ACCURACY AND TOLERANCES OF WEIGH-IN-MOTION SYSTEMS

A systematic study of in-motion weighing of some 800 trucks that were selected from the traffic stream on I-10 near Seguin, Texas, yielded data sets that were analyzed to define the attainable accuracy within which wheel, axle, axle-group, and gross vehicle weights could be estimated by a properly calibrated in-pavement weigh-in-motion system. Each truck that was weighed passed successively over the Radian weigh-in-motion system transducers at high (plus or minus 50 mph), intermediate (plus or minus 30 mph), and low (less than or equal to 10 mph) speed and then stopped on a special axle/wheel reference scale for successive static weighing of each wheel. Tolerances for a 95 percent confidence level were derived after the system had been calibrated to yield a zero mean of differences in the weigh-in-motion wheel weight estimates and the corresponding static wheel weights. The concept of use tolerances, which allow for the probable error in both the static wieight measurement and the weigh-in-motion weight estimate, is presented. Tolerances for high-speed weigh-in-motion, intermediate-speed weigh-in-motion, and low-speed weigh-in-motion scales at the experimental site are tabulated.