Statistical Determination of Agricultural Vehicle Induced Bridge Distribution Factor Threshold

Most studies to date on live-load distribution factor (DF) determination have focused on the examination of bridges under normal highway-type vehicles. Many bridges are in use on rural roadways where agricultural vehicle travel is common and therefore the impacts of their atypical load types should be explicitly considered. Developing systematic methodologies accounting for their effects are also critical for more reasonable safety assessment of rural bridges. The primary objective of this study is to develop a statistical framework to determine DFs for rural bridges subjected to agricultural vehicles. A sample bridge in a rural area of Iowa was selected for demonstration of the proposed statistical framework. The proposed procedure consists of multiple parts including live load field testing, finite element simulations, and statistical analyses. A network of multiple strain sensors were installed at the critical locations on the bridge to monitor strain time histories resulting from passes of farm vehicles with known characteristics. Strains were utilized to compute field measured DFs and also used to calibrate analytical models. As part of the model simulation, farm vehicles commonly found in the United States were chosen and applied to the model to calculate their DFs. Statistical thresholds for the exterior and interior girders were calculated by performing a statistical analysis of the computed data. The thresholds were compared to the AASHTO code-specified DFs (for both the Standard Specification and the LRFD Specification), indicating the interior girder limit was below yet the exterior limit was above the AASHTO values.