An Agentic Approach for Active Data Collection, Travel Behavior Modeling, and Weather-Sensitive Demand Prediction
Narges Ahmadi, Yubo Jiao, Jônatas Augusto Manzolli, Jiangbo Yu, Luis Miranda-Moreno
A three-part workflow combining a chatbot-administered survey, machine-learning data processing, and LLM-based forecasting to predict weather-sensitive transportation mode choice. Across 454 student observations under five weather conditions, the best vision-augmented LLM configuration reached 71.5% five-class accuracy, showing how multimodal context can strengthen travel demand forecasting.