Aug 2026· Sustainability· Vol 18, pp. 8077· 0 citations· 49 references
TL;DR
A multi-source machine learning framework for segment-direction-level prediction in the Denizli city center showed that temporal and traffic-state variables dominate predictions, while weather and public transport provide complementary value.
Abstract
Accurate short-term travel time prediction is foundational for sustainable urban mobility and intelligent transportation systems on urban arterial corridors, where travel conditions are shaped by interacting traffic, weather, and public transport factors. This study proposes a multi-source machine learning framework for segment-direction-level prediction in the Denizli city center. Floating car data (FCD), Traffic Control Center (TCC) inductive loop detector measurements, historical weather, and public transport indicators were integrated into a 15 min time-segment structure. The final dataset includes 60 segment-direction targets. Performance was evaluated using Linear Regression, Random Forest, LightGBM, and LSTM under a chronological train-validation-test design. Tree-based ensemble models produced the most stable overall performance, with LightGBM and Random Forest yielding similarly low pooled test errors. Segment-level analyses revealed clear spatial and temporal heterogeneity, showing no single model is universally superior across all links. By providing reliable traffic-state information, the framework enables efficient traffic management and may indirectly reduce delay, fuel use, and emissions; these environmental effects were not quantified. SHAP-based interpretation showed that temporal and traffic-state variables dominate predictions, while weather and public transport provide complementary value.
This study examines the prediction of bus travel times within urban corridors, using an extensive dataset from transit management databases and on-street sensors. The analysis focuses on a range of variables, including corridor configuration, general traffic conditions, and intrinsic bus transit factors. Employing Multiple Linear Regression (MLR), Artificial Neural Networks (ANN), Support Vector Regression (SVR), and Random Forest (RF), bus travel times are modelled across four distinct urban corridors in A Coruña, Spain, considering dynamic variables like traffic flow rate, general travel time, and average stream patronage per bus stop, with patronage showing the strongest influence on bus travel time in three of the four corridors, similar to the sum of all general traffic variables. Furthermore, these models are applied to a joint dataset encompassing all corridors, incorporating static variables such as bus stops per kilometre, and, for the first time in the field, the percentage of corridors with adjacent parking and percentage with one lane, both considered as an interaction. This allows predictions of bus travel time changes due to corridor modifications and travel times for new bus routes in unserved areas. Findings reveal that, for our dataset, none of the three ML approaches has consistently proven to be preferable in bus travel time predictions, while MLR provides competitive results, balancing accuracy and interpretability despite its flexibility constraints. The study underscores the importance of selecting models based on data range and transit context, advocating for simplicity in constrained scenarios. This developed methodology provides a valuable planning tool for transport agencies, adaptable to other urban contexts, and highlights the benefits of optimising corridor configurations to enhance bus travel time performance.
Yaiza Montero-Lamas, A. Fernandez, M. Gestal et al.· European Transport Research...· 0 citations
Although the majority of Intelligent Transportation Systems (ITS) and risk-prediction frameworks keep considering pavement quality and dynamic traffic behavior as separate phenomena, road infrastructure degradation and traffic flow instability both contribute to risky driving situations. This study combines segment-level Pavement Condition Index (PCI) data with actual traffic observations from New York City to present an integrated, data-driven methodology for simulating infrastructure-induced unsafe driving circumstances. A supervised machine learning model is developed by combining measures of traffic congestion, speed variation, and pavement deterioration to estimate hourly instability risk, which serves as a proxy for risky driving behavior. Evaluating on a temporally separated test set, the infrastructure-aware model achieves an ROC-AUC of 0.9804 and a PR-AUC of 0.9074, outperforming a traffic-only baseline (ROC-AUC 0.8785, PR-AUC 0.5833). These instability indicators correspond to high-level behavioral patterns commonly observed in ITS monitoring contexts, without relying on raw visual data. Model explainability using SHAP indicates that pavement condition and congestion are the most influential features, with comparable contributions to instability prediction. Predicted risk probabilities are geospatially mapped to identify infrastructure-driven hotspots, and the ORQCIAM framework demonstrates how such risk outputs can inform infrastructure-aware routing and maintenance prioritization. The findings reveal that machine learning enhanced with pavement condition data offers a data-driven approach for predicting hazardous driving situations and supporting infrastructure-aware decision-making, demonstrating how infrastructure-aware risk estimates might help with routing analysis and repair priority in future ITS applications.
