Adelino Ferreira
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Adelino Ferreira has a total of 99 co-authors. Here are the top co-authors by number of shared publications:
- João Santos
University of Twente
12 shared publications - Tiago Tamagusko
University of Coimbra
11 shared publications Francisco Duarte
University of Coimbra
10 shared publicationsGerardo Flintsch
Center for Sustainable Transportation Infrastructure, Virginia Tech Transportation Institute, The Charles Via, Jr. Department of Civil and Environmental Engineering, Virginia Polytechnic Institute and State University, Blacksburg, VA, USA
8 shared publications- Diogo Correia
CFE, Centre for Functional Ecology: Science for People & Planet, Department of Life Sciences, University of Coimbra, Calçada Martim de Freitas, 3030-456 Coimbra, Portugal
6 shared publications - Francisco Duarte
PhD student, Pavement Mechanics Laboratory, Research Center for Territory, Transports and Environment, Department of Civil Engineering, University of Coimbra, Coimbra, Portugal
6 shared publications
Latest Publications
- Open Access
- 29 October 2026
- Published by Springer Nature
- p. 427-435
Identifying Cycling Accident Hotspots Through Spatial-Temporal Network Kernel Density Estimation: Evidence from Madrid
This study applies the Spatiotemporal Network Kernel Density Estimation (ST-NKDE) methodology to the analysis of cycling accidents in Centro of Madrid between 2019 and 2024. Unlike traditional methods, which either disregard the temporal dimension or the spatial constraints of the road network, this approach integrates both factors, allowing for a more accurate representation of accident dynamics. The analysis was carried out in two stages: in the exploratory phase, heat maps identified priority intervention areas, while in the subsequent phase, ST-NKDE provided a detailed representation of accident densities, with bandwidth selection determined through leave-one-out cross-validation. The optimal parameters found were 1,100 m for spatial bandwidth and 80 days for temporal bandwidth, reflecting the dispersion of accidents. The temporal analysis revealed a bimodal distribution of accidents, with lower densities during the winter months. Two high-density hotspots were identified, particularly along Calle de Fuencarral and Calle de Alcalá. The proposed approach highlights the significance of incorporating spatial-temporal dimensions for a more robust analysis of accident risk, providing valuable evidence to support urban planning and sustainable mobility policies, such as traffic calming measures and road safety improvements.
- 29 October 2026
- Published by Springer Nature
- p. 427-435
Open AccessIdentifying Cycling Accident Hotspots Through Spatial-Temporal Network Kernel Density Estimation: Evidence from Madrid
This study applies the Spatiotemporal Network Kernel Density Estimation (ST-NKDE) methodology to the analysis of cycling accidents in Centro of Madrid between 2019 and 2024. Unlike traditional methods, which either disregard the temporal dimension or the spatial constraints of the road network, this approach integrates both factors, allowing for a more accurate representation of accident dynamics. The analysis was carried out in two stages: in the exploratory phase, heat maps identified priority intervention areas, while in the subsequent phase, ST-NKDE provided a detailed representation of accident densities, with bandwidth selection determined through leave-one-out cross-validation. The optimal parameters found were 1,100 m for spatial bandwidth and 80 days for temporal bandwidth, reflecting the dispersion of accidents. The temporal analysis revealed a bimodal distribution of accidents, with lower densities during the winter months. Two high-density hotspots were identified, particularly along Calle de Fuencarral and Calle de Alcalá. The proposed approach highlights the significance of incorporating spatial-temporal dimensions for a more robust analysis of accident risk, providing valuable evidence to support urban planning and sustainable mobility policies, such as traffic calming measures and road safety improvements.
