TY  - JOUR
AU  - Liu, Ruping
AU  - Yin, Xiaoqi
AU  - Zhou, Jing
AU  - Ma, Chenxi
AU  - Li, Chen
PY  - 2026
DA  - 2026/09/29
TI  - A Multi-Scale Traffic-State Gated Attention Network for Bus Arrival Time Prediction
JO  - ICCK Transactions on Intelligent Systematics
T2  - ICCK Transactions on Intelligent Systematics
JF  - ICCK Transactions on Intelligent Systematics
VL  - 3
IS  - 3
SP  - 196
EP  - 208
DO  - 10.62762/TIS.2026.906270
UR  - https://www.icck.org/article/abs/TIS.2026.906270
KW  - bus arrival time prediction
KW  - intelligent transportation
KW  - LSTM
KW  - traffic-state gating
KW  - multi-scale temporal modeling
KW  - feature attention
AB  - Accurate bus arrival time prediction supports passenger information services and transit dispatching, but irregular sampling and time-varying traffic conditions make the task strongly nonlinear. This paper proposes a Spatial Attention Multi-Scale Traffic-State Gated Long Short-Term Memory (SA-MS-TG-LSTM) network. Seven trajectory features are constructed from Global Positioning System (GPS) records, including acceleration, congestion index, and speed volatility. Scale-preserving Spatial Feature Attention (SFA) first reweights the feature dimensions, and three Traffic-State Gated Long Short-Term Memory (TG-LSTM) branches subsequently encode the latest 5, 10, and 15 time steps. Their outputs are concatenated and mapped to the remaining travel time. Experiments are conducted on 8,834 GPS records from 293 trips on a single bus route (Route 86). Under a controlled setting in which all recurrent models use a hidden size of 16, one recurrent layer, and at most 50 training epochs, SA-MS-TG-LSTM achieves the best aggregate performance, with a root mean square error (RMSE) of 1.056 min, mean absolute error (MAE) of 0.762 min, mean absolute percentage error (MAPE) of 12.06%, and coefficient of determination ($R^2$) of 0.9631. Compared with the Convolutional Neural Network--Gated Recurrent Unit (CNN-GRU), it reduces RMSE, MAE, and MAPE by 6.30%, 5.57%, and 11.26%, respectively. Trip-level analysis further confirms statistically significant MAE improvements over a standard Long Short-Term Memory (LSTM) model and TG-LSTM. These results demonstrate the effectiveness of the proposed method on the evaluated route, while broader multi-route validation remains an important direction for future work.
SN  - 3068-5079
PB  - Institute of Central Computation and Knowledge
LA  - English
ER  - 

@article{Liu2026A,
  author = {Ruping Liu and Xiaoqi Yin and Jing Zhou and Chenxi Ma and Chen Li},
  title = {A Multi-Scale Traffic-State Gated Attention Network for Bus Arrival Time Prediction},
  journal = {ICCK Transactions on Intelligent Systematics},
  year = {2026},
  volume = {3},
  number = {3},
  pages = {196-208},
  doi = {10.62762/TIS.2026.906270},
  url = {https://www.icck.org/article/abs/TIS.2026.906270},
  abstract = {Accurate bus arrival time prediction supports passenger information services and transit dispatching, but irregular sampling and time-varying traffic conditions make the task strongly nonlinear. This paper proposes a Spatial Attention Multi-Scale Traffic-State Gated Long Short-Term Memory (SA-MS-TG-LSTM) network. Seven trajectory features are constructed from Global Positioning System (GPS) records, including acceleration, congestion index, and speed volatility. Scale-preserving Spatial Feature Attention (SFA) first reweights the feature dimensions, and three Traffic-State Gated Long Short-Term Memory (TG-LSTM) branches subsequently encode the latest 5, 10, and 15 time steps. Their outputs are concatenated and mapped to the remaining travel time. Experiments are conducted on 8,834 GPS records from 293 trips on a single bus route (Route 86). Under a controlled setting in which all recurrent models use a hidden size of 16, one recurrent layer, and at most 50 training epochs, SA-MS-TG-LSTM achieves the best aggregate performance, with a root mean square error (RMSE) of 1.056 min, mean absolute error (MAE) of 0.762 min, mean absolute percentage error (MAPE) of 12.06\%, and coefficient of determination (\$R^2\$) of 0.9631. Compared with the Convolutional Neural Network--Gated Recurrent Unit (CNN-GRU), it reduces RMSE, MAE, and MAPE by 6.30\%, 5.57\%, and 11.26\%, respectively. Trip-level analysis further confirms statistically significant MAE improvements over a standard Long Short-Term Memory (LSTM) model and TG-LSTM. These results demonstrate the effectiveness of the proposed method on the evaluated route, while broader multi-route validation remains an important direction for future work.},
  keywords = {bus arrival time prediction, intelligent transportation, LSTM, traffic-state gating, multi-scale temporal modeling, feature attention},
  issn = {3068-5079},
  publisher = {Institute of Central Computation and Knowledge}
}