Mechanistic empirical homogenisation of British Pendulum Testers

Abstract

The British Pendulum Test (BPT) is commonly used to assess the skid resistance characteristics of road surfaces. However, the wide range of slider stiffness calibration limits of the BPT tester specified by standards can lead to significant inconsistencies in British Pendulum Numbers (BPNs) generated using different machines. This renders the test impractical for precise evaluation of pavement skid resistance. This study therefore proposes a mechanistic-empirical homogenisation technique which is easy to implement and does not require heavy computational power to resolve the issue. Linear equations between slider force variation and BPN output are developed using a detailed finite element simulation model. The linear relationships proposed in the study can transform BPNs from different BPT machines, with a standard deviation of 10 BPN units, into a reference BPN result, with a standard deviation of less than 1 BPN unit, for most test specimens. The proposed method is found to perform well across specimen gradations by reducing the maximum percentage difference from nearly 25% to less than 5% without requiring computationally intensive simulations. This technique can be used to reduce the variation in BPT measurements, leading to a more precise assessment of low-speed skid resistance properties of roads.