Study on the accuracy of axle load spectra used for pavement design

Weigh-in-Motion (WIM) systems are used in order to reduce the number of overloaded vehicles. Data collected from WIM provide characteristics of vehicle axle loads that are crucial for pavement design as well as for the development of pavement distress prediction models. The inaccuracy of WIM data lead to erroneous estimation of traffic loads and in consequence inaccurate prediction of pavement distress process. The objective of the paper is to present a new methodology of heavy traffic axle load spectra (ALS) correction due to weighing errors (systematic and random) that occur in WIM systems. The theoretical solution which is proposed in the paper was validated successfully. The method enables correction of erroneous data to make traffic load statistics used for pavement design more reliable and precise, with no necessity to remove high number of records, as it is used in other methods. The practical meaning of the newly developed method was emphasised by analysis of the effect of relative and random error of WIM data on pavement fatigue life estimation, as well as on the estimated percentage of overloaded vehicles. Mechanistic-empirical approach (M-EPDG) was used for this purpose.