Weigh-In-Motion Testbed in New York City Provides Lessons Towards an Autonomous Weighing Enforcement of Oversized Vehicles to Reduce Damages to the Infrastructure.
New York City, New York, United States
Development of Autonomous Enforcement Approach using Advanced Weigh-In-Motion (A-WIM) System to Minimize Impact of Overweight Trucks on Infrastructure
Summary Information
Oversized vehicles pose a great safety risk by causing damage to roads and bridges and therefore require careful monitoring at weighing stations. This study evaluated the damage induced by overweight (OW) trucks in New York City (NYC) and implemented the advanced weigh-in-motion (WIM) system for future autonomous OW enforcement practice in collaboration with the NYC Department of Transportation. The study aggregated the bridge and pavement data from the National Bridge Inventory (NBI) database and bridge inspection reports dating back to 2016, and developed deterioration models for different types of structures to estimate their socio-economic impact on major highways.
The research team established a testbed along the triple cantilever of the Brooklyn-Queens Expressway (BQE) corridor to install, collect, and interpret various types of WIM sensors to determine the truck weight spectra effectively. Two WIM sites were proposed to monitor the traffic and truck weight for the BQE and to understand the overweight trucks on the triple cantilever. Two types of WIM sensors, namely piezoelectric and Quartz sensors were installed, and a calibration test was performed to determine the calibration factors and to test the accuracy of the implemented system. The WIM installation was performed in September 2019.
- Opt for quartz sensors rather than traditional piezo-type sensors at WIM stations. It was found in this study that the quartz sensors provide more reliable and accurate data compared to traditional piezo-type sensors. At lower gross vehicle weight (GVW) ranges < 60 kips, the PVDF sensors (piezoelectric polymer surrounded by a flat brass casing) overestimated the GVW by up to 78 percent. At higher GVW ranges > 80 kips, the PVDF sensors underestimated the GVW by 8 percent. At the GVW ranges between 60 kips and 80 kips, the PVDF sensors overestimated the GVW by 14 percent. The minimum and maximum GVW errors for quartz sensors were between -5.1 percent and 1.5 percent, respectively.
- Smooth the roadway segment where major pavement defects are observed to improve weighing accuracy. Cracks running between lanes might reduce the service life of the sensors and water could permeate the interface between sensor epoxy and pavement. The field implementation suggests that the smoothness would help improve the weighing accuracy.
- Use a calibration truck of the same type and/or configuration in which the owners are the most interested. When calibrating the sensors, generally, three- or four-axle single unit truck (FHWA Class six or seven) or five-axle semi-tractor trailer (FHWA Class nine) should be used for calibration. In this study, a Class Nine truck was used for calibration.
- Run the calibration truck multiple times over each lane where the WIM sensors are installed. During test runs, the truck shall maintain position in the middle of the lane to maximize the signal strength. Additionally, when passing over the sensors, the truck shall maintain a constant speed (no accelerating or breaking).
- Aggregate bridge inspection and pavement maintenance data for a more thorough and accurate deterioration model development. This would help identify the deterioration modes of different structure types, validate the deterioration models of different structure elements, and provide more accurate life-cycle infrastructure costs for analysis.
