«R. SHANKAR NAIR R. Shankar Nair R. Shankar Nair, Ph.D., P.E., S.E. is a principal and senior vice president of Teng & Associates, Inc. in Chicago. In ...»
Federal Emergency Management Agency (FEMA). Recommended Quality Assurance Criteria for Steel Moment Frame Construction, Report No. FEMA 353, prepared by the SAC Joint Venture for FEMA, Washington, D.C., 2000c.
Federal Emergency Management Agency (FEMA). Prestandard and Commentary for Seismic Rehabilitation of Buildings, Report No. FEMA 356, FEMA, Washington, D.C., 2000a.
General Services Administration (GSA). GSA Progressive Collapse Analysis and Design Guidelines for New Federal Office Buildings and Major Expansion Projects, prepared by Applied Research Associates for GSA, Washington, D.C., 2003.
Hamburger, R, Mayes, R, Parker, J. “Impact of High Strength Steel on Building Applications, 3 Applications,” August, 2003 Advanced Technology Institute, Proceedings of Workshop on High Strength Low Alloy Steels in Defense Applications, Chicago, IL Partin, B.K., Bomb Damage Analysis of the Alfred P. Murrah Federal Building, Oklahoma City, Oklahoma, 1995, Congressional Register Sozen, M., Thornton, C., Mlakar, P., Corley, G., “The Oklahoma City Bombing: Structure and Mechanics of the Murrah Building,” Journal of the Performance of Constructed Facilities, ASCE, Reston, VA, Vol. August, 1998, pp. 120-136
PERFORMANCE-BASED ENGINEERING OF
BUILDINGS FOR EXTREME EVENTS
ANDREW WHITTAKER, S.E., RONALD HAMBURGER, S.E., AND MICHAEL MAHONEY
BIOGRAPHIESAndrew Whittaker: Andrew Whittaker is an Associate Professor of Civil Engineering at the University at Buffalo, with research and design-professional interests in earthquake and blast engineering. He is a licensed Structural Engineering in the State of California. Dr. Whittaker serves as a member of the ASCE committees that address loadings on structures, blast engineering, and earthquake protective systems, ACI Committee 349 on reinforced concrete nuclear structures, BSSC Technical Subcommittee 12 on seismic isolation and passive energy dissipation systems for new buildings, as Vice President of the Consortium of Universities for Research in Earthquake Engineering, and as the Structural Team Lead for the ATC-58 project on performance-based earthquake engineering. Dr. Whittaker was a member of the 3-person MCEER-NSF reconnaissance team that collected forensic data at Ground Zero in the weeks after the September 11, 2001, terrorist attacks.
Ron Hamburger: Ronald Hamburger is a structural engineer and principal with Simpson Gumpertz & Heger in San Francisco. Mr. Hamburger has nearly 30 years of experience in structural design, evaluation, upgrade, research, code and standards development, and education. Mr. Hamburger serves as chair of the BSSC Provisions Update Committee and the AISC Connection Prequalification Review Panel, and is vice-chair of the AWS D1.1 Seismic Task Committee. Further, he is a member of the ASCE-7 committee, is President-Elect of the National Council of Structural Engineering Associations and is the Project Director for the ATC-58 project on performance-based earthquake engineering. Mr. Hamburger was a member of the joint FEMA/ASCE Building Performance Assessment Team that studied the collapse of New York’s World Trade Center on September 11, 2001.
Michael Mahoney: Michael Mahoney is a Senior Geophysicist with the Mitigation Division of the Department of Homeland Security’s Federal Emergency Management Agency. He currently serves as the Acting Director of the National Earthquake Program Office in Federal Emergency Management Agency. Since 1991, Mr. Mahoney has been responsible for many of the technical activities under the National Earthquake Hazards Reduction Program (NEHRP), including the FEMA/SAC Steel Moment Frame Buildings Study and FEMA’s current Performance Based Seismic Design initiative. From 1984 to 1991, Mr. Mahoney was with FEMA's Office of Loss Reduction, within the National Flood Insurance Program. From 1978 to 1984, he was employed as a Loss Prevention Consultant with Factory Mutual Engineering. He holds Bachelors and Masters Degrees in physics.
