![]() | Only 14 pages are availabe for public view |
Abstract Introduction: International roughness index is one of the criteria used to evaluate pavement condition flexible or rigid as it describes the roughness of pavement and so ride quality.Rigid pavement is considered a suitable option as it has less life cycle cost compared to flexible pavement and can sustain loads heavier compared to flexible pavement. The research problem: Developing a reliable model to predict international roughness index over time as a dependant variable. And then the research aims: Using LTPP data base to develop two models to predict IRI over time and the first is a regression model and the other is using artificial neural network algorithm and compare the results. Steps of study: Collect the data of IRI,distresses and climatic data from LTPP database and casting the missing then proceesing the data to be ready for analysis.Using the processed data a regression analysis is conducted and artificial neural network based model and evaluated and acomplete sensitivity analysis is conducted. The study concludes: Artificial neural network model gives better correlation and coeeficent of determination compared to regression model and also distresses and initial IRI values are significant while predicting IRI over time. |