Investigation Into Collection Variability of Surface Crack Data for Network-Level Asphalt Pavement Evaluation

Abstract

Surface cracking is a major type of pavement distress that is of interest to pavement engineers. Cracks are generally categorized in relation to patterns, orientations, and locations. Each type of crack is associated with one or more failure modes of pavements. Accurately detecting and rating surface cracks is crucial in pavement condition surveying. Currently, many State Highway Agencies employ automated survey methods to collect pavement condition data at network level. As network-level pavement evaluation is based on a single run, the traditional way of characterizing data variation based on multiple runs is no longer valid. Therefore, it is necessary to introduce a new method to evaluate the variability of pavement condition data at the network level. In this study, the variability of network-level surface crack data was evaluated by means of network-level sample parallel tests. The parallel test was conducted from 2018 to 2020 using two vehicles equipped with identical automated survey systems which consisted of a 3-D imaging system and automated distress identification system. A matrix-based method was proposed to evaluate the variations of crack data obtained from two testing vehicles. Crack data investigated in this study included fatigue cracks, longitudinal wheel-path and non-wheel-path cracks, and transverse cracks. Results indicated that change of testing speed could potentially influence the variation of automated crack data. The variations between severity levels for fatigue cracking were higher than other types of cracks. The variation of crack data decreased with the increase of reporting intervals.