Roughness and condition prediction models for airfield pavements using digital image processing

Based on the assessment of three runways at mid-size Chilean airports, the aim of this article is twofold. First, it discusses the potential empirical relationship that may exist between two parameters traditionally used to measure pavement roughness and condition, i.e. the International Roughness Index (IRI) and Pavement Condition Index (PCI), both measured using industry-standard, semi-automated processes. Second, it proposes an automated and parsimonious methodology for estimating the IRI and PCI, based on a digital image processing algorithm that accurately determines the total amount of cracking in pavements and which essentially requires no human curation. On one hand, a direct correlation between IRI and PCI has been obtained that can be considered statistically significant. On the other hand, highly reasonable estimates for the IRI and PCI can be achieved based solely on the total cracking percentage of a unit sample. The models proposed can be used for airport pavement management purposes.