Table 5 Top 10 most relevant articles in...

Table 5

Top 10 most relevant articles in automotive

Source	Objectives	AI model	Maintenance aspect	Short outcomes
Jakobsson et al. (2020)	Predict the remaining useful life of heavy mining vehicles based on their usage data	NN	Predictive maintenance	The model accurately forecasts vehicle component life using operational data, enabling proactive maintenance scheduling and reducing unexpected heavy vehicle breakdowns
Gong et al. (2023)	Diagnose and monitor electric vehicle faults using indirect measurements combined with AI	AI-driven signal processing	Predictive maintenance	The proposed methodology quickly and accurately detects EV motor and battery faults using only indirect signals, improving maintenance response times and reducing the need for additional sensors
Gugaratshan et al. (2023)	To continuously update reliability, availability, maintainability models of vehicles using in-service sensor data	ML	Maintenance planning	Demonstrated that incorporating real-time multi-variate sensor data into reliability models significantly improves failure prediction and optimizes maintenance timing for critical vehicle subsystems (e.g. brakes, engine)
Chen et al. (2021)	Predict automotive maintenance needs by integrating deep learning models with GIS data	Deep Learning (e.g. CNN/LSTM)	Predictive maintenance	Successfully predicts when and where vehicle maintenance is required by analyzing operational routes and environmental conditions, allowing timely, location-aware maintenance actions for vehicle fleets
Giordano et al. (2022)	Evaluate and derive best practices for implementing predictive maintenance in an automotive production environment	ML	Predictive maintenance	Highlighted effective data-driven maintenance strategies (and common pitfalls) in a real automotive case study, providing guidance on data quality, model selection and integration for future implementations
Giobergia et al. (2018)	Use large volumes of sensor data from vehicles to predict component failures in advance	Data mining and ML	Predictive maintenance	Showed that mining high-dimensional vehicle sensor data can uncover early warning patterns for component failures, thereby improving fault detection and enabling maintenance to be scheduled before breakdowns occur in automotive fleets
Liu et al. (2021)	To diagnose faults in electric/hybrid vehicles using enhanced machine learning algorithms	Improved ML algorithm (e.g. enhanced SVM/ANN)	Predictive maintenance	The improved machine learning model accurately identifies battery and motor anomalies in new energy vehicles, contributing to more reliable operation and timely maintenance of electric vehicle fleets
Kavousi-Fard et al. (2020)	To detect anomalies in vehicle operation data for both security and reliability purposes	DL	Anomaly detection	The model detects abnormal patterns in vehicle data (potentially indicating component faults or cyber-attacks) with high accuracy, providing an early warning system that can trigger maintenance or security interventions promptly
Ji et al. (2021)	Identify anomalies in the clutch operation of hybrid electric vehicles using machine learning	ML	Condition monitoring	Successfully detected unusual clutch engagement patterns that indicate wear or faults in hybrid vehicle transmissions, enabling targeted preventive maintenance for the clutch system
Aeddula et al. (2024)	Apply AI-driven predictive maintenance in autonomous vehicle fleets as part of a product-service system strategy	AI-driven analytics framework	Predictive maintenance/maintenance planning	The research identified practical considerations, limitations and potential improvements for implementing AI-based maintenance in autonomous vehicle services, demonstrating the approach’s significance in enhancing fleet reliability and informing effective PSS design

Source	Objectives	AI model	Maintenance aspect	Short outcomes
Jakobsson et al. (2020)	Predict the remaining useful life of heavy mining vehicles based on their usage data	NN	Predictive maintenance	The model accurately forecasts vehicle component life using operational data, enabling proactive maintenance scheduling and reducing unexpected heavy vehicle breakdowns
Gong et al. (2023)	Diagnose and monitor electric vehicle faults using indirect measurements combined with AI	AI-driven signal processing	Predictive maintenance	The proposed methodology quickly and accurately detects EV motor and battery faults using only indirect signals, improving maintenance response times and reducing the need for additional sensors
Gugaratshan et al. (2023)	To continuously update reliability, availability, maintainability models of vehicles using in-service sensor data	ML	Maintenance planning	Demonstrated that incorporating real-time multi-variate sensor data into reliability models significantly improves failure prediction and optimizes maintenance timing for critical vehicle subsystems (e.g. brakes, engine)
Chen et al. (2021)	Predict automotive maintenance needs by integrating deep learning models with GIS data	Deep Learning (e.g. CNN/LSTM)	Predictive maintenance	Successfully predicts when and where vehicle maintenance is required by analyzing operational routes and environmental conditions, allowing timely, location-aware maintenance actions for vehicle fleets
Giordano et al. (2022)	Evaluate and derive best practices for implementing predictive maintenance in an automotive production environment	ML	Predictive maintenance	Highlighted effective data-driven maintenance strategies (and common pitfalls) in a real automotive case study, providing guidance on data quality, model selection and integration for future implementations
Giobergia et al. (2018)	Use large volumes of sensor data from vehicles to predict component failures in advance	Data mining and ML	Predictive maintenance	Showed that mining high-dimensional vehicle sensor data can uncover early warning patterns for component failures, thereby improving fault detection and enabling maintenance to be scheduled before breakdowns occur in automotive fleets
Liu et al. (2021)	To diagnose faults in electric/hybrid vehicles using enhanced machine learning algorithms	Improved ML algorithm (e.g. enhanced SVM/ANN)	Predictive maintenance	The improved machine learning model accurately identifies battery and motor anomalies in new energy vehicles, contributing to more reliable operation and timely maintenance of electric vehicle fleets
Kavousi-Fard et al. (2020)	To detect anomalies in vehicle operation data for both security and reliability purposes	DL	Anomaly detection	The model detects abnormal patterns in vehicle data (potentially indicating component faults or cyber-attacks) with high accuracy, providing an early warning system that can trigger maintenance or security interventions promptly
Ji et al. (2021)	Identify anomalies in the clutch operation of hybrid electric vehicles using machine learning	ML	Condition monitoring	Successfully detected unusual clutch engagement patterns that indicate wear or faults in hybrid vehicle transmissions, enabling targeted preventive maintenance for the clutch system
Aeddula et al. (2024)	Apply AI-driven predictive maintenance in autonomous vehicle fleets as part of a product-service system strategy	AI-driven analytics framework	Predictive maintenance/maintenance planning	The research identified practical considerations, limitations and potential improvements for implementing AI-based maintenance in autonomous vehicle services, demonstrating the approach’s significance in enhancing fleet reliability and informing effective PSS design

Note(s): NN: Neural network; AI: artificial intelligence; ML: machine learning; CNN: convolution neural network; LSTM: long short-term memory; GIS: geographic information system; SVM: support vector machine; ANN: artificial neural network; DL: deep learning

[ViewLarge]

Sharing Unavailable