- Open Access
- 29 October 2026
- Published by Springer Nature
- p. 487-495
Network Design Model Applied to Cycling Infrastructure
The urban fabric should be planned inclusively, giving priority to its most vulnerable users: pedestrians and cyclists. Yet, in practice, policymaking continues to privilege motorised transport. Decisions regarding active mobility are often based on limited studies, with little or no quantitative information about real benefits, safety, or utilisation potential. This study addresses this gap by proposing methodologies adapted from Network Design theory, tailored to cycling mobility. Rather than relying on travel time as the sole determinant of the user’s benefit, the models integrate key factors such as safety, comfort, and slope to estimate user benefit and prioritise interventions. Two approaches are developed: the first sequentially tests different possible typologies of interventions with the objective of prioritising greater levels of segregation in the proposed cycling infrastructure, while the second applies road hierarchy to assign solutions more consistent with the functional role of each street through its road hierarchy class. The results show that the first model achieves broader network coverage, while the second ensures safer and technically consistent solutions, though with lower inclusivity. These approaches provide a structured, objective, and adaptable decision-support tool for policymakers, capable of delivering clear proposals, prioritisation strategies, and cost estimates. While not definitive solutions, they represent a step forward in integrating cycling into the core of urban planning decisions and point to pathways for future refinement.
- 29 October 2026
- Published by Springer Nature
- p. 487-495
Open AccessNetwork Design Model Applied to Cycling Infrastructure
The urban fabric should be planned inclusively, giving priority to its most vulnerable users: pedestrians and cyclists. Yet, in practice, policymaking continues to privilege motorised transport. Decisions regarding active mobility are often based on limited studies, with little or no quantitative information about real benefits, safety, or utilisation potential. This study addresses this gap by proposing methodologies adapted from Network Design theory, tailored to cycling mobility. Rather than relying on travel time as the sole determinant of the user’s benefit, the models integrate key factors such as safety, comfort, and slope to estimate user benefit and prioritise interventions. Two approaches are developed: the first sequentially tests different possible typologies of interventions with the objective of prioritising greater levels of segregation in the proposed cycling infrastructure, while the second applies road hierarchy to assign solutions more consistent with the functional role of each street through its road hierarchy class. The results show that the first model achieves broader network coverage, while the second ensures safer and technically consistent solutions, though with lower inclusivity. These approaches provide a structured, objective, and adaptable decision-support tool for policymakers, capable of delivering clear proposals, prioritisation strategies, and cost estimates. While not definitive solutions, they represent a step forward in integrating cycling into the core of urban planning decisions and point to pathways for future refinement.
- Open Access
- 30 May 2026
- Published by Taylor & Francis in Transportation Letters
- p. 1-23
Characterizing the urban road network for automated mobility: a scalable typology for evidence-based policy
The advent of Connected and Automated Vehicles (CAVs) promises to revolutionize transportation, offering substantial improvements in safety, efficiency, and mobility (Anderson et al. 2016; Carp 201...
- 30 May 2026
- Published by Taylor & Francis in Transportation Letters
- p. 1-23
Open AccessCharacterizing the urban road network for automated mobility: a scalable typology for evidence-based policy
The advent of Connected and Automated Vehicles (CAVs) promises to revolutionize transportation, offering substantial improvements in safety, efficiency, and mobility (Anderson et al. 2016; Carp 201...
