Chimich DD, Rutledge BA, Elkin BS, Siegmund GP (2018). Variability of walkway tribometer measurements. 8th World Congress of Biomechanics: Dublin, Ireland.
Walkway tribometers are devices that measure the slip resistance of walking surfaces for the purpose of assessing slip risk and flooring performance. To be useful, these measurements must be repeatable and linked to human slip risks. ASTM International provides walkway tribometer suppliers with a standard practice (ASTM F2508-16) and a set of four reference surfaces to validate their tribometer design by properly ranking and statistically differentiating the reference surfaces to match their human slip risk potential (Powers et al., 2010). This practice also provides users with a procedure to calibrate their tribometer against the supplier’s validation. To be considered a calibrated walkway tribometer, the mean slip resistance value for each reference surface within calibration testing must lie within the supplier’s 95th percentile confidence interval (95th%CI) for that surface. It is our experience and understanding that few users have met this calibration requirement with their tribometers.
Our goal here was to quantify the variability within the calibration procedure defined in ASTM F2508-16 for three common walkway tribometers (Mark IIIB, English XL, and BOT 3000E). Repeated calibrations (n=9 or 10) were performed using each tribometer on a single set of reference surfaces (RSA-Granite, RSB-Porcelain, RSC-Vinyl composition tile (VCT), RSD-Ceramic). The first five calibrations for each tribometer were done over five consecutive days, and the remaining four or five calibrations were spaced at least one week apart. A tribometer calibration consists of 16 slips (four in each of four orthogonal directions) on each reference tile, and generates a slip resistance value (mean of the 16 slips) for each surface. One operator performed the Mark IIIB and BOT tests; another operator performed the XL tests.
Across all 28 tribometer calibrations, only 12 of the 112 slip resistance values fell within the supplier’s 95th%CI (Figure 1). Only once did two slip resistance values from the same calibration fall within the 95th%CIs. For a given surface, the ranges of slip resistance values were 3 to 20 (7.5 ± 5.1) times wider than the supplier’s 95th%CIs (Figure 1).
None of the three tribometers we tested met the 95th%CI requirement of ASTM F2508-16 for all four surfaces, despite multiple calibrations. Even under controlled conditions of one tribometer (per design), one user and one set of reference surfaces, the range of slip resistance values were many times larger than the supplier’s 95th%CI. Based on these data, the 95th%CI requirement in ASTM F2508-16 is too narrow. Future tribometer standards should consider the variability shown here, plus additional variability that may arise from different users, different sets of reference surfaces, and different samples of the tribometer models.