Arash Hosseini
Arash Hosseini has a total of 24 co-authors. Here are the top co-authors by number of shared publications:
- Ahmed Faheem
Department of Civil and Environmental Engineering, Khalifa University of Science and Technology, Abu Dhabi, UAE
3 shared publications Aria Fathi
3 shared publicationsH. Titi
Associate Professor, Dept. of Civil and Environmental Engineering, Univ. of Wisconsin-Milwaukee, 3200 N. Cramer St., Milwaukee, WI 53211
2 shared publications- Mahdi Saghafi
Center for Transportation Infrastructure Systems (CTIS), The University of Texas at El Paso, El Paso, USA
2 shared publications Saurav Kumar
2 shared publicationsScot Schwandt
Senior Pavement Engineer, Kiewit Infrastructure Engineers, 9780 Mount Pyramid Ct., Englewood, CO 80112
2 shared publications
Latest Publications
- 1 June 2022
- Published by American Society of Civil Engineers (ASCE) in Journal of Transportation Engineering, Part B: Pavements
- Vol. 148 (2) , 04022025
Deterioration Modeling of Flexible Pavements Based on As-Produced and As-Constructed Properties
The goal of this study is to develop a framework for the life-cycle understanding of flexible pavements. New advancements in data analytics allow for the utilization of pavement life-cycle data (historical, environmental, and structural) to evaluate the effects of material, construction, and loading parameters on the in-service performance of the pavements. In this study, the data were georeferenced to establish a connection between pavement parameters such as construction and production quality factors, traffic loading, material properties, pavement structure, and climate conditions to the long-term performance of flexible pavements. The data used in this paper were sampled from the Wisconsin Department of Transportation (WisDOT). Data were filtered to include pavement sections of comparable traffic load and environmental conditions to avoid potential bias in the analysis. Information on 42 highways with a total length of 260.5 mi was collected and analyzed for this study. Pavement deterioration metamodels were developed on high-resolution data using three machine learning (ML) techniques. For the purpose of construction of the metamodels, ML techniques including decision tree regression (DTR), random forest (RF), and gene-expression programming (GEP) were utilized by using coded subroutines in Python. The outcomes of DTR, RF, and GEP approaches showed promising results in the modeling of pavement performance by considering the effects of mix production quality factors such as air voids of the mixture (VA), individual lots voids in mineral aggregates (VMA), in-place density of asphalt mixture (%Gmm), asphalt content (AC), surface thickness, and age of pavements. This approach provides a basis for comprehensive life-cycle evaluation of the highway network without disrupting the state of practice. It relies on connecting data already being collected by the transportation agencies. The relational connection of such data allows for a pavement management system that is capable of continuously reflecting the pavement network performance on design, control, and maintenance activities.
- 1 June 2022
- Published by American Society of Civil Engineers (ASCE) in Journal of Transportation Engineering, Part B: Pavements
- Vol. 148 (2) , 04022025
Deterioration Modeling of Flexible Pavements Based on As-Produced and As-Constructed Properties
The goal of this study is to develop a framework for the life-cycle understanding of flexible pavements. New advancements in data analytics allow for the utilization of pavement life-cycle data (historical, environmental, and structural) to evaluate the effects of material, construction, and loading parameters on the in-service performance of the pavements. In this study, the data were georeferenced to establish a connection between pavement parameters such as construction and production quality factors, traffic loading, material properties, pavement structure, and climate conditions to the long-term performance of flexible pavements. The data used in this paper were sampled from the Wisconsin Department of Transportation (WisDOT). Data were filtered to include pavement sections of comparable traffic load and environmental conditions to avoid potential bias in the analysis. Information on 42 highways with a total length of 260.5 mi was collected and analyzed for this study. Pavement deterioration metamodels were developed on high-resolution data using three machine learning (ML) techniques. For the purpose of construction of the metamodels, ML techniques including decision tree regression (DTR), random forest (RF), and gene-expression programming (GEP) were utilized by using coded subroutines in Python. The outcomes of DTR, RF, and GEP approaches showed promising results in the modeling of pavement performance by considering the effects of mix production quality factors such as air voids of the mixture (VA), individual lots voids in mineral aggregates (VMA), in-place density of asphalt mixture (%Gmm), asphalt content (AC), surface thickness, and age of pavements. This approach provides a basis for comprehensive life-cycle evaluation of the highway network without disrupting the state of practice. It relies on connecting data already being collected by the transportation agencies. The relational connection of such data allows for a pavement management system that is capable of continuously reflecting the pavement network performance on design, control, and maintenance activities.
