Quality control of weigh-in-motion data incorporating threshold values and rational procedures

Abstract

Weigh-in-motion (WIM) stations constitute a key source of traffic data for use in mechanistic–empirical pavement design. One of the major improvements provided by the Mechanistic–Empirical Pavement Design Guide (MEPDG) is in traffic characterization. Instead of converting all truck axles to 18,000lb equivalent single axles (ESALs), the Mechanistic–Empirical Pavement Design Guide (MEPDG) simulates every truck axle, and the associated stresses and strains imposed on the pavement structure, from a wide range of axle load spectra (ALS). This paper presents an objective approach to quality control (QC) of WIM data that includes threshold checks that detect implausible values of individual variables in the truck weight records and rational checks that examine patterns in axle load distributions and relationships among the variables. Instead of using subjective visual comparisons of gross vehicle weight (GVW) distributions, this research implements a peak-range check, peak-shift check, and correlation analysis to quantify the ALS comparison process of rational checks. A number-of-axles check that calculates the average number of axles per vehicle class is also introduced herein. The entire QC procedure has been applied to three years of data from 12 WIM stations in Alabama that used bending plate sensors. As a result, 23.8% of data were filtered out, and all data from one WIM station were removed. Therefore, QC of WIM data is strongly recommended, regardless of the extent of WIM system calibration. Furthermore, it is also recommended that the rational checks module be integrated in the data collection process for rapid detection of systematic errors.