The Role of Artificial Intelligence in Safety Assessment Across Various Transport Modes

Recent studies in transportation safety put emphasis on the usage of extensive data throughintelligent systems to scale down the accidents among users. Numerous applications of Machine Learning (ML)and Artificial Intelligence (AI) have been created to tackle safety challenges and enhance the efficiency oftransportation systems. Despite some limitations in knowledge exchange across different transport modes, thispaper explores the application of machine learning (ML) and artificial intelligence (AI) techniques in road,rail, maritime, and aviation transport to enhance safety. The study aims to identify best practices andexperiences that can be adapted across these sectors. The methodologies examined include statistical andeconometric techniques, algorithmic strategies, classification and clustering methods, artificial neural networks(ANN), as well as optimization and dimensionality reduction approaches. Findings reveal a growing interestamong transportation researchers and practitioners in leveraging AI for crash prediction, incident and failuredetection, pattern recognition, driver/operator assistance, route optimization, and problem-solving. Among themost used and effective techniques across all transport modes are ANN, Support Vector Machines (SVM),Hidden Markov Models, and Bayesian models. The selection of analytical methods largely depends on thespecific safety assessment objectives. Notably, the road transport sector exhibits the broadest range of AI andML applications and a significantly higher and continuously growing volume of research compared to othertransport modes.