduction, optimization of maintenance schedules, and accurate predictions of remaining useful life.
Details
Artificial intelligence;
Deep learning;
Maintenance management;
Generative artificial intelligence;
Data analysis;
Breakdowns;
Automobile industry;
Machine learning;
Time series;
Explainable artificial intelligence;
Keywords;
Learning algorithms;
Preventive maintenance;
Vehicles;
Literature reviews;
Bibliometrics;
Sensors;
Optimization;
In vehicle;
Systematic review;
Predictive maintenance;
Schedules