Blast, earthquakes, fire and hurricanes are extreme events for building construction and warrant innovative structural engineering solutions. The state-of-the-practice and new developments in performance-based earthquake engineering (PBEE) are discussed, with emphasis on hazard intensity measures, engineering demand parameters, and performance levels. The new performance-based earthquake engineering methodology is extended to performance-based blast engineering. Sample intensity measures, engineering demand parameters, and performance levels are proposed for blast engineering. Some similarities and differences between performance approaches for blast and earthquake engineering are identified.
Performance-based engineering of buildings and infrastructure for gravity and windstorm loadings has been indirectly undertaken for more than 20 years since the introduction of strength (concrete structures) and load-andresistance-factor (steel structures) design in the 1960s and 1970s. Such engineering of buildings and infrastructure has been based on force-based analysis and design checking of components using ∑ αi Li ≤ φ C (1) where α i are load factors, Li are load effects (e.g., dead load, live load), φ is a capacity reduction factor for the action that is being checked (e.g., moment, axial load, shear force) and C is the component capacity that is determined using a materials standard such as the AISC Load and Resistance Factor Design Manual (AISC 2002).
Factored component force demands are required to be less than or equal to de-rated component force capacities. The values of α i and φ were selected to ensure that the probability of component failure, measured here as component demands greater than component capacities, is extremely low. Global performance of a framing system is measured by performance at the component level. No statements are made regarding the relationship between component and system failure.
Extreme loadings on buildings and infrastructure are produced by natural and man-made hazards including strong earthquakes, hurricane and tornado winds, blast, fire and equipment malfunctions. Setting aside equipment malfunctions for the purpose of this paper, the extreme loadings of Figure 1 should be resisted by buildings and infrastructure without collapse for sufficient time so as to allow the occupants the time required to exit the structure.
Figure 1. Extreme loadings and effects on building structures Blast and earthquake loadings are short-term loadings with durations measured in milli-seconds and seconds, respectively.
For such loadings, component and system ductility can be utilized to avoid system collapse. For relatively long-duration loads such as hurricane wind loadings on buildings, strength alone must be used to avoid collapse. Because component and system ductility are related to framing system displacements and deformations, and not component forces, performance-based engineering for extreme blast and earthquake loadings must be displacement or deformation-based based rather than force-based per (1).
The following sections of this paper provide summary information on the state-of-the-practice in performance-based earthquake engineering and the framework for on-going and future developments in performance-based earthquake engineering. Aspects of the performance-based earthquake engineering framework that might prove useful in the development of performance-based guidelines for blast engineering of buildings are identified. Some similarities and differences between performance approaches for blast and earthquake engineering are identified.
PERFORMANCE-BASED EARTHQUAKE ENGINEERING OF BUILDING STRUCTURESPractice of performance-oriented earthquake engineering The traditional prescriptive provisions for seismic design contained in U.S. building codes (e.g., ICBO 1997; FEMA 2000b) and under development since the late 1920s (ATC 1995) could be viewed as performance-oriented in that they were developed with the intent of achieving specific performance, that is, avoidance of collapse and protection of life safety. It was assumed by those engineers preparing the codes that buildings designed to these prescriptive provisions would (1) not collapse in very rare earthquake; (2) provide life safety for rare earthquakes; (3) suffer only limited repairable damage in moderate shaking; and (4) be undamaged in more frequent, minor earthquakes. The shortcomings of the prescriptive procedures include fuzzy definitions of performance and hazard and the fact that the procedures do not include an actual evaluation of the performance capability of a design to achieve any of these performance objectives. Further, records of earthquake damage to buildings over the past 70+ years following minor, moderate and intense earthquake shaking has demonstrated that none of the four performance objectives has been realized reliably. Deficiencies in the prescriptive provisions in terms of accomplishing the four target objectives have been identified following each significant earthquake in the United States and substantial revisions to the prescriptive provisions have then been made.
