A Statistical Spatial Repeatability Algorithm for Multiple Sensor Weigh-in-Motion

The use of an array consisting of multiple WIM sensors is well established as a means of increasing the overall accuracy of in-motion axle weighing. This paper proposes a new algorithm for processing the outputs from such arrays and finding an improved estimate of the static axle weights. The mean pattern of forces from the axles of many vehicles are repeatable – this is the phenomenon of Statistical Spatial Repeatability (SSR). The recorded WIM force is therefore, on average, over- or under-estimating the static axle weight. Removing this bias and averaging the corrected measurements is the basis for the new algorithm. The newly developed SSR algorithm is assessed using experimental data from the WIM-Hand project test site near Arnheim in the Netherlands. The accuracy of the new algorithm was therefore assessed by applying the algorithm twice, once for each half of the multiple-sensor array, and comparing the calculated static weights. The difference was found to be small, which demonstrates that the calculation is accurate.