Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

UniST-Pred: A Robust Unified Framework for Spatio-Temporal Traffic Forecasting in Transportation Networks Under Disruptions

Yue Wang, Areg Karapetyan, Djellel Difallah, Samer Madanat

TL;DR

UniST-Pred is proposed, a unified spatio-temporal forecasting framework that first decouples temporal modeling from spatial representation learning, then integrates both through adaptive representation-level fusion, demonstrating its competitive performance against existing well-established models despite a lightweight design.

Abstract

Spatio-temporal traffic forecasting is a core component of intelligent transportation systems, supporting various downstream tasks such as signal control and network-level traffic management. In real-world deployments, forecasting models must operate under structural and observational uncertainties, conditions that are rarely considered in model design. Recent approaches achieve strong short-term predictive performance by tightly coupling spatial and temporal modeling, often at the cost of increased complexity and limited modularity. In contrast, efficient time-series models capture long-range temporal dependencies without relying on explicit network structure. We propose UniST-Pred, a unified spatio-temporal forecasting framework that first decouples temporal modeling from spatial representation learning, then integrates both through adaptive representation-level fusion. To assess robustness of the proposed approach, we construct a dataset based on an agent-based, microscopic traffic simulator (MATSim) and evaluate UniST-Pred under severe network disconnection scenarios. Additionally, we benchmark UniST-Pred on standard traffic prediction datasets, demonstrating its competitive performance against existing well-established models despite a lightweight design. The results illustrate that UniST-Pred maintains strong predictive performance across both real-world and simulated datasets, while also yielding interpretable spatio-temporal representations under infrastructure disruptions. The source code and the generated dataset are available at https://anonymous.4open.science/r/UniST-Pred-EF27

UniST-Pred: A Robust Unified Framework for Spatio-Temporal Traffic Forecasting in Transportation Networks Under Disruptions

TL;DR

UniST-Pred is proposed, a unified spatio-temporal forecasting framework that first decouples temporal modeling from spatial representation learning, then integrates both through adaptive representation-level fusion, demonstrating its competitive performance against existing well-established models despite a lightweight design.

Abstract

Spatio-temporal traffic forecasting is a core component of intelligent transportation systems, supporting various downstream tasks such as signal control and network-level traffic management. In real-world deployments, forecasting models must operate under structural and observational uncertainties, conditions that are rarely considered in model design. Recent approaches achieve strong short-term predictive performance by tightly coupling spatial and temporal modeling, often at the cost of increased complexity and limited modularity. In contrast, efficient time-series models capture long-range temporal dependencies without relying on explicit network structure. We propose UniST-Pred, a unified spatio-temporal forecasting framework that first decouples temporal modeling from spatial representation learning, then integrates both through adaptive representation-level fusion. To assess robustness of the proposed approach, we construct a dataset based on an agent-based, microscopic traffic simulator (MATSim) and evaluate UniST-Pred under severe network disconnection scenarios. Additionally, we benchmark UniST-Pred on standard traffic prediction datasets, demonstrating its competitive performance against existing well-established models despite a lightweight design. The results illustrate that UniST-Pred maintains strong predictive performance across both real-world and simulated datasets, while also yielding interpretable spatio-temporal representations under infrastructure disruptions. The source code and the generated dataset are available at https://anonymous.4open.science/r/UniST-Pred-EF27
Paper Structure (33 sections, 14 equations, 5 figures, 7 tables)

This paper contains 33 sections, 14 equations, 5 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Real-World Value and Collaboration.
  3. Related Work
  4. Spatio-Temporal Traffic Forecasting with Coupled Architectures
  5. Decoupled Perspectives: Efficient Temporal Modeling and Task-Adaptive Spatial Encoding
  6. Preliminaries & Problem Statement
  7. Methodology
  8. Temporal Modeling via Feature and Time Mixing:
  9. (a) Mixer operators.
  10. (b) Stacked temporal operator.
  11. Spatial Representation Learning with Graph-Based Encoding:
  12. (a) Relation composition (task-adaptive graph generation).
  13. (b) Graph propagation (per-channel GCN).
  14. Adaptive Fusion via Squeeze-Excitation Residual Integration:
  15. Final prediction:
  16. ...and 18 more sections

Figures (5)

  • Figure 1: The architecture of the proposed UniST-Pred model.
  • Figure 2: Per-location MAPE distribution for UniST-Pred, STEP, and TSMixer on three datasets, stratified by high and low traffic flow. (Light colors indicate lower error.)
  • Figure 3: Qualitative comparison of ground truth and model predictions.
  • Figure 4: Explainability map for (b) all connected network and (c) Scenario 1 (disconnecting Oakland Bay Bridge). The zoom-in part is the Oakland Bay Bridge and its surrounding area. Darker color indicates higher contribution to the prediction.
  • Figure 5: Ground truth and UniST-Pred traffic flow prediction under three bridge removal scenarios, shown for the overall network average (top row) and two representative roads (middle and bottom).