Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Development of pavement roughness models using Artificial Neural Network (ANN)
Pavement roughness in terms of the International Roughness Index (IRI) plays an important role in determining the riding quality of road networks. Moreover, it is considered as a pavement performance measure and a determinant of the optimal time for pavement Maintenance and Rehabilitation (M & R). Due to the high cost of evaluating pavement roughness using automated devices, it is crucial to find a suitable alternative method with low cost and high accuracy, especially in developing countries. The main objective of this research was to develop pavement roughness models using the Artificial Neural Network (ANN) based on smartphone measurements. The effects of pavement age, traffic loading, and traffic volume on the IRI values were investigated. The results of ANN model development showed that ANN is promising to predict the future IRI with a relatively low average error of less than 10%. M & R alternatives were also suggested based on the present and predicted IRI values. The predicted alternatives using ANN showed a relatively low average error (less than 15%) compared to the actual M & R alternatives. The comparison result between regression and ANN models showed that developed ANN models were more accurate in IRI prediction than the regression models.
Development of pavement roughness models using Artificial Neural Network (ANN)
Pavement roughness in terms of the International Roughness Index (IRI) plays an important role in determining the riding quality of road networks. Moreover, it is considered as a pavement performance measure and a determinant of the optimal time for pavement Maintenance and Rehabilitation (M & R). Due to the high cost of evaluating pavement roughness using automated devices, it is crucial to find a suitable alternative method with low cost and high accuracy, especially in developing countries. The main objective of this research was to develop pavement roughness models using the Artificial Neural Network (ANN) based on smartphone measurements. The effects of pavement age, traffic loading, and traffic volume on the IRI values were investigated. The results of ANN model development showed that ANN is promising to predict the future IRI with a relatively low average error of less than 10%. M & R alternatives were also suggested based on the present and predicted IRI values. The predicted alternatives using ANN showed a relatively low average error (less than 15%) compared to the actual M & R alternatives. The comparison result between regression and ANN models showed that developed ANN models were more accurate in IRI prediction than the regression models.
Development of pavement roughness models using Artificial Neural Network (ANN)
Alatoom, Yazan Ibrahim (Autor:in) / Al-Suleiman (Obaidat), Turki I. (Autor:in)
International Journal of Pavement Engineering ; 23 ; 4622-4637
10.11.2022
16 pages
Aufsatz (Zeitschrift)
Elektronische Ressource
Unbekannt
| DOAJ | 2022
Nondestructive Flexible Pavement Evaluation Using ILLI-PAVE Based Artificial Neural Network Models
| British Library Conference Proceedings | 2006
Pavement Roughness Modeling Using Back-Propagation Neural Networks
| Online Contents | 2004