Transverse Reduction Factor of Multi-lane Bridges Considering Lane Load Disparity

To consider the influence of lane load disparity on the transverse reduction factor of multi-lane bridges, a novel multi-coefficient model of the transverse reduction factor was proposed, and lane-load meeting approach based on the independent repeated trials was improved to calibrate these coefficients. First, a lane load correction coefficient that reflects the lane load disparity and a multi-lane combination coefficient that reveals the occurrence probability of lane-load meeting events were introduced. These shaped the multi-coefficient model of the transverse reduction factor. Then, the commonly known lane-load meeting approach based on the independent repeated trials was improved to consider the differences in truck volume and truck load over multiple traffic lanes. In addition, critical parameters that influence the calibration of the transverse reduction factor were studied. These included the average time for a truck to pass through a control cross-section, the return period of the traffic load, lane truck volume, and lane load distribution. Finally, measured weigh-in-motion data from a freeway were used to calibrate the transverse reduction factor, and the results were compared with those derived from the conventional approach and bridge design specification JTG D60-2015. The research shows the improved lane-load meeting approach is a general solution that allows the questionable assumptions used in the conventional approach (e.g., normal distribution of truck weight, identical truck weight distribution to different lanes, and 3.5-times standard deviation between the maximum and mean truck weights) to be discarded. Parametric studies indicate the average time of a truck passing through the control cross-section and the return period of the traffic load have slight effects on the results, whereas the truck weight feature and its distribution over multiple lanes have the most significant effects. This highlights the importance of precise modeling of lane loads. Statistics of truck loads show the truck volume and distribution of truck loads are obviously different in various traffic lanes, and the transverse reduction factors given by the conventional approach and the design code overestimate the realistic condition, which is as much as 19% and thus may lead to waste of design and management resources. However, the transverse reduction factor calibrated by the proposed approach and model form may clearly reveal the contribution of truck loads in each lane and accurately reflect the transverse reduction raw of lane load under the combined loading of different numbers of traffic lanes. This would have wide engineering applications. © 2020, Editorial Department of China Journal of Highway and Transport. All right reserved.