Revised Methodology for Computing International Friction Index

An investigation evaluated the principles underlying the concept of the international friction index (IFI) to revise the current ASTM methodology for calibration of the IFI model parameters. The current ASTM standard for computing the IFI assumes a linear correlation between speed-corrected measurements of all friction-measuring devices that operate at different slip ratios with those of the standard device, the dynamic friction tester (DFT), which operates at 100% slip. This investigation revealed that a linear correlation existed between measurements converted to the standard slip speed of 60 km/h (FR60 values) obtained from devices that operate at 100% slip conditions and those of the DFT. However, the relationships between FR60 values obtained from friction-measuring devices that operated in the slip range of 10% to 20% and those of the DFT were observed as nonlinear. The FR60 values obtained with the latter category of friction-measuring devices when transformed by a power transformation produced a linear relationship with the standard F60 values obtained by using the DFT. A revised methodology for computing the IFI was proposed. This revised methodology yielded higher predictive capability than the current ASTM procedure and calibration parameters A and B that were speed invariant. The revised methodology also captured the actual friction–speed relationship of any given friction-measuring device on a given pavement surface by incorporating not only surface characteristics but also device-specific characteristics into the definition of the speed constant.