Proactive sinkhole risk assessment in urban asphalt pavements using wavelet transform and the international roughness index

This paper proposes a dual method integrating the wavelet transform and the International Roughness Index (IRI) for proactive detection of sinkholes in urban roadways. Longitudinal profiles of asphalt pavements were employed, based on the premise that sinkhole-induced amplitudes and wavelengths are manifested in the profiles. Four sinkholes, with wavelengths from 2 to 20 m, were derived from reported cavity data and engineering inferences and superimposed onto in-service profiles (IRI 0.7 to 4.6 m/km) to simulate sinkhole progression. The discrete wavelet transform was applied to 152.4-m profiles at 0.15-m intervals, decomposing them into seven wavelet levels. Localized increases in wavelet coefficients enabled identification of sinkhole locations, and the method successfully distinguished sinkholes from other pavement distresses with similar wavelengths. Wavelet energy facilitated sinkhole size estimation. The findings confirm IRI as a valuable complementary tool for assessing roadway safety. These findings have potential applications in both pavement management and construction management domains. • Dual method with wavelet transform and IRI proposed for sinkhole detection. • Wavelet coefficients capture roadway sinkhole locations effectively. • Wavelet energy provides quantitative estimation of sinkhole size. • Progressive sinkholes detected even on distressed asphalt pavements. • IRI complements wavelet analysis by quantifying roadway safety.