Integration of Rolling Wheel Deflectometer Deflection Measurements into Pavement Management Systems: Use of Multivariate Statistical Methods and Fuzzy Logic

The Louisiana Department of Transportation and Development (DOTD) conducted a research project with the rolling wheel deflectometer (RWD) on approximately 1,400 mi of asphalt concrete roadways. The major outcomes of the study were evaluation of the effectiveness of the RWD as a structural assessment tool on Louisiana highways and the introduction of the RWD index (RI). The purpose of this study was to define RI ranges for treatment selections or pavement structural conditions linked to current methods of the Louisiana DOTD's pavement management system; multivariate (factor analysis) and descriptive [analysis of variance (ANOVA)] statistical methods and fuzzy logic functions were used. The factor analysis revealed that the RI could be used as a single index for assessing pavement structural conditions. ANOVA revealed that the RI could not be used to establish the four treatment methods typically used by the Louisiana DOTD but that it could be used to infer whether a pavement was structurally sound or deficient. If the pavement were determined to be structurally deficient, additional structural evaluation, such as with a falling weight deflectometer, would be required. Fuzzy logic and set theory were used to establish fuzzy functions for structurally sound and structurally deficient pavement conditions along with RI threshold ranges per thickness group. Equations were derived to produce RI threshold values for pavement thicknesses less than 12 in. The equations may be used for pavements thicker than 12 in., but with caution. For large segments of roadways, algorithms were used to assess whether pavements were structurally sound or structurally deficient on the basis of RI.