Urban traffic flow forecasting is essential for intelligent transportation management,
particularly in data-constrained urban environments. This study presents a reproducible
benchmark framework for hourly traffic flow prediction using the Metro Interstate Traffic
Volume dataset, comprising 48,204 observations collected between 2012 and 2018 with
weather and holiday attributes. To ensure methodological rigor, classical statistical
approaches (Seasonal Naïve, AutoReg, and ARIMA/SARIMA) are compared with machine
learning models, namely Random Forest and Long Short-Term Memory. A time-seriesaware validation strategy with chronological data partitioning is employed to prevent
information leakage. Temporal, lagged, and rolling statistical features are generated
exclusively from historical observations. Model performance is evaluated using Mean
Absolute Error, Root Mean Square Error, Mean Absolute Percentage Error, and the
coefficient of determination (R²). The study provides a transparent and reproducible
comparison of forecasting approaches and discusses their applicability to developing urban
transportation systems.
N. Mammadov, Azar Huseyn, E. Abasov et al.· Problems of Information Tech...· 0 citations
With the increasing complexity of urban traffic systems, reliable forecasting of road conditions has become an important requirement for congestion management and intelligent transportation applications. This study develops a multivariate and multi-horizon traffic speed forecasting framework using real-world traffic monitoring data collected from 75 street segments in Shenzhen, China. LSTM and PatchTST are selected as representative recurrent and Transformer-based forecasting models for comparative evaluation. Average travel speed is considered the prediction target, while traffic index, total sample travel length, and total sample travel time are incorporated as additional input variables. Historical observations from the previous 48 hours are used to forecast traffic speeds over three future horizons, including 30, 60, and 120 minutes. To increase the dependability of experimental outcomes, the dataset is split chronologically, and several random seeds are used. MAE, RMSE, MAPE, and R2 are used to assess the model's performance. According to the experimental findings, PatchTST consistently outperforms LSTM in predicting across all prediction horizons. The MAE values of PatchTST are 2.0710, 2.1960, and 2.3552 for 30-, 60-, and 120-minute forecasting tasks, corresponding to reductions of 6.81%, 7.86%, and 5.89% compared with LSTM. Meanwhile, PatchTST obtains R² values of 0.8886, 0.8795, and 0.8670, respectively. Although the forecasting accuracy of both models decreases with longer prediction horizons, PatchTST maintains relatively lower errors and better stability. The results indicate that the patch-based representation and attention mechanism provide advantages in capturing both local variations and long-term temporal dependencies in urban traffic sequences.
Wentao Hu, Er Zhou· Applied and Computational En...· 0 citations
Tourism-heavy urban mobility nodes often experience short, intense congestion spikes where visitor surges interact
with commuter-related traffic. Despite growing interest in smart mobility management, few studies integrate real-time
computer vision sensing with uncertainty-aware machine learning to support operational decision-making at tourism-intensive
urban nodes. This study aims to develop and validate a real-time, uncertainty-aware traffic prediction framework that
provides actionable, short-horizon (15-minute) congestion intelligence in such areas. Town Quay (Southampton, UK) is used
as a case study. Three months of continuous webcam footage (July–September 2025) were processed with a YOLO + SORT
pipeline to generate 15-minute vehicle counts. Counts were enriched with contextual features capturing weather
(clear/foggy/rainy), day–night cycle, and day type (business vs. leisure). CatBoost and LightGBM were trained using a
chronological split (first two months for training, third month for testing). Uncertainty quantification combined conformal
prediction intervals with Gaussian Mixture Model (GMM). Traffic volumes varied systematically with context, with the
highest average loads occurring during clear-weather daytime business periods. The framework achieved MAE = 43.62, R² =
0.902, and CCC = 0.94. These findings demonstrate that integrating computer-vision traffic sensing with contextual feature
engineering and uncertainty-aware learning yields reliable, interpretable 15-minute congestion forecasts. This enables
proactive, risk-informed interventions (e.g., port-arrival coordination, adaptive traffic control) that help protect the physical
carrying capacity and heritage-setting quality of tourism-intensive areas while maintaining everyday urban mobility.
Enas Elshebli, F. Erdós, aradarajan Vijayakumar et al.· Geo Journal of Tourism and G...· 0 citations
An integrated prediction-and-visualisation pipeline that transforms complex data distributions into actionable visual analytics, such as interpretable station-to-station demand heatmaps via interactive GIS Folium layers is implemented, providing an operationally robust framework to support smart-city transportation management and build more sustainable urban transit systems.
Berna Çalışkan· Journal of Data Analytics an...· 0 citations