Author: Devin K. Harris

Publisher:

Published: 2020

Total Pages: 0

ISBN-13:

Load rating is the process of determining the safe load-carrying capacity of a bridge; however, when plans and details are insufficient to determine the overall capacity of the structure, alternative methods must be used to infer what the live load capacity is. Two viable methods allowed by the AASHTO Manual for Bridge Evaluation are the commonly used but subjective engineering judgement and the experimentally based proof testing. However, these methods suffer from limitations. Engineering judgement typically is not based on physical phenomena and creates a degree of risk in unconservative estimates or unnecessarily restricts traffic and commerce if estimates are overly conservative. On the contrary, proof testing can cause damage during testing, tends to be expensive, and cannot be extrapolated to future performance. Thus, the objective of this study was to develop rational engineering approaches for load rating structures within the Virginia Department of Transportation (VDOT) inventory for which limited as-built information is available. The initial phase of the investigation focused on categorizing the VDOT inventory to determine the types of structures that are likely to be missing information necessary for an analytical load rating, which were identified to be short span reinforced concrete slab or T-beam designs. Subsequent phases emphasized two main approaches to load rating: (i) structural identification frameworks based on finite element model updating; and (ii) leveraged vibration response characterization. Both approaches emphasized estimating unknown characteristics of these types of structures for use in a traditional analytical load rating. These unknown parameters include modulus of elasticity and strength of concrete as well as cross-sectional area of steel reinforcement. These estimates can ultimately be used to provide a rational estimate of load ratings. All approaches were evaluated on two slab and two T-beam structures in varying condition states, which had sufficient plans available, but were treated as having varying degrees of unknown details. The results illustrated that the finite element model updating method generated load ratings that were within 0% to -17% of the load ratings developed according to conventional calculations, with negative differences indicating lower rating factor estimates; and the vibration-based simplified method led to results with a percent difference ranging from 16% to -16%. It was also shown that instrumenting bridges with a limited number of sensors is sufficient for successful implementation of the developed methods. The results from the study have been synthesized into recommendations for VDOT to perform load ratings of structures with insufficient plans or information, with the goal of minimizing the degree and complexity of experimental measurement as well as simplifying the tools for performing the analyses of these structures as much as feasible.