DescriptionThe bridge load rating provides a basis for determining the safe load carrying capacity of a bridge to ensure the bridge serviceability and safety. In New Jersey, a large number of steel truss bridges, including historical bridges, are currently in service. In the case that the as-built plans of a bridge are missing or not available, a particular challenge is raised for the agencies and engineers to determine the capacity of the structural members. Relying on the engineering judgement alone may lead to an inefficient load posting. Thus, a reliable methodology, which could also reduce the amount of work needed for the agencies to do the field investigations, is needed to load rate the steel truss bridges with no plans.
The proposed load rating procedure involves using the clustering methodology. The clustering methodology estimates the member sizes based on the similar bridges which were built in the same decade and have the same structural type, as well as similar geometries. This thesis focuses on finding the correlations between the member sizes and bridge geometries. The information of six (6) bridges from NJDOT bridge inventory were analyzed to construct the parametric study. From the study, it is found that the member sizes have strong correlations with geometries, such as stringer spacing, floorbeam length, bridge width, etc. The unknown member sizes then estimated based on the bridges with plans that have similar bridge geometries. The estimation results are then validated by the field inspection, proving the effectiveness of the clustering methodology. The accuracy of the clustering approach can be further improved by incorporating more bridges with plans into the cluster.
The load rating of one bridge without plans is performed in this study. In order to refine the rating factors, the Finite Element Model (FEM) and the advanced technology of Weight-in-Motion (WIM) system were also applied in this thesis. It is found that the FEM can significantly reduce the live load effects on the floorbeam compared with the line girder analysis using American Association of State Highway and Transportation Officials (AASHTO) LRFD Specification Girder Distribution Factor (GDF) equation. Furthermore, the WIM data was applied to find site-specific live load factor in order to better address the live load uncertainties within the specific region. The final load rating results indicate that there is no need to post load limitation for the target bridge without plans.