The objective of Performance-Based Seismic Design (PBSD) is to accurately predict, in definable terms, the performance of a building during any intensity of earthquake ground motion that may occur at the building site over the lifetime, or design life, of the building. Definable performance can be accomplished by designing the building to meet a wide range of performance objectives. A single performance objective consists of a level of performance in terms of damage, coupled with a level of earthquake hazard. As an example, a building may be designed to be at the brink of collapse during an earthquake that is expected to occur once every 2,500 years. Performance-based design offers the building owner the opportunity to make informed decisions about the future performance of the building. These decisions involve the consideration of the social and economic implications of each possible performance objective. The owner can weigh the probability of damage occurring, the cost of repair, and the cost of business interruption losses against the cost of initially designing and constructing the building to a higher level of performance.
The 1989-1990 SEAOC Seismology Committee took the first step in the development of a PBSD methodology for new buildings when it created a Vision 2000 Sub-Committee. This sub-committee was charged with the task of defining the conceptual framework for PBSD. As a demonstration of support for PBSD, this subcommittee was replaced in 1992 with a SEAOC Vision 2000 committee. This committee further defined the research needs and framework for PBSD. The product of this SEAOC effort is the SEAOC Vision 2000 report (OES, 1995). This groundbreaking document outlines for the first time the general procedure for the design of buildings based on multiple performance objectives.
The Vision 2000 matrix of performance objectives is shown in Table 2.1. The earthquake hazard is probabilistically defined and divided into four categories: (1) Frequent, (2) Occasional, (3) Rare, and (4) Very Rare earthquakes. These categories cover a range from 70% to 5% probability of exceed an ce in 50 years. Building performance is separated into four categories: (1) Fully Operational, (2) Operational, (3) Life Safety, and (4) Near Collapse. A description of the general damage expected for each of these performance levels is given in Table 2.2. Damage for each performance level is described for vertical load-cairying elements, lateral load-carrying elements, and architectural elements. Table 2.1 also defines a minimum set of recommended performance objectives. The Basic Objective is recommended for "typical" or nonessential buildings and defines performance objectives that range from Fully Operational performance during a 43-year earthquake and Near Collapse performance during a 970-year earthquake. As the importance of the building and its contents or occupants increases, the building is designed to have less damage. As an example, for safety-critical buildings, such as hospitals, the building is expected to provide Fully Operational performance during a 475-year earthquake and Operational performance during a 970-year earthquake.
The Vision 2000 performance-based design methodology addresses each step of the design process, from the selection of performance objectives to the construction and maintenance of the building. A flowchart illustrating the methodology is shown in Figure 2.1 and a detailed description of each step in the PBSD methodology is provided in Table 2.3. A crucial part of the design process is the Initial Design Phase involving the first three steps of the PBSD methodology. These steps involve the selection of the performance objectives, the development of ground motions, and the conceptual design. These three steps will dictate the scope of the remaining tasks and ultimately, the performance of the building. Selecting the performance objectives is a decision to be made by the building owner, in consultation with the structural engineer and architect, who must consider the economic and social implications of the future performance of the building. The earthquake ground motions developed for the site must consider the selected performance objectives and the location of the building, including the soil type and the seismic characteristics of the surrounding region. During the conceptual design phase, a structural system is selected that is the most efficient and effective in meeting the desired performance goals. Each of the aforementioned tasks will influence the design approach and the verification procedures. For example, a building whose structural system is composed of base isolators will require that design and verification methods account for the nonlinear behavior of the isolators.
An important part of the performance-based design method is the Design Verification process. Three design verification steps are incorporated in the PBSD methodology and are new elements that are not traditionally found in building design procedures. Following the conceptual, preliminary, and final design steps, an analysis must be performed to verify that the building design will meet the selected performance objectives. Depending on the selected performance objectives and the type and configuration of the structural system, the verification process will involve performing linear and/or nonlinear analyses to obtain structural response quantities. These quantities are then measured against acceptable values that have been defined for the selected levels of performance.
The final stages of the PBSD methodology involve assuring the building owner that the building will perform as specified in the performance objectives. The design review stage, which may be performed by the local building department or another agent selected by the building department, ensures that the final building design meets the applicable code requirements and intended performance objectives. For safety critical buildings, such as hospitals, it may be necessary to perform an independent third-party "peer" review of the final design. Once the final design is approved, the building must be constructed as intended by the structural engineer, with the appropriate control measures taken to assure the quality of the construction. In addition, the building department and building owner must manage the usage of the building to prevent loads and modifications that may result in detrimental building performance.
