This report presents the results of a study on the calculation of load and resistance factors for the AASHTO LRFD Bridge Design Specifications. Information on various load models, and procedures for determining reliability indices, are included. The development of a new load and resistance factor design (LRFD) code for the design of bridges in the United States required the calculation of factors that were consistent with both theory and the performance of existing bridges. Load factors account for the variability in live and dead loads that a structure will endure during its design life. Resistance factors account for imperfect knowledge regarding material characteristics (especially strength), structural member geometries, and the static and dynamic behaviour of bridges, and the effect this lack of knowledge has on the ability of structures to withstand loads. Because bridge design in the United States through the 1980s was based on the working stress (allowable stress) and load factor methods (neither of which had formal, probabilistically determined factors for both loads and remittances), significant new information was needed to provide the range of factors used in design. NCHRP Project 12-33, "Development of a Comprehensive Bridge Specification and Commentary," was initiated in 1988 with the objective of developing a comprehensive new design code that could eventually replace the AASHTO Standard Specifications for Highway Bridges (which was considered disjointed, fragmented, and not state of the art). The product of Project 12-33 was published by AASHTO in 1994 as the LRFD Bridge Design Specifications, and a summary of the project is published in NCHRP Research Results Digest 198. A significant part of the project was the development and calibration of the load and resistance factors, and that work is the basis for this report. The research was performed by the University of Michigan, in Ann Arbor, Michigan, under a subcontract to Modjeski and Masters, Inc., of Mechanicsburg, Pennsylvania. The research results are presented in a form that allows researchers and bridge engineers to understand the loads that were considered during the course of the project, the types of structural resistance that were investigated, the concept of reliability and the target reliability indices chosen for the design code, and, finally, the load and resistance factors recommended for inclusion in the design specifications. The report also describes issues related to the state of the practice-that is, how the factors selected were intended to result in structures that performed as satisfactorily as those designed and built using the "older" methods of working stress or load factors. Detailed information is provided regarding the database of bridges that served to calibrate the new factors; this database represents bridges of many geometries, materials, and span lengths from across the nation. The report provides the basis for the continuous refinement of the bridge design code as more and better data are generated related to loads, load variability, materials, workmanship, and bridge performance. (A)
Abstract