Performance expectations for mission-critical buildings began to evolve in the mid-1970s following severe damage to a number of emergency response facilities, most notably hospitals, in the 1971 San Fernando earthquake.
Earthquake engineers decided that those buildings deemed to be essential for post-earthquake response and recovery (e.g., hospitals, fire stations, communications centers and similar facilities) should be designed to remain operational following severe earthquakes, and assumed that this would be achieved by boosting the required strength of such buildings by 50% compared with comparable non-essential buildings and requiring more rigorous quality assurance measures for the construction of essential facilities1. Since that time, the prescriptive provisions have evolved slowly but still include few direct procedures for predicting the performance of a particular building design, or for adjusting the design to affect the likely performance, other than through application of arbitrary importance factors that adjust the required strength.
Large economic losses and loss of function in mission-critical facilities following the 1989 Loma Prieta and 1994 Northridge earthquakes spurred the development of performance-based seismic design procedures with the goal of developing resilient, loss-resistant communities. In the early 1990s, experts design professionals and members of the academic community, ostensibly structural and geotechnical engineers, recognized that new and fundamentally different design approaches were needed because the prescriptive force-based procedures were a complex compendium of convoluted and sometimes contradictory requirements, were not directly tied to the performance they were intended to achieve, were not reliable in achieving the desired protection for society, were sometimes excessively costly to implement, and were not being targeted at appropriate performance goals in most cases.
Although the 50% increase in strength served to reduce damage to the structural framing, there is no evidence to support the assumption that the essential facility would be operational after severe earthquake shaking.
Funding in the early to mid-1990s from the Federal Emergency Management Agency (FEMA) to the Applied Technology Council (ATC) and the Building Seismic Safety Council (BSSC) led to the development of the NEHRP Guidelines and Commentary for Seismic Rehabilitation of Buildings (FEMA 1997). This development effort marked a major milestone in the evolution of performance-based seismic design procedures and articulated several important earthquake-related concepts essential to a performance-based procedure. The key concept was that of a performance objective, consisting of the specification of the design event (earthquake hazard), which the building is to be designed to resist, and a permissible level of damage (performance level) given that the design event is experienced. Another important feature of the NEHRP Guidelines (FEMA 273/274) was the introduction of standard performance levels, which quantified levels of structural and nonstructural damage, based on values of standard structural response parameters. The NEHRP Guidelines also specified a total of four linear and nonlinear analysis procedures, each of which could be used to estimate the values of predictive response parameters for a given level of shaking, and which could then be used to evaluate the building’s predicted performance relative to the target performance levels contained in the performance objective. Figure 2 below illustrates the qualitative performance levels of FEMA 273/274 superimposed on a global force-displacement relationship for a sample building. The corresponding levels of damage are sketched in the figure. Brief descriptions of the building damage and business interruption (downtime) for the three performance levels are given in Table 1.
Table 1. Building performance levels per FEMA 273/274/356 (after Comartin) Figure 3 illustrates the FEMA 273/274 nonlinear static procedure for performance assessment.
First, the earthquake hazard is characterized by one or more elastic acceleration response spectra. A nonlinear mathematical model of the building is prepared and subjected to monotonically increasing forces or displacements to create the capacity curve of Figure 3, which is generally plotted in terms of base shear (ordinate) versus roof displacement (abscissa). A maximum roof displacement is calculated for each design spectrum using an equivalent SDOF nonlinear representation of the building frame. Component deformation and force actions for performance assessment are then established for the given roof displacement using the results of the nonlinear static analysis. Component deformation and force demands are then checked against component deformation and force capacities, which are summarized for the performance levels of Figure 2 in the materials chapters of FEMA 273. If component demands do not exceed component capacities, the building performance objective are assumed to have been met.