Adaptive Neuro-Fuzzy Modeling for Traffic Volume Prediction

Abstract

In the realm of modern urban planning and transportation management, the accurate prediction of traffic volumes has emerged as an indispensable tool for efficient traffic flow and strategic infrastructure development. As cities continue to grow and traffic congestion becomes increasingly complex, the need for precise traffic volume forecasting has become paramount. This research addresses this critical need by leveraging the Adaptive Neuro-Fuzzy Inference System (ANFIS) to model and predict traffic volumes with remarkable accuracy. ANFIS's unique ability to capture intricate patterns within data, particularly in the context of varying vehicle categories and daily fluctuations, makes it an ideal candidate for this task. With a rich dataset spanning 31 working days and encompassing five vehicle categories, including two-wheelers, four-wheelers, heavy vehicles, light vehicles, and other vehicles, this study aims to showcase the potential of ANFIS as a pioneering solution for enhanced traffic prediction accuracy. It is observed that developed model has 89.91% accuracy level. By fusing advanced machine learning techniques with real-world traffic data, this research contributes to the advancement of transportation planning and management, ultimately leading to more optimized traffic systems and sustainable urban development.