Leveraging Smartphone Sensors and GIS for Pavement Condition Assessment: A Case Study of Roads in Kathmandu Valley, Nepal

Abstract

Rapid urbanization and inadequate maintenance have caused significant deterioration of asphalt concrete roads across Nepal, particularly in the Kathmandu Valley. Conventional road condition assessment techniques such as bump integrators and laser profilers are accurate but often expensive, time consuming and labor-intensive, making them impractical for frequent pavement evaluations under budget constraints. This study evaluates the road condition of selected road sections in the Kathmandu Valley using smartphone-based International Roughness Index (IRI) measurements obtained through the RoadRoid application and estimates the corresponding Pavement Condition Index (PCI). Six selected road sections across the Kathmandu valley covering a total of 52.5 km were surveyed. The geo-referenced IRI data collected from RoadRoid were categorized into standard pavement condition classes (Good, Acceptable, Poor, and Very Poor), and visualized in spatial maps prepared using Quantum Geographic Information System (QGIS) software. PCI values were estimated through a regression model developed in a prior study to compare traditional and smartphone-based assessments. Spatial visualization confirmed strong alignment between IRI-based and PCI-based maps, demonstrating that higher IRI values correspond to lower PCI scores in most of the sections. Moreover, Lane-wise correlation analysis was performed between selected road sections which indicated variability in roughness between lanes and suggested the need for separate lane evaluation. Overall, this study validates that the smartphone-based IRI measurements integrated with Geographic Information Systems (GIS) can be a reliable and affordable approach for pavement status assessment and maintenance prioritization especially in a resource constrained country like Nepal.