- Open Access
- 29 May 2026
- Published by Springer Nature in International Journal of Pavement Research and Technology
Optimizing Pavement Maintenance with AI: A Data-Driven Framework Integrating Optimization and Machine Learning
As global infrastructure ages under fiscal constraints, traditional Pavement Management Systems often rely on sequential workflows that ignore the dynamic impact of interventions on future deterioration. This paper introduces a data-driven framework integrating a recursive multi-year forecasting engine with network-level Genetic Algorithm (GA) optimization. Unlike static models, this approach establishes a dynamic feedback loop where maintenance decisions at year t update the pavement state inputs for the prediction model at year $$t+1$$ . The framework couples a Light Gradient Boosting Machine (LightGBM), with hyperparameters tuned via Bayesian optimization (Optuna), trained on Long-Term Pavement Performance (LTPP) data ( $$R^2=0.88$$ , $$RMSE=0.19$$ ) with a GA optimization module for resource allocation. Using a section-based splitting strategy to prevent data leakage, results show this recursive integration outperforms reactive strategies, improving network quality by 3% while reducing maintenance costs by 22%. This framework demonstrates the potential of recursive data-driven models to optimize life-cycle costs and reliability in modern pavement management.- 29 May 2026
- Published by Springer Nature in International Journal of Pavement Research and Technology
Open AccessOptimizing Pavement Maintenance with AI: A Data-Driven Framework Integrating Optimization and Machine Learning
As global infrastructure ages under fiscal constraints, traditional Pavement Management Systems often rely on sequential workflows that ignore the dynamic impact of interventions on future deterioration. This paper introduces a data-driven framework integrating a recursive multi-year forecasting engine with network-level Genetic Algorithm (GA) optimization. Unlike static models, this approach establishes a dynamic feedback loop where maintenance decisions at year t update the pavement state inputs for the prediction model at year $$t+1$$ . The framework couples a Light Gradient Boosting Machine (LightGBM), with hyperparameters tuned via Bayesian optimization (Optuna), trained on Long-Term Pavement Performance (LTPP) data ( $$R^2=0.88$$ , $$RMSE=0.19$$ ) with a GA optimization module for resource allocation. Using a section-based splitting strategy to prevent data leakage, results show this recursive integration outperforms reactive strategies, improving network quality by 3% while reducing maintenance costs by 22%. This framework demonstrates the potential of recursive data-driven models to optimize life-cycle costs and reliability in modern pavement management. - Open Access
- 27 May 2026
- Published by Elsevier in Construction and Building Materials
- Vol. 533
Integrated viscoelastic-statistical analysis of the fatigue-critical temperature in two polymer-modified asphalt mixes
- 27 May 2026
- Published by Elsevier in Construction and Building Materials
- Vol. 533
Open AccessIntegrated viscoelastic-statistical analysis of the fatigue-critical temperature in two polymer-modified asphalt mixes
- Open Access
- 23 May 2026
- Published by Elsevier in Construction and Building Materials
- Vol. 533
Interfacial adhesion analysis between granular-asphalt binder and granular-mastic systems
- 23 May 2026
- Published by Elsevier in Construction and Building Materials
- Vol. 533
Open AccessInterfacial adhesion analysis between granular-asphalt binder and granular-mastic systems
- Open Access
- 15 May 2026
- Published by Elsevier in Case Studies on Transport Policy
- Vol. 25
Bikeable: Estimating cyclists’ perceived safety from street-level imagery to support transport planning in Lisbon and London
- 15 May 2026
- Published by Elsevier in Case Studies on Transport Policy
- Vol. 25
Open AccessBikeable: Estimating cyclists’ perceived safety from street-level imagery to support transport planning in Lisbon and London
- Open Access
- 4 October 2025
- Published by Springer Nature
- p. 153-159
Asphalt Pavement Performance Prediction Using Ensemble Learning Methods
Ensuring road quality is crucial for the safety and comfort of all road users. The International Roughness Index (IRI) is a fundamental index for assessing the performance of flexible pavements. However, many conventional methods to gauge this index often fall short in accuracy and reliability. To address this, the present research harnessed ensemble machine learning methods, incorporating data from 55 asphalt pavement sections from the Long-Term Pavement Performance InfoPave database. The models were trained using pavement structure, age, traffic, and climatic factors to predict IRI values. Results indicate that ensemble learning methods effectively predict pavement performance, recording R2 values ranging from 0.82 to 0.87. Notably, the initial IRI emerged as the most significant predictor. Moreover, models showed heightened precision for pavements displaying IRI values under 2.0 m/km, indicating their aptness for early-stage flexible road pavements. These findings underscore the potential to refine these models further and the necessity for more diversified datasets.
