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

Yue Wang; Areg Karapetyan; Djellel Difallah; Samer Madanat

cs.LGFeb 15, 2026

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

Authors: Yue Wang, Areg Karapetyan, Djellel Difallah, Samer Madanat

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

Figures & tables

Figure 1 : The architecture of the proposed UniST-Pred model.

	SimSF-Bay			PEMS-Bay			NYCTaxi
Method	RMSE	MAE	MAPE (%)	RMSE	MAE	MAPE (%)	RMSE	MAE	MAPE (%)
ARIMA	8.5 (-58%)	4.68 (-49%)	45.8 (-25%)	6.50 (-35%)	3.36 (-43%)	8.34 (-46%)	43.22 (-69%)	15.09 (-69%)	45.3 (-8%)
LSTM	6.4 (-44%)	4.23 (-44%)	53.6 (-36%)	4.96 (-15%)	2.37 (-19%)	5.70 (-21%)	22.30 (-40%)	10.34 (-55%)	122 (-66%)
TCN	5.2 (-31%)	3.40 (-30%)	45.3 (-25%)	5.11 (-18%)	2.76 (-31%)	5.97 (-24%)	15.88 (-16%)	5.45 (-14%)	42.3 (-2%)
TS-Mixer	3.9 (-8%)	2.60 (-9%)	37.1 (-8%)	4.91 (-14%)	2.69 (-29%)	6.06 (-25%)	14.82 (-10%)	5.20 (-10%)	45.0 (-8%)
TS-Mixer-ext	4.0 (-10%)	2.58 (-8%)	37.4 (-9%)	4.90 (-14%)	2.69 (-29%)	6.00 (-25%)	14.82 (-10%)	5.20 (-10%)	45.0 (-8%)

Table 1 : Performance comparison on SimSF-Bay, PEMS-Bay, and NYC-Taxi datasets. The best results are bold and the second best results are underlined . The values in the brackets denote the relative performance gap w.r.t. UniST-Pred, computed as

100\%\cdot\frac{M_{\text{UniST-Pred}}-M_{\text{method}}}{M_{\text{method}}}

, where

M

represents the scores (negative values indicate worse performance).

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.

Method	SimSF-Bay	PEMS-Bay	NYCTaxi
STEP	12,952,470	61,138,261	5,670,166
UniST	3,208,208	16,558,467	167,994
Reduction (%)	75.23	72.91	97.04

Table 2 : Comparison of model parameter counts between STEP and UniST across datasets.

Horizon		TS-Mixer	STEP	UniST-Pred
		TS-Mixer	STEP	UniST-Pred
$H^{\prime}=3$	RMSE	4.14	4.38	3.73
	MAE	2.67	2.84	2.40
	MAPE (%)	38.23	41.28	33.29
$H^{\prime}=6$	RMSE	4.33	4.52	3.89
	MAE	2.96	2.92	2.49

Table 3 : Performance comparison of TS-Mixer, STEP and UniST-Pred on SimSF-Bay across different forecasting horizons.

Component removal
Method	RMSE $\downarrow$	MAE $\downarrow$	MAPE (%) $\downarrow$
UniST-Pred (full)	3.61	2.37	34.2
w/o spatial	6.06	3.43	65.5
w/o temporal	5.64	4.11	49.8
w/o fusion	4.06	2.57	36.2
Component replacement

Table 4 : Ablation results of UniST-Pred on SimSF-Bay.

Figure 4 : Comparison of ground-truth traffic flow and UniST-Pred predictions across the six disruption scenarios, shown for the network-wide average (top row) and two representative roads (middle and bottom rows).

Panel of Figure 5 — Figure 5 : Explainability map for (c) Scenario 1: disconnecting Oakland Bay Bridge, (f) Scenario 4: disconnecting US-101, (b) and (e): all connected network. The zoom-in parts in (a) (b) (c) are the Oakland Bay Bridge and its surrounding area. The zoom-in parts in (d) (e) (f) are US-101 and its surrounding area. Darker color indicates higher contribution to the prediction.

	TS-Mixer	STEP	UniST-Pred
Scenario 1	3.89	3.93	3.61
Scenario 2	3.87	3.95	3.59
Scenario 3	3.87	3.91	3.60
Scenario 4	3.89	3.91	3.59
Scenario 5	3.87	3.91	3.58
Scenario 6	3.88	3.90	3.58

Table 5 : RMSE of UniST-Pred, TSMixer, and STEP across the disruption scenarios and the fully connected network.

