A METHOD TO CALIBRATE AND CORRELATE ROAD ROUGHNESS DEVICES USING ROAD PROFILES.

摘要

Road Roughness is very important input information for road planning and maintenance. It is measured for the purposes of road performance prediction, vehicle operating cost prediction, road safety evaluation, newly-constructed road acceptance, and maintenance cost calculation. Since there are various types of roughness measuring devices and indices as well as variations in roughness measurement over time, the main purpose of this study was to find a method which can be used to correlate between roughness measurement of one device and that of another type of device, and to calibrate the roughness measurement of the same device at different times.;Three sets of Maysmeter calibration data were analyzed and a simulation of these Maysmeter responses was modeled using root-mean-square vertical acceleration of two baselengths. Based on a model resulting from combining the obtained Maysmeter simulation models, standard roughness scale was formulated. The most practical method for obtaining this standard scale consists of measuring road profile with a surface dynamic (SD) profilometer, and inputing these profiles to the standard roughness scale model.;Due to the impracticality of using the SD profilometer in some circumstances, the rod and level method was evaluated and found to be an effective substitution of the SD profilometer, particularly for developing countries, as the method requires relatively low capital cost and minimum repair and is labor intensive. Since the rod and level study was done on only 10 sections, further study is needed (with more sections, as well as with unpaved sections).;Various calibration methods were evaluated to determine the most effective method for generating a standard roughness index. It is concluded that the method which is composed of a set of real sections and a roughness change monitoring system is the most practical. A profile processing technique, the root-mean-square vertical acceleration, was developed in this study. This technique can generate a set of roughness indices by varying the stepsize or baselength (b) at which a vertical acceleration (VA) is calculated.