An Investigation About Prediction of Stability Property of Bituminous Hot Mixtures by Regression Models

Yıl 2016, Cilt: 2 Sayı: 2, 77 - 89, 01.06.2016

Deniz Arslan Hikmet Bayırtepe

Öz

In this study, the stability property which is one of the basic characteristics of the bituminous hot mixtures was aimed to be predicted through the physical properties of the mixtures and the materials used. A data set of 4680 observations relating to the briquette and core samples were collected by this aim. Regression, principal component, cluster and discriminant analyses were performed on the data set. Cluster analysis has shown that the data set can be divided into four main clusters. This outcome was also wholly supported by the discriminant analysis. According to the stepwise regression results, the variance of the stability was determined to be explained up to 73.8 % by the independent variables.