New buildings designed in accordance with current design standards satisfy a limited set of performance objectives. The 1996 Structural Engineers Association of California (SEAOC) Recommended Lateral Force Requirements and Commentary (SEAOC, 1996), also known as the 1996 SEAOC Blue Book, requires buildings to be designed to meet only a single performance objective. The earthquake hazard is quantified as the ground motion that occurs once every 475 years, i.e. 10% probability of exceedance in 50 years. For this level of earthquake hazard, the objective of the building design is to "safeguard against major failures and loss of life, not to limit damage, maintain function, or provide for easy repair" (SEAOC, 1996). Building design using the 1997 Uniform Building Code (UBC) (ICBO, 1997) is almost identical to the recommendations of the SEAOC Blue Book and, therefore, buildings designed in accordance with the 1997 UBC requirements are also designed for a single performance objective.
In 1997, the Building Seismic Safety Council (BSSC) developed the NEHRP Recommended Provisions (BSSC, 1997). The NEHRP Provisions take new building design one step closer to the Vision 2000 PBSD method, by offering a dual-performance objective approach. Buildings designed in accordance with the 1997 NEHRP Provisions are expected to provide life-safety performance during the design earthquake, and for seismic events greater than the design earthquake, there will be a "low likelihood of structural collapse" (BSSC, 1997). Previous editions of the NEHRP Provisions (BSSC, 1994) are similar to the 1996 SEAOC Blue Book and the 1997 UBC in that the design earthquake ground motion is defined as an earthquake with a 10% probability of exceedance in 50 years. It was the opinion of the BSSC that buildings designed in accordance with previous editions of the NEHRP Provisions provide a minimum margin against collapse of 1.5 times the 10% in 50 years design earthquake. Although the margin against collapse is a function of the building type, material, and detailing, it was their opinion, that 1.5 is a conservative estimate of the margin against collapse. Therefore, rather than defining the design earthquake in terms of a uniform likelihood of earthquake occurrence, such as 10% in 50 years, the design earthquake in the 1997 NEHRP Provisions was defined based on a uniform margin against collapse. The design earthquake in the 1997 NEHRP Provisions is defined as an earthquake whose response spectrum is two-thirds (1/1.5) of the Maximum Considered Earthquake (MCE) response spectrum. The MCE is defined by ground motion maps provided along with the 1997
NEHRP Provisions. These maps were developed by combining probabilistic ground motion maps developed by Frankel, et al. (1996) and deterministic ground motion maps, as described in the 1997 NEHRP Provisions (BSSC, 1997). Although higher levels of earthquake ground motion could occur at a building site, for most regions of the United States, the maximum considered earthquake is defined as a 2475-year earthquake, which is an earthquake with a 2% probability of being exceeded in 50 years. Since the intent of the provisions is to have a low-likelihood of collapse during seismic events greater than the design earthquake, the BSSC felt that this definition of the maximum considered earthquake provided an adequate margin against collapse and an economically practical design.
Recently, the PBSD methodology has been applied to the evaluation, retrofit, and rehabilitation of existing buildings in a framework that captures the spirit of the Vision 2000 PBSD methodology. The first of these documents was published in 1996 by the Applied Technology Council (ATC), under contract from the California Seismic Safety Commission (CSSC), as a report entitled Seismic Evaluation and Retrofit of Concrete Buildings (ATC, 1996), also referred to as the ATC-40 report. The funding for this report was part of California Proposition 122 that was passed in 1990 and established the Earthquake Safety and Public Buildings Rehabilitation Fund of 1990. The report is focused primarily at the number of pre-1970's California government buildings that are of nonductile concrete construction and is just one of the products proposed by the 1991
Seismic Safety Commission (SSC) report, Breaking the Pattern: A Research and Development Plan to Improve Seismic Retrofit Practices for Government Buildings (CSSC, 1991). The basis for the development of the methodology and commentary in the ATC-40 document is provided in SSC reports 94-02 and 94-01 (CSSC, 1994a; CSSC, 1994b). The methodology presented in the ATC-40 report is cast in a framework that offers the building owner and design engineer the flexibility to select multiple performance objectives for the building evaluation.
