<p>Road traffic management is crucial in shaping urban mobility using sophisticated technologies and data analytics to enhance road traffic movement, security, and environmental sustainability. ML, DL, and FL approaches are subsets of AI technologies that play a significant role in urban traffic regulation. In road traffic regulation, within the context of traffic patterns, outliers or anomalies are the data points or observations that diverge markedly from the rest of the data collection. These outliers can arise from various factors and have different implications for traffic analysis, prediction, and control. This survey discusses the significant role that ML, DL, and FL play in outlier detection to predict traffic patterns. This comprehensive review not only outlines the current applications but also contributes to understanding and developing future perspectives in road traffic management. This survey provides a detailed examination of these techniques in road traffic management, outlining the methodologies employed in the reviewed studies. It also covers a qualitative assessment of the role of outliers in managing road traffic. This study extends previous work by detailing how outlier detection techniques are utilized within road traffic management. It provides a comparative analysis of outlier detection methods based on various parameters, a perspective not thoroughly explored in earlier studies. Furthermore, this review investigates the impact of distance metrics on the effectiveness of outlier detection in traffic analysis, an area that previous surveys have not adequately covered. The survey offers new insights into optimizing traffic management systems through advanced analytical techniques by addressing these gaps.</p>