Urban Traffic Forecasting with Integrated Travel Time and Data Availability in a Conformal Graph Neural Network Framework
This work addresses traffic forecasting for transportation authorities, offering incremental improvements in accuracy and uncertainty estimation.
The study tackled urban traffic flow prediction by integrating travel times and data availability into a Graph Neural Network framework with adaptive conformal prediction for uncertainty handling, resulting in a 24% improvement in MAE and 8% in RMSE over the next-best model and simulated travel times closely matching real-world data.
Traffic flow prediction is a big challenge for transportation authorities as it helps plan and develop better infrastructure. State-of-the-art models often struggle to consider the data in the best way possible, as well as intrinsic uncertainties and the actual physics of the traffic. In this study, we propose a novel framework to incorporate travel times between stations into a weighted adjacency matrix of a Graph Neural Network (GNN) architecture with information from traffic stations based on their data availability. To handle uncertainty, we utilized the Adaptive Conformal Prediction (ACP) method that adjusts prediction intervals based on real-time validation residuals. To validate our results, we model a microscopic traffic scenario and perform a Monte-Carlo simulation to get a travel time distribution for a Vehicle Under Test (VUT), and this distribution is compared against the real-world data. Experiments show that the proposed model outperformed the next-best model by approximately 24% in MAE and 8% in RMSE and validation showed the simulated travel time closely matches the 95th percentile of the observed travel time value.