ng: Reports on Leading Edge Engineering from the 1996 NAE Symposium on Frontiers of Enginee http://www.nap.edu/catalog/5576.html TABLE 2 Performance Objectives for Buildings Earthquake Return period Probability of Condition of standard Exceedance Occupancy Building Frequent 43 50%in30 Fully operational Occasional 50%in50 10% in 50 years Life safety 970 10% in 100 years Near collapse Source: SEAOC(1995) Again considering an average building, four states of damage have been related to four earthquake intensities(Table 2). Expected levels of damage to structural members, architectural elements, mechanical systems, and the building contents also have been defined for each damage condition. durin the design process, the engineer would consider each earthquake level, and check that the calculated structural response is consistent with the expected performance Although the objectives of the performance-based design procedures ar well defined, implementation is not a simple matter of considering a few more load cases during design. The structural engineering community must address a large number of technical issues for which consensus opinions have not emerged. Of primary importance is the development of a clear understanding f what aspects of structural response trigger damage. Building codes have traditionally defined force levels, but damage levels more often are defined related to displacements, and the influence of earthquake duration cannot be lored. Therefore, reliable analytical tools are needed to calculate the distor tion of the structure when it is subjected to various levels of earthquake excitation. Specialized nonlinear analytical models have been used search for more than 20 years, but they are not sufficient to accomplish the objectives of performance-based design. The results of these analyses ar extremely sensitive to the choice of input parameters, most algorithms are limited to modeling two-dimensional response, the influence of nonstructural lements typically is ignored, and the nonlinear analysis models are compu nationally intensive Even if the structural engineering community were able to develop com- prehensive and efficient modeling tools, determination of the earthquake risk at a given location would remain a major concern. Seismologists continue to develop new theories about source mechanisms (Allen, 1995), earthquakes occur along previously unidentified faults, and large-amplitude velocity and displacement pulses have been identified in near-field ground motions(Iwan 1995). In addition, the 1985 Mexico and 1989 Loma Prieta earthquakes have highlighted the influence of local soil conditions on building response Copyright National Academy of Sciences. All rights reservedDevelopment of Performance-Based Seismic Design Procedures 11 Again considering an average building, four states of damage have been related to four earthquake intensities (Table 2). Expected levels of damage to structural members, architectural elements, mechanical systems, and the building contents also have been defined for each damage condition. During the design process, the engineer would consider each earthquake level, and check that the calculated structural response is consistent with the expected performance. Although the objectives of the performance-based design procedures are well defined, implementation is not a simple matter of considering a few more load cases during design. The structural engineering community must address a large number of technical issues for which consensus opinions have not emerged. Of primary importance is the development of a clear understanding of what aspects of structural response trigger damage. Building codes have traditionally defined force levels, but damage levels more often are defined related to displacements, and the influence of earthquake duration cannot be ignored. Therefore, reliable analytical tools are needed to calculate the distor￾tion of the structure when it is subjected to various levels of earthquake excitation. Specialized nonlinear analytical models have been used in re￾search for more than 20 years, but they are not sufficient to accomplish the objectives of performance-based design. The results of these analyses are extremely sensitive to the choice of input parameters, most algorithms are limited to modeling two-dimensional response, the influence of nonstructural elements typically is ignored, and the nonlinear analysis models are compu￾tationally intensive. Even if the structural engineering community were able to develop com￾prehensive and efficient modeling tools, determination of the earthquake risk at a given location would remain a major concern. Seismologists continue to develop new theories about source mechanisms (Allen, 1995), earthquakes occur along previously unidentified faults, and large-amplitude velocity and displacement pulses have been identified in near-field ground motions (Iwan, 1995). In addition, the 1985 Mexico and 1989 Loma Prieta earthquakes have highlighted the influence of local soil conditions on building response. TABLE 2 Performance Objectives for Buildings Earthquake Return Period Probability of Condition of Standard Designation (years) Exceedance Occupancy Building Frequent 43 50% in 30 years Fully operational Occasional 72 50% in 50 years Operational Rare 475 10% in 50 years Life safety Very rare 970 10% in 100 years Near collapse Source: SEAOC (1995). Copyright © National Academy of Sciences. All rights reserved. Frontiers of Engineering: Reports on Leading Edge Engineering from the 1996 NAE Symposium on Frontiers of Engineering http://www.nap.edu/catalog/5576.html