Using a multi-phase model to predict flushing of sprayed seal pavements.

Flushing is a pavement defect that has a negative effect on the structural integrity and performance of a pavement surface. The aim of this study was to develop a model to predict the occurrence and progression of flushing on sprayed seal pavements. Data analysis and model development were performed using data sourced from New Zealand's Long-Term Pavement Performance programme. The developed model consists of two phases to model (1) the probability of flushing initiating on a pavement and (2) the progression of flushing. The probability of flushing initiation was modelled using a logistic model format, and the progression of flushing was modelled using a linear model format. Testing of the developed model revealed that the logistic model predicted the probability of flushing initiation at 76% accuracy, whereas the linear model had statistically robust predictions of flushing. The developed model is recommended for use in sprayed seal pavement management processes. [ABSTRACT FROM AUTHOR]