Weigh-in-Motion Equipment – Experiences and Challenges

The Norwegian Public Roads Administration focus on issues to improve weight controls, and the quality of statistic weigh data. An important part of the project is field tests of different methods and equipment for collecting Weigh-In-Motion (WIM) data, with high speed WIM equipment, and vehicle to infrastructure (V2I) communication where the vehicle reports its own weight. The tests are focusing on data quality (WIM vs static weight), user interface and usability, procedure and need for calibration and lifetime of sensors. Another important part of the project is the use of WIM data to pre-select vehicles at the control station. WIM data are combined with license plate numbers from automatic number plate recognition (ANPR) cameras, and the measured weight is compared with the permitted weight. If a vehicle is overloaded then a warning is shown on the control station. This allows legal drivers to pass without being stopped. The authors have conducted several tests over a three years period. Two different WIM sensors technologies from four different manufacturers has been tested. The focus of the tests have been on accuracy, lifetime of sensors, calibration and usability of software and equipment. The authors have tested equipment for high speed WIM for pre selection of vehicles at control stations. The results shows that lineas quarts sensors are better than piezo electric cables. The calibration procedures are time end cost consuming. Lifetime of the sensors are very short. This can be improved by deeper installation where it is possible. It is a problem that the accuracy of the sensors decrease over time. For one system the error in average gross weight increased to about 15% over a period of 6 months. The error in the other system had increased to 15% in three months, and after calibration the error increased to 15% in just over one month. A full scale test of a V2I system was conducted in Oslo in February 2015, as a proof of concept. One instrumented vehicle was used to verify the communication protocols. The information flow was steady and the focus was therefore on the reported weight. The internal difference in the self-declared truck weight for each axle was very low – with less than 1% difference between the test runs. Compared with the static weight the self-declared truck weight was approximately 5% too low. This was considered as a good start for the concept. A full scale tests with four vehicles are planned to start in September. The trucks will be equipped with communication units for sharing the current weight for a two month period. This setup will provide us with a wider basis for evaluating the data quality provided by the system. The authors believe that in the future WIM systems will be used together with V2I solutions, and that this will improve vehicle weight control further. The project is a cooperation between the Norwegian Public Roads Administration and the Traffic Engineering Research Centre.