Development of Asphalt Pavement Temperature Models for Mediterranean Climate Condition

Abstract

This paper examines the models for predicting pavement temperatures at specific depths
and formulates new models for the four seasons by using Multiple Linear Regression (MLR) to
predict pavement temperature by using specific depths, time, and air temperature as the
independent variables. The dataset for this study contains 7200 measured pavement
temperatures. Thermal instruments were used to measure asphalt pavement temperature and
other variables every two hours during the four seasons of the year in an attempt to model
pavement temperature by utilising MLR. The prediction aims to establish a pavement
temperatures model for the four seasons. Furthermore, the regression square (R2) values
predicted by the MLR model are 0.84, 0.83, 0.84 and 0.92 for the summer, winter, spring, and
autumn, respectively. All the models presented in this study have a significantly high coefficient
of correlation. The models were validated using the data collected in the Gaza strip for the period
from March 2012 to February 2013, and the results are satisfactory. Therefore, the resulting
models can be used to predict asphalt temperature at varying depths and time.