Search

Predicting faulting failure is useful in the optimal concrete pavement design. In this study, artificial neural networks and the random forest method have been used to predict the amount of this failure. The general prediction model was created by inserting 32 available input variables into artificial neural networks. An integer two objectives optimisation problem was designed to select features that significantly affect the faulting. After applying this method, 19 important variables were identified and used to develop two simplified models based on artificial neural networks and the random forest method. It is shown that the simplified model developed by artificial neural networks is the best model to accurately predict the faulting considering the number of input variables. The cumulative number of days when the precipitation is more than 12.7 mm, the elastic modulus of concrete slab, the number of days passed since the pavement was built, base thickness, the cumulative ESALs in the traffic lane, and the annual average number of days when the temperature is more than 32°C were identified as the most important parameters in predicting faulting using the random forest method. A Sensitivity analysis has been then performed on these variables and optimal values were determined. [ABSTRACT FROM AUTHOR]

Acid sludge is an environmental pollutant that is produced daily in large amounts in used engine oil refineries. Recycling this waste is the primary goal of this paper. This study used 'washing with water' and 'neutralization with quicklime' methods to prepare acid sludge for use in bituminous materials as huge potential disposal. The neutralised sludge by the 'washing with water' method was added to bitumen in the wet blending process at 10, 20, and 30 wt%. This product is brittle without proper adhesion and cohesion and cannot meet the standard limits as an acceptable binder. It performs like long-term aged bitumen, reducing penetration and ductility and increasing the aging potential and viscosity. In another method, the neutralised sludge by adding quicklime was added to the asphalt mixture to replace fine aggregates in 50, 75, and 100 wt%. Using this product decreases the aging potential and moisture susceptibility up to 36.4, and increases rutting resistance up to 34.7%, indirect tensile strength up to 17.3%, and resilient modulus up to 17.3%. Despite the presence of large amounts of heavy metals in acid sludge, using this product in asphalt mixtures is an economical and completely hygienic solution for disposing of acid sludge. [ABSTRACT FROM AUTHOR]

Three-dimensional numerical finite element (FE) biphasic models verified against the experimental data were developed aiming to study the accuracy of FE modelling in the prediction of viscoelastic properties of crumb rubber-modified asphalt binders. In the numerical models, crumb rubber-modified asphalt binders are considered as a heterogeneous composite material consisting of the original asphalt binder and rubber particles which respectively have viscoelastic and hyperelastic behaviours. Three different dosages of crumb rubber including 10, 15 and 20 wt% of the original asphalt are added. To capture rubber particles distribution within the bituminous matrix, a random generation method by using the MOA software (French acronym for random object modeller) was implemented based on rubber's minimum and maximum particle sizes. In order to determine the viscoelastic properties of the original asphalt binder as well as the crumb rubber-modified ones, the dynamic shear rheometer (DSR) was used. The excellent agreement between the numerical FE modelling and experimental results witnessed the capability of biphasic finite element models to predict the visco-hyperelastic behaviour of crumb rubber-modified asphalt binders having an error lower than 9%. [ABSTRACT FROM AUTHOR]