- 1 December 2021
- Published by American Society of Civil Engineers (ASCE) in Journal of Transportation Engineering, Part B: Pavements
- Vol. 147 (4) , 04021058
Machine Learning Approach to Predict International Roughness Index Using Long-Term Pavement Performance Data
On-time pavement maintenance and rehabilitation are required to maintain or improve pavement roughness, which is an indicator of pavement performance and road safety. Better prediction of maintenance time can help in budget planning and allocation for highways as well as a better and safer driving experience for drivers. In this research, the International Roughness Index (IRI) for asphalt concrete pavement is predicted based on the 12,637 observations in the Long-Term Pavement Performance (LTPP) data set for 1,390 roads and highways in 50 states of the US and the District of Columbia from 1989 to 2018. To identify the research gaps and to better understand the state-of-the-art research in IRI prediction, a systematic literature review (SLR) has been performed to develop a comprehensive view of machine learning techniques used for IRI prediction. We used a machine learning algorithm that can handle missing data in the LTPP data set. Extreme gradient boosting (XGBoost) was used to predict the IRI. Also, the support vector regression (SVR) and random forest (RF) models were used to compare the results. Our results show that XGBoost provides a better model fit in terms of mean absolute error and coefficient of determination. Moreover, our results show that No.-200-passing, hydraulic conductivity, and equivalent single-axle loads in thousands (KESAL) are the most important factors in predicting the IRI.
- 1 December 2021
- Published by American Society of Civil Engineers (ASCE) in Journal of Transportation Engineering, Part B: Pavements
- Vol. 147 (4) , 04021058
Machine Learning Approach to Predict International Roughness Index Using Long-Term Pavement Performance Data
On-time pavement maintenance and rehabilitation are required to maintain or improve pavement roughness, which is an indicator of pavement performance and road safety. Better prediction of maintenance time can help in budget planning and allocation for highways as well as a better and safer driving experience for drivers. In this research, the International Roughness Index (IRI) for asphalt concrete pavement is predicted based on the 12,637 observations in the Long-Term Pavement Performance (LTPP) data set for 1,390 roads and highways in 50 states of the US and the District of Columbia from 1989 to 2018. To identify the research gaps and to better understand the state-of-the-art research in IRI prediction, a systematic literature review (SLR) has been performed to develop a comprehensive view of machine learning techniques used for IRI prediction. We used a machine learning algorithm that can handle missing data in the LTPP data set. Extreme gradient boosting (XGBoost) was used to predict the IRI. Also, the support vector regression (SVR) and random forest (RF) models were used to compare the results. Our results show that XGBoost provides a better model fit in terms of mean absolute error and coefficient of determination. Moreover, our results show that No.-200-passing, hydraulic conductivity, and equivalent single-axle loads in thousands (KESAL) are the most important factors in predicting the IRI.
- 1 December 2020
- Published by American Society of Civil Engineers (ASCE) in Journal of Materials in Civil Engineering
- Vol. 32 (12) , 04020378
Effects of Quality Control Parameters on In-Service Performance of Flexible Pavements
This study investigates the relation of quality measures of asphalt mixture properties during production and construction to long-term in-service performance of flexible pavements. Data from flexible pavement sections taken from the Long-Term Pavement Performance (LTPP) program along with the pavement sections of 30 Wisconsin projects were compiled in a single relational database to establish a georeferenced architecture that embraces the natural variability within the database. An in-depth analysis was performed on seven selected projects using two modeling programs that are based on statistical regression and mechanistic-empirical methods. The three research approaches allowed for assessing the effects of the variation in quality measures during construction and production: air void (Va); voids in mineral aggregates (VMA); and in-place density. The analysis showed that the deviation from the standard target for in-place density influences fatigue, longitudinal, and transverse cracking, as well as the International Roughness Index (IRI). The study showed the shortcomings of over-averaging pavement characteristics in the modeling of predicted in-service performance. Last, this study showed that implementing the proposed georeferenced relational database architecture within LTPP would increase the impact of this program.
- 1 December 2020
- Published by American Society of Civil Engineers (ASCE) in Journal of Materials in Civil Engineering
- Vol. 32 (12) , 04020378
Effects of Quality Control Parameters on In-Service Performance of Flexible Pavements
This study investigates the relation of quality measures of asphalt mixture properties during production and construction to long-term in-service performance of flexible pavements. Data from flexible pavement sections taken from the Long-Term Pavement Performance (LTPP) program along with the pavement sections of 30 Wisconsin projects were compiled in a single relational database to establish a georeferenced architecture that embraces the natural variability within the database. An in-depth analysis was performed on seven selected projects using two modeling programs that are based on statistical regression and mechanistic-empirical methods. The three research approaches allowed for assessing the effects of the variation in quality measures during construction and production: air void (Va); voids in mineral aggregates (VMA); and in-place density. The analysis showed that the deviation from the standard target for in-place density influences fatigue, longitudinal, and transverse cracking, as well as the International Roughness Index (IRI). The study showed the shortcomings of over-averaging pavement characteristics in the modeling of predicted in-service performance. Last, this study showed that implementing the proposed georeferenced relational database architecture within LTPP would increase the impact of this program.
