Development of Surface Roughness Standards for Pathways Used by Wheelchair Users: Final Report

APPENDIXES

(Nonmandatory information)

X.1 PRECISION AND BIAS

X1.1 Precision:

X1.1.1 The precision of the computed PRI is limited by the procedures used in making the longitudinal profile measurement

X1.1.2 PRI precision depends on the interval between adjacent profile elevation measures. Reducing the interval typically improves the precision. An interval of 1.0 mm (0.04 in.) or smaller is recommended. For some surface types, a shorter interval will improve precision. More information about the sensitivity of PRI to the profile data interval is being developed.

X1.1.4 PRI precision is limited by the degree to which a traveled path on the pathway can be profiled. Errors in locating the traveled path longitudinally and laterally can influence the PRI values, because the PRI will be computed for the profile of the traveled path as measured, rather than the travel path as intended. These errors are reduced by using longer profiles.

X1.1.5 If measurements are taken so that the least significant digit is 0.01 mm or smaller, Computational errors due to rounding can be safely ignored.

X1.2 Bias:

X1.2.1 The bias of the computed PRI is typically limited by the procedures used in making the longitudinal profile measurement.

X1.2.2 PRI bias depends on the interval between adjacent profile elevation measures. An interval of 1.0 mm (0.04 in.) or smaller is recommended. Shorter intervals improve precision but have little effect on bias. More information about the sensitivity of PRI to the profile data interval is being determined.

X1.2.3 Many forms of measurement error cause an upward bias in PRI due to variations in profile elevation from measurement error not correlated with the profile changes. Some common sources of positive PRI bias are: height-sensor round-off, mechanical vibrations in the instrument that are not corrected and electronic noise. Bias is reduced by using profiler instruments that minimize these errors.