Fusion of weigh-in-motion and global positioning system data to estimate truck weight distributions at traffic count sites

Abstract

Truck weight data is needed for a wide range of applications including but not limited to pavement design, weight enforcement, traffic monitoring, and freight transportation planning. Unfortunately, the low spatial resolution of weight sensors along the transportation network can limit these and other potential applications. The main contribution of this paper is a methodology to estimate gross vehicle weight (GVW) distributions at traffic count sites, which collect traffic volumes but currently do not have the ability to directly measure vehicle weight. This paper presents a method for estimating GVW distributions of five-axle tractor-trailers (“3-S2”) at traffic count sites by fuzing weight data from weigh-in-motion (WIM) sites with position data from global positioning system (GPS) equipped trucks. Truck travel patterns derived from a truck GPS database were used to determine the degree to which a WIM and traffic count site are spatially related. A GVW distribution was then estimated by combining Gaussian mixture models (GMM) estimated at WIM sites defined to be spatially related to the traffic count site. A leave-one-out cross validation framework allowed for comparisons of estimated and measured GVW distributions at each WIM site. Coincidence ratios and two-sample Kolmogorov-Smirnov (KS) tests were used as comparison metrics for a case study of 112 WIM sites in California. The proposed methodology provided better goodness-of-fit between observed and estimated GVW distributions compared to a baseline approach which defined the spatial relation between sites using great circle distances (GCD).

Keywords:

• Gaussian mixture model
• global positioning systems (GPS)
• gross vehicle weight (GVW)
• spatial interpolation
• weigh-in-motion (WIM)

Acknowledgements

The authors kindly acknowledge the support and feedback received from Stephen G. Ritchie and Andre Tok of the Institute of Transportation Studies at the University of California, Irvine. The authors also thank the reviewers and editor for their comments and suggestions.