- 4 October 2025
- Published by Springer Nature
- p. 153-159
Open AccessAsphalt Pavement Performance Prediction Using Ensemble Learning Methods
Ensuring road quality is crucial for the safety and comfort of all road users. The International Roughness Index (IRI) is a fundamental index for assessing the performance of flexible pavements. However, many conventional methods to gauge this index often fall short in accuracy and reliability. To address this, the present research harnessed ensemble machine learning methods, incorporating data from 55 asphalt pavement sections from the Long-Term Pavement Performance InfoPave database. The models were trained using pavement structure, age, traffic, and climatic factors to predict IRI values. Results indicate that ensemble learning methods effectively predict pavement performance, recording R2 values ranging from 0.82 to 0.87. Notably, the initial IRI emerged as the most significant predictor. Moreover, models showed heightened precision for pavements displaying IRI values under 2.0 m/km, indicating their aptness for early-stage flexible road pavements. These findings underscore the potential to refine these models further and the necessity for more diversified datasets.
- Open Access
- 19 September 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (18) , 10220
Fatigue Testing in Asphalt Mixes: Emerging Trends and Findings from an Integrated Literature Review
This study compiled a dataset of published works relating to fatigue testing in asphalt mixes, covering 2020–2025. The dataset was subjected to bibliometric and textual analyses, including a systematic review, to explore emerging trends and patterns in experimental protocols. Bibliometrix, VOSviewer, and IRaMuTeQ were employed to map the scientific landscape of 368 articles. Following PRISMA guidelines, the 100 most-cited articles were reviewed to identify prevailing test setups and parameters. The results showed a growing scientific production (9.1% per year), concentrated in a few high-impact journals and dominated by China, with emphasis on sustainability. A comparison between scientific output and a road quality index revealed a disconnect between academic research and field implementation. Five thematic clusters emerged: sustainable pavement management, mechanical characterization, binder modification, performance modeling, and evaluation of innovative materials. Indirect tensile and four-point bending tests were the most common loading modes. Considerable variability in protocols, frequent omissions of methodological details, and limited statistical treatment were also observed. The study highlighted the importance of standardized reporting and robust analysis, offering a reproducible framework to understand fatigue behavior and support future research.
- 19 September 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (18) , 10220
Open AccessFatigue Testing in Asphalt Mixes: Emerging Trends and Findings from an Integrated Literature Review
This study compiled a dataset of published works relating to fatigue testing in asphalt mixes, covering 2020–2025. The dataset was subjected to bibliometric and textual analyses, including a systematic review, to explore emerging trends and patterns in experimental protocols. Bibliometrix, VOSviewer, and IRaMuTeQ were employed to map the scientific landscape of 368 articles. Following PRISMA guidelines, the 100 most-cited articles were reviewed to identify prevailing test setups and parameters. The results showed a growing scientific production (9.1% per year), concentrated in a few high-impact journals and dominated by China, with emphasis on sustainability. A comparison between scientific output and a road quality index revealed a disconnect between academic research and field implementation. Five thematic clusters emerged: sustainable pavement management, mechanical characterization, binder modification, performance modeling, and evaluation of innovative materials. Indirect tensile and four-point bending tests were the most common loading modes. Considerable variability in protocols, frequent omissions of methodological details, and limited statistical treatment were also observed. The study highlighted the importance of standardized reporting and robust analysis, offering a reproducible framework to understand fatigue behavior and support future research.
- Open Access
- 30 May 2025
- Published by MDPI AG in Sustainability
- Vol. 17 (11) , 5047
Correction: Rita et al. Using Deep Learning and Google Street View Imagery to Assess and Improve Cyclist Safety in London. Sustainability 2023, 15, 10270
Regarding the author correction request, it has been clarified that the original paper was based on a master’s project at Imperial College London
- 30 May 2025
- Published by MDPI AG in Sustainability
- Vol. 17 (11) , 5047
Open AccessCorrection: Rita et al. Using Deep Learning and Google Street View Imagery to Assess and Improve Cyclist Safety in London. Sustainability 2023, 15, 10270
Regarding the author correction request, it has been clarified that the original paper was based on a master’s project at Imperial College London