Development of a prior model to predict the cracking performance of asphalt mixture in general for asphalt material selection and mix design.

Zhang, Runhua; Chen, Hui; Guo, Hongzhen; Zhou, Mengmeng; Jiandong, Huang

The main objective of this proposed study is to explore the correlations between the performance parameters measured from the state-of-practice mixture laboratory cracking tests and develop a prior model to predict the mixture cracking performance using the identified significant mix design variables for asphalt material selection and mix design stages. Total 20 plant-produced asphalt mixtures with varying design information were evaluated and the disparate mixture cracking performance tests were performed on these mixtures to characterise their performance. The analysis results indicate that the Flexibility Index (FI) parameter shows the overall good correlations with other cracking indices and thus is selected as the index to modelling mixture cracking performance in general. The stepwise regression model was employed to identify the significant mix design variables which were then used for the development of the prediction models to predict the mixture cracking performance in the form of FI parameter. The developed models can be used as a prior tool for material selection and mix design procedure, eliminating the time and effort for sample preparations and running the tests.

REGRESSION analysis; PREDICTION models; INFORMATION design

International Journal of Pavement Engineering, 2023, Vol 24, Issue 1, p1

1029-8436

Academic Journal

10.1080/10298436.2023.2251080