Appendix figures & tables2 assets

Supplementary material from the paper’s appendix.

Appendix

Property	PEMS-BAY	NYC Taxi	SimSF-Bay
Traffic variable	Speed	Flow	Flow
Spatial structure	Sensor graph	Fixed zones / grid	Dynamic road network
# of nodes	325	200	7709
# of edges	2369	712	8781
# of time steps	5.2k+	2.6m+	168 (per scenario)
Time interval	5 min	30 min	5 min

Appendix

Table 6 : Statistics of datasets used in this study.

Model Parameters
Item	SimSF-Bay	PEMS-Bay	NYCTaxi
Input length ( $H$ )	9	2016	35
Prediction horizon ( $H^{\prime}$ )	1	12	1
Number of channels ( $C$ )	4	4	4
Number of GT layers ( $L_{g}$ )	2	2	2
GCN in channel	2	2	1

Appendix

Table 7 : Model parameters and implementation details

Explore similar work

Efficient Traffic State Prediction With Dynamic Joint Spatio-Temporal Relation Inference
Apr 10, 2025Zhifeng Hao, Kai Hu, Juncai Zhang +2Accurate Traffic Forecasting Spatial Encoding
School of Cyberspace Security, Hainan University, Haikou, China
GAMMA-Net: Adaptive Long-Horizon Traffic Spatio-Temporal Forecasting Model based on Interleaved Graph Attention and Multi-Axis Mamba
Apr 18, 2026Dongyi He, Yuanquan Gao, Bin Jiang +1Accurate Traffic Forecasting Temporal Graph Neural Networks
School of Artificial Intelligence, Chongqing University of Technology, Chongqing, 400054, China
Unveiling Stochasticity: Universal Multi-modal Probabilistic Modeling for Traffic Forecasting
Apr 17, 2026Weijiang Xiong, Robert Fonod, Nikolas GeroliminisAccurate Traffic Forecasting Traffic
Urban Transport Systems Laboratory (LUTS), EPFL, Lausanne, 1015, Switzerland

Apr 10, 2025, cs.CVJ/K move · Enter open · S save

Efficient Traffic State Prediction With Dynamic Joint Spatio-Temporal Relation Inference

Zhifeng Hao, Kai Hu, Juncai Zhang, Zhidan Zhao+1

School of Cyberspace Security, Hainan University, Haikou, China

Accurate Traffic Forecasting Spatial Encoding Graph Neural Networks

Open paper AskPDF arXiv

Traffic prediction is difficult due to the complex interplay of temporal evolution, spatial interactions, and delayed spatio-temporal propagation over road networks. Existing methods either model spatial and temporal dependencies separately or employ unified spatio-temporal structures, but they often insufficiently characterize how neighboring sensors at historical timestamps influence a target node, while complex joint models may incur high computation. This paper proposes STEI-PCN, an efficient pure convolutional network based on spatio-temporal encoding and relation inference. It first builds a local causal joint spatio-temporal graph to restrict candidate interactions, then uses absolute position and relative distance encodings to infer dynamic edge weights. A single-layer graph convolution with a position-aware gated activation unit captures local joint dependencies, and temporal dilated causal convolutions complement long-range temporal patterns. A multi-view prediction module fuses raw, local propagation, and long-range temporal representations for direct multi-step forecasting. Experiments on PeMS03, PeMS04, PeMS07, PeMS08, and PeMS-Bay under multiple horizons show that STEI-PCN achieves competitive accuracy with moderate parameters and low training and inference costs. Ablation and fluctuation analyses further verify the contributions of the main components and empirically analyze the effects of the training-stage constraints under sharp speed changes. Our code is available at a GitHub link https://github.com/Jessez2/STEI-PCN.

Abstract

Explore similar work

Efficient Traffic State Prediction With Dynamic Joint Spatio-Temporal Relation Inference

GAMMA-Net: Adaptive Long-Horizon Traffic Spatio-Temporal Forecasting Model based on Interleaved Graph Attention and Multi-Axis Mamba

Unveiling Stochasticity: Universal Multi-modal Probabilistic Modeling for Traffic Forecasting