Road stability is a crucial indicator in determining the level of service of land transportation infrastructure. Two commonly used approaches for evaluating road conditions are the International Roughness Index (IRI) and the Pavement Condition Index (PCI). The IRI quantitatively measures road surface roughness based on longitudinal profiles, while the PCI visually assesses surface damage by classifying the type and severity of distress. Although these methods differ methodologically, the relationship between them can be utilized to develop more efficient predictive models of road condition. This study aims to analyze the relationship between IRI and PCI using a multimodel statistical approach, including linear, logarithmic, exponential, and second-order polynomial regression. The data used consist of IRI and PCI values from both the left and right sides of the road segments, obtained through empirical testing. The analysis results indicate a negative correlation between IRI and PCI, with the second-order polynomial model yielding the lowest Mean Squared Error (MSE) compared to other models. These findings suggest that non-linear models are better suited to represent the complex relationship between road roughness and surface deterioration. Therefore, IRI has the potential to serve as an early predictive indicator of pavement condition, supporting data-driven decision-making in road maintenance management.
Saputra et al. (Mon,) studied this question.