- Open Access
- 1 October 2019
- Published by Elsevier in Journal of Rock Mechanics and Geotechnical Engineering
- Vol. 11 (5) , 1055-1065
Feasibility of using electrokinetics and nanomaterials to stabilize and improve collapsible soils
- 1 October 2019
- Published by Elsevier in Journal of Rock Mechanics and Geotechnical Engineering
- Vol. 11 (5) , 1055-1065
Open AccessFeasibility of using electrokinetics and nanomaterials to stabilize and improve collapsible soils
- 18 July 2019
- Published by American Society of Civil Engineers (ASCE)
- p. 31-41
Parametric Study of Pavement Deterioration Using Machine Learning Algorithms
- 18 July 2019
- Published by American Society of Civil Engineers (ASCE)
- p. 31-41
Parametric Study of Pavement Deterioration Using Machine Learning Algorithms
- 18 July 2019
- Published by American Society of Civil Engineers (ASCE)
- p. 424-435
Image Processing Method to Estimate the Wearing Condition of Slurry Seal Mixtures
- 18 July 2019
- Published by American Society of Civil Engineers (ASCE)
- p. 424-435
Image Processing Method to Estimate the Wearing Condition of Slurry Seal Mixtures
- 11 June 2019
- Published by Springer Nature in SN Applied Sciences
- Vol. 1 (7) , 703
Estimation of rocking capacity of soil-structure systems using a hybrid inverse solver
- 11 June 2019
- Published by Springer Nature in SN Applied Sciences
- Vol. 1 (7) , 703
Estimation of rocking capacity of soil-structure systems using a hybrid inverse solver
- Open Access
- 4 October 2017
- Published by Ital Publication in Civil Engineering Journal
- Vol. 3 (9) , 690-700
Electrokinetic Remediation of Zinc and Copper Contaminated Soil: A Simulation-based Study
Electrokinetic is an effective and innovative method to remediate different kinds of soils, especially low permeable fined-grain soils such as silty and clayey soils. In this method, by applying a direct-current electric field into a contaminated soil resulted in different transport phenomena, the soil is remediated. This paper's objective is to propose a numerical model for Electrokinetic remediation of zinc and copper contaminated soils. Different transport phenomena including ion migration, electroosmosis flow, and diffusion were taken into account in the model. Chemical reactions such as precipitation/dissolution, adsorption onto the soil surface, and water chemical equilibrium were considered as well. Furthermore, instead of simplified boundary conditions (Neumann or Dirichlet) that cannot properly reflect the reality of the Electrokinetic remediation process, the realistic boundary conditions were used with consideration of flux and electrolysis reaction at the electrodes. The simulation results compared with the available experimental data in the literature. The coefficient of determination and the index of agreement indicated that the present model is consistent with the tests' results. Thus, the assumptions considered in the present study are acceptable.
- 4 October 2017
- Published by Ital Publication in Civil Engineering Journal
- Vol. 3 (9) , 690-700
Open AccessElectrokinetic Remediation of Zinc and Copper Contaminated Soil: A Simulation-based Study
Electrokinetic is an effective and innovative method to remediate different kinds of soils, especially low permeable fined-grain soils such as silty and clayey soils. In this method, by applying a direct-current electric field into a contaminated soil resulted in different transport phenomena, the soil is remediated. This paper's objective is to propose a numerical model for Electrokinetic remediation of zinc and copper contaminated soils. Different transport phenomena including ion migration, electroosmosis flow, and diffusion were taken into account in the model. Chemical reactions such as precipitation/dissolution, adsorption onto the soil surface, and water chemical equilibrium were considered as well. Furthermore, instead of simplified boundary conditions (Neumann or Dirichlet) that cannot properly reflect the reality of the Electrokinetic remediation process, the realistic boundary conditions were used with consideration of flux and electrolysis reaction at the electrodes. The simulation results compared with the available experimental data in the literature. The coefficient of determination and the index of agreement indicated that the present model is consistent with the tests' results. Thus, the assumptions considered in the present study are acceptable.