INFLUENCE OF PAVEMENT ROUGHNESS ON DEFLECTIONS MEASURED BY THE TRAFFIC SPEED DEFLECTOMETER DEVICE

Abstract: Structural evaluation based on deflections is a fundamental methodology for determining the structural performance of pavements in modern pavement management systems. By using techniques such as surface deflection analysis, it is possible to gain a detailed understanding of the pavement's structural response under dynamic loads. The Traffic Speed Deflectometer device (TSDd) stands out in this context as a device that allows for continuous measurement of pavement deflection, offering several benefits compared to traditional methods. However, since it operates at traffic speed, the data obtained by the TSDd may be influenced by pavement roughness due to dynamic loading. The objective of this study was to evaluate the influence of pavement roughness, in terms of the International Roughness Index (IRI), on the deflection data obtained by the TSDd. Deflection data from the Falling Weight Deflectometer (FWD) and TSDd, as well as the IRI, were collected on a Brazilian flexible pavement highway. Simple and multiple linear regression statistical methods were applied in two stages: a global data analysis and a segmented analysis by IRI level. In the global analysis, it was possible to observe the impact of predictor variables on the models and compare the predicted values of both methods. The comparison result showed a strong correlation indicating similar behavior between the models. Additionally, the p-value for the IRI variable in the model was greater than 0.05, indicating no statistical significance for this variable in the regression. In the analysis by IRI level, a decrease in correlation was observed with increasing irregularity, but the effect of this difference was small, with R² ranging from 0.71 to 0.60. This analysis revealed two behaviors, allowing for additional analysis with multiple regression. However, even with the inclusion of the IRI variable, the model did not show statistical significance. Thus, the study concludes that longitudinal irregularity does not have a statistically significant influence on the deflections measured by the TSDd.