Evaluating the Effect of Climate Change in Pavement Performance Modeling Using Artificial Neural Network Approach: Fid-Bau Portal

Evaluating the Effect of Climate Change in Pavement Performance Modeling Using Artificial Neural Network Approach

Salma, S. / Hakan, Y. / Rulian, B. / Jacob, N.

Climate change is expected to affect the performance of highway pavements. However, the literature review to date does not consider how the pavement performance will be affected by extreme weather (i.e., extreme temperature and extreme rainfall). The quantification of pavement deterioration due to extreme weather can be beneficial to evaluate the effect of climate change on pavement performance. This study aims to develop a pavement deterioration model for jointed plain concrete pavement (JPCP) using the artificial neural network (ANN) modeling techniques. The models will be developed using data collected from the Long Term Pavement Performance (LTPP) database for wet, freeze climatic region. The input variables are initial pavement condition (i.e., initial IRI), pavement structural and mechanical properties (i.e., age, concrete pavement thickness, base/subbase thickness, average contraction spacing, and base/subbase materials type), traffic [Cumulative ESAL (CESAL)], and climate attributes (i.e., average annual air temperature, total annual precipitation, annual freezing index, and annual freeze-thaw), and the output is pavement deterioration indicator [international roughness index (IRI)]. The effect of different climate scenarios can be examined by conducting several models with inputs that reflect several climate scenarios. The best model will be used to investigate how much more pavement will deteriorate if climate attributes change to extreme conditions. This research will address a few gaps in the literature, including the scarcity of studies on long-term IRI prediction using LTPP data and studies on the effect of climate attributes in pavement deterioration. This analysis can be beneficial to the long-term policymaking in road infrastructure.