ADAPTIVE DEEP-LEARNING FRAMEWORK FOR REAL-TIME HOURLY BUS BOARDING DEMAND CLASSIFICATION IN URBAN TRANSIT

Public transportation systems face significant fluctuations in passenger demand, with over 60% of daily boardings occurring during peak hours and nearly 30% of commuters experiencing delays due to overcrowding. In densely networked urban transit systems, about 70% of route optimization efforts fail without accurate boarding predictions. Traditional manual demand estimation techniques such as timeslot passenger counts, field route surveys, and ticket ledger analysis are often hindered by human error, time-consuming procedures, and limited scalability, resulting in inconsistent data and ineffective planning. To address these challenges, this study introduces an AI-driven boarding demand prediction model that utilizes Support Vector Machine (SVM) and Deep Neural Network (DNN) classifiers. The system integrates real-time data inputs, including timestamp, route number, traffic delays, and historical boarding volumes, to train and evaluate the models using accuracy-based metrics for optimal prediction performance. Furthermore, the AI-based results are benchmarked against three conventional analysis methods—time-slot monitoring, on-site route surveys, and ticket data review to ensure result reliability and model robustness. The proposed solution delivers a high-accuracy, real-time prediction framework that supports dynamic scheduling and minimizes the mismatch between supply and demand. This integration of machine learning into public transport planning offers a scalable, data-centric alternative to manual approaches, paving the way for smarter, more efficient transit systems.