The ATC-40 matrix of commonly selected performance objectives, shown in Table 2.4, is similar to the Vision 2000 matrix. The notable differences between the ATC-40 and Vision 2000 performance objectives, include the absence of the Frequent (43-year) earthquake level in the ATC-40 matrix, and the difference in performance level names, such as Collapse Prevention (Vision 2000) and Structural Stability (ATC-40). A general description of the damage expected for each of the ATC-40 performance levels is given in Table 2.5. Damage is categorized according to structural and non-structural damage for each performance level. As in Vision 2000, ATC-40 defines a level of minimum performance to be investigated for all buildings. This is termed the Basic Safety Objective and is indicated in Table 2.4. Buildings satisfying the Basic Safety Objective will provide Life Safety performance during a 475-year earthquake and Structural Stability performance during a 970-year earthquake. Note that the Maximum earthquake hazard level in the ATC-40 report is defined as an earthquake with a 5% probability of exceedance in 50 years, which is different from the MCE defined in the 1997 NEHRP Provisions.
In 1997, a second document implementing a PBSD methodology for the evaluation, retrofit, and rehabilitation of existing buildings was published by the Federal Emergency Management Agency (FEMA). This document and its supporting commentary is entitled NEHRP Guidelines for the Seismic Rehabilitation of Existing Buildings (ATC, 1997), and is referred to as the FEMA 273 report The work in this report involved the American Society of Civil Engineers (ASCE) and ATC under project management provided by the Building Seismic Safety Council (BSSC). The Development of Guidelines for Seismic Rehabilitation of Buildings, Phase I: Issues Identification and Resolution (ATC, 1992) and the Proceedings of the Workshop To Resolve Seismic Rehabilitation Sub-Issues (ATC, 1993) documents provided recommendations and direction for the FEMA 273 report. FEMA 273 covers a broad range of building materials, including steel, concrete, masonry, and wood construction. The predecessor of the FEMA 273 report, the NEHRP Handbook for the Seismic Evaluation of Existing Building (BSSC, 1992), referred to as the FEMA 178 report, was developed by ATC in 1992 to evaluate existing buildings. However, the FEMA 178 report considers only a single-performance objective, which is life-safety performance for a 10% in SO years earthquake.
The matrix of available performance objectives presented in FEMA 273 and the recommended Basic Safety Objective are shown in Table 2.6. Buildings satisfying the Basic Safety Objective will provide Life Safety performance during the Basic Safety Earthquake 1 (BSE-1), which is approximately a 475-year earthquake, and Collapse Prevention performance during the Basic Safety Earthquake 2 (BSE-2), which is approximately a 2475-year earthquake. Note that the BSE-2 earthquake is the same as the MCE earthquake defined in the 1997 NEHRP Provisions. The BSE-1 earthquake is the same as the design earthquake in the 1997 NEHRP Provisions, which is defined as two-thirds of the BSE-2 earthquake. A description of the general and non-structural damage expected for each of the FEMA 273 performance levels is given in Table 2.7.
FEMA 273 and ATC-40 offer the building owner and design engineer the opportunity to select multiple performance objectives for the seismic rehabilitation of a building. An obvious discrepancy between ATC-40 and FEMA 273 is the definition of the Basic Safety Objective. The ATC-40 document recommends the structure remain stable during a 5% in 50 year earthquake, while the FEMA 273 report recommends structural stability during a 2% in 50 year earthquake. The difference between 2% and 5% seems small, however, the building demands calculated using a 2% in 50 year earthquake can be very different from the demands calculated using a 5% in 50 year earthquake. The difference between the two hazard levels can be appreciated by comparing the corresponding return periods. Table 2.8 lists the earthquake hazard definitions from Vision 2000, ATC-40, and FEMA 273. For each earthquake hazard definition, the table shows the return period and the probability of exceedance in 50 and 100 years. Clearly, there needs to be a consensus opinion developed, if possible, that defines performance levels, earthquake hazard levels, and the Basic Safety Objective, for both new design and rehabilitation of existing buildings.
ATC-40 and FEMA 273 represent significant steps towards implementing the PBSD method. However, these reports agree that rehabilitation of a building according to either set of guidelines does not guarantee satisfactory performance. These documents do not address the randomness in the variables affecting demands and capacities, and the uncertainty in the analysis methods used to estimate demands and capacities. Therefore, the building owner and structural engineer do not have a means of assessing the real expectation that the design will satisfy the performance objectives. In addition, the effect of uncertainty on building performance has not been expressed in quantifiable terms and thus, these guidelines do not provide a framework for addressing and reducing uncertainties.
Was this article helpful?