Searching for Precision in Pavement Evaluation: A Comparative Analysis of Smartphones for Measuring the International Roughness Index (IRI)

Abstract

This study assesses the feasibility of using smartphones to measure the International Roughness Index (IRI) by comparing them with the Roughometer III instrument on a section of the Loja-Malacatos highway. Data on IRI were collected using a mobile app on 51 smartphones mounted in 51 different vehicles, which were then compared with measurements from the Roughometer III to evaluate accuracy and reliability. The results indicate that although the smartphones did not achieve the precision of the specialized device, with a maximum R^2 of 0.334 (RMSE = 1.6 m/km) certain models showed potential for making quick and cost-effective IRI estimates. The study revealed significant variations in accuracy among the devices, influenced by the type and condition of the vehicle, the technical specifications of the smartphone, and the environmental conditions during data collection. The mean absolute error (MAE) and the root mean square error (RMSE) were calculated for each smartphone, identifying those whose measurements deviated least from those obtained by the Roughometer III. Despite the observed limitations, the results suggest a moderate potential of smartphones for estimating IRI, particularly in contexts where specialized equipment is unavailable. The research emphasizes the need to optimize calibration methodologies and to develop specific software that improves the accuracy of data collected with these devices. In conclusion, smartphones represent an accessible and promising tool for the preliminary evaluation of pavement roughness, enabling more frequent and economical monitoring of road infrastructure and allowing greater community involvement in pavement quality management.