Hierarchical Flow Decomposition for Turning Movement Prediction at Signalized Intersections

2026-04-10

AI Summary

This study addresses the challenge of accurately predicting turning movements at signalized intersections, where directional traffic flows exhibit high volatility. To this end, the authors propose the HFD-TM framework, which introduces a novel hierarchical flow decomposition architecture: it first forecasts more stable and interpretable through-movements along traffic corridors and subsequently disaggregates them into individual turning flows. A physics-informed loss function enforcing flow conservation is incorporated to maintain physical consistency. By integrating deep learning with physical constraints, the method achieves a mean absolute error (MAE) of 2.49 vehicles per 15-minute interval on a six-intersection LiDAR dataset in Nashville—outperforming Transformer and GRU baselines by 5.7% and 27.0%, respectively—and trains 12.8 times faster than DCRNN, significantly enhancing both prediction accuracy and computational efficiency.

Technology Category

• Planning, Routing, and Scheduling: Model-Based Reasoning
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Machine Learning: Time-Series/Data Streams

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Accurate prediction of intersection turning movements is essential for adaptive signal control but remains difficult due to the high volatility of directional flows. This study proposes HFD-TM (Hierarchical Flow-Decomposition for Turning Movement Prediction), a hierarchical deep learning framework that predicts turning movements by first forecasting corridor through-movements and then expanding these predictions to individual turning streams. This design is motivated by empirical traffic structure, where corridor flows account for 65.1% of total volume, exhibit lower volatility than turning movements, and explain 35.5% of turning-movement variance. A physics-informed loss function enforces flow conservation to maintain structural consistency. Evaluated on six months of 15-minute interval LiDAR (Light Detection and Ranging) data from a six-intersection corridor in Nashville, Tennessee, HFD-TM achieves a mean absolute error of 2.49 vehicles per interval, reducing MAE by 5.7% compared to a Transformer and by 27.0% compared to a GRU (Gated Recurrent Unit). Ablation results show that hierarchical decomposition provides the largest performance gain, while training time is 12.8 times lower than DCRNN (Diffusion Convolutional Recurrent Neural Network), demonstrating suitability for real-time traffic applications.

Problem

• turning movement prediction
• signalized intersections
• traffic flow volatility
• adaptive signal control

Innovation

• hierarchical flow decomposition
• turning movement prediction
• physics-informed loss
• real-time traffic control
• LiDAR traffic data