FALL 2015

ASPIRE is a quarterly magazine published by PCI in cooperation with the associations of the National Concrete Bridge Council. The editorial content focuses on the latest technology and key issues in the Concrete Bridge Industry.

Issue link:

Contents of this Issue


Page 35 of 55

FHWA FHWA Strategy to Increase Use of Refined Analysis by Reggie Holt and Dr. Brian Kozy, Federal Highway Administrationd Engineering practitioners of today, with the aid of ever-advancing computer technology, are able to solve engineering problems of great complexity, and produce designs/evaluations that are more refined and more reliable than in the past. However, our nation's governing bridge design specifications and the profession as a whole have not yet fully exploited the capabilities of this new generation of analytical tools. Many bridge engineers and owners appear to favor a general philosophy of keeping analyses as simple as possible to minimize errors or to remain true to the accepted, proven engineering practices. Consequently, they have avoided embracing regular use of refined analysis methods. In 2009, an international technology scan sponsored by Federal Highway Administration (FHWA), the American Association of State Highway and Transportation Officials (AASHTO), and the National Cooperative Highway Research Program (NCHRP) determined that engineers in the United States significantly lagged behind their European counterparts in the use of advanced modeling tools and procedures to design and assess bridges. The scan team recommended increased use of refined analysis for bridge design and evaluation, and encouraged the use of refined analysis to avoid unnecessary posting, rehabilitation, or replacement. Unfortunately, practical implementation of these recommendations has been limited. What is Refned Analysis? The generic term "refined analysis" is often used to describe a more-detailed, sophisticated structural modeling approach, which typically involves computerized finite-element analysis (FEA). A significant number of references to refined analysis are made in the AASHTO LRFD Bridge Design Specifications, however no formal definition is provided. These references are typically along the lines of "in lieu of a refined analysis, the following can be used," implying that the provided approximate (simplified) analysis procedure is deemed sufficient for most cases, but refined analysis should be considered if more complexity is involved. The AASHTO LRFD specifications also acknowledges and specifies limits of applicability for many approximate procedures indicating that in some cases refined analysis is required. Therefore, using the AASHTO LRFD specifications as the governing bridge analsys specification, one could define refined analysis as: Any analysis that provides more accurate results or addresses complex structural components/systems or behaviors that fall outside the limits of the AASHTO LRFD specifications' approximate procedures. Based on the previous definition, analytical procedures that would not be considered refined would include the following: * Line girder analysis using distribution factors * Moment magnification procedure for compression elements * Strut-and-tie models of concrete elements * Strip method of deck analysis and design * Cross-sectional frame analysis for box girders * Equations for effective flange width of composite decks considered refined (Figures 1-3) include the following: * System modeling that accounts for load distribution to girder lines * Sectional modeling that accounts for shear lag, local stresses, or distortion * Models explicitly defining diaphragms/cross frames or the deck as a surface(rather than a grid) in two dimensions * Models using plastic hinges, such as by pushover analysis Why Use Refned Analysis? In some cases refined analysis is required to complete the design verification according to the AASHTO LRFD specifications. These are instances for which the code-specified approximate methods do not apply. Furthermore, there are reasons why using a refined analysis might be advantageous, such as capturing behavior not adequately accounted for by approximate methods and/ or outside the limits of the AASHTO LRFD specifications.

Articles in this issue

Archives of this issue