Traffic load modelling and factors influencing the accuracy of predicted extremes

Abstract

Design and assessment of highway bridge structures requires accurate prediction of the maximum load effects, i.e. shear forces and bending moments etc., which may be expected during the proposed or remaining life of the structure. Traditionally these effects are calculated using conservative deterministic loading models prescribed by codes of practice. The inherent conservatism of these models lies in their need to be widely applicable. While this conservatism is relatively insignificant in design, it may be critical in assessment. In recent years advances in Weigh-in-Motion (WIM) technology have led to the increased availability of accurate and unbiased site-specific traffic records. These records have been successfully employed in the derivation of site-specific loading models and in calculation of load effects in assessment of bridge structures. The results of these assessments are accepted to be less conservative than those performed using generalised codified loading models. Given this reduction in the conservatism of the calculation it is important to quantify the implication of factors such as data inaccuracy or traffic growth on the calculated maximum load effects. This paper briefly describes the mathematical modelling involved in traffic simulation using WIM statistics. The results of direct simulations performed using WIM data are compared with those obtained through the statistical simulation technique termed Monte Carlo simulation, which is regularly employed where insufficient measured data exists. The implications of the accuracy of the recorded WIM data and the duration of recording on the predicted load effect are assessed along with the sensitivity of the extreme to the method of prediction. The effect of traffic evolution with time in terms of increased volumes of flow and weight limits are also explored.