Project Title/ID Number Building Performance and Loss Measures—1222002
Start/End Dates 10/1/02—9/30/03
Project Leader James Beck (Caltech/Faculty)
Team Members Keith Porter (Caltech/Faculty), Rustem Shaikhutdinov (Caltech/Grad Student)
Project goals and objectives

The objectives of this project are to develop and illustrate the damage- and loss-analysis portions of PEER’s PBEE methodology for modern structures with high-value equipment and contents. We are focusing on elucidating and illustrating the analysis of two important aspects of seismic performance—operational failure and life-safety failure—associated with equipment and content damage. We are also examining repair cost and repair duration, although in less detail, and including in our analyses the performance of some architectural elements, probably gypsum wallboard partitions on metal-stud framing.
Role of this project in supporting PEER’s vision

This project contributes to the development of a detailed PEER PBEE evaluation methodology by developing a probabilistic relationship P[DV|DM] using the UC Science Building testbed as an example, where the DVs (loss measures) include operational failure and life-safety failure. The evaluation of P[DV] will be demonstrated by applying the methodology to this testbed. This project will help to identify development needs for the PEER PBEE evaluation methodology.
Methodology employed

The overall methodology is illustrated in Figure 1. The present study includes calculating the damage response (g[DM] from p[DM|EDP] and g[EDP]) and calculating performance (g[DV] from the developed fragility model p[DV|DM] and g[DM]). Here, g denotes mean annual frequency and p denotes probability. We will use the ABV method implemented by Monte Carlo Simulation to produce g[DV].


Figure 1. PEER PBEE methodology
Larger View
Brief description of past year’s accomplishments and more detail on expected Year 6 accomplishments

In Year 5, we completed a sensitivity analysis of the effects of uncertainties on repair costs for the Van Nuys testbed. The uncertainties in each of the four analysis stages in Figure 1 were considered. This study was published as a PEER report and as a journal article in Earthquake Spectra. In year 5, we also formulated the loss model, including the choice of DMs and DVs, for the UC Science Building testbed, while the hazard and structural analyses were completed by others (Somerville and Mosalam, respectively). Also, during year 5 we coordinated with Hutchinson and Makris to specify the necessary component fragility functions (p[DM|EDP]) for laboratory equipment. We are currently waiting for Hutchinson and Makris to deliver these component fragility functions based on their laboratory tests. In year 6, we will complete the development of the PBEE methodology for modern structures with high-value equipment and contents (begun in year 5) and illustrate it using at least one sample laboratory of the UC Science Building testbed. In year 6, we will also coordinate with Deierlein and Miranda to receive the component fragility function(s) for the architectural elements, and we will carry out the damage analysis and loss analysis for the equipment and architectural elements. This project will produce the first elucidation of the DM and DV analysis portions of PEER’s PBEE methodology for this facility type, and be part of the first end-to-end performance of the entire methodology for a real facility, illustrated in a single document. It will help to identify development needs for the methodology.
Other similar work being conducted within and outside PEER and how this project differs

ASCE/FEMA 356 is also concerned with the overall performance of facilities, but its treatment of equipment systems is limited.
Plans for Year 7 if this project is expected to be continued

One plan for Year 7 is to test whether using realistic simulated ground-motion records give the same DVs as Somerville’s suite of real earthquake records. If they do, then in at least some cases, simulated ground motions can be used instead of recorded ground-motion time histories. This would have important implications for future work. For this task, we will repeat the damage and loss calculations based on structural response calculations to be produced by Mosalam using simulated ground motions, and compare the probability distributions p[DV | simulated ground motions] with p[DV | recorded ground motions].
Describe any instances where you are aware that your results have been used in industry

Expected milestones
  • June 30, 2003: First elucidation of the DM and DV analysis portions of PEER’s methodology for this facility type, that is, modern structures with high-value equipment and contents.
  • September 30, 2003: First end-to-end performance of the entire PEER PBEE evaluation methodology for a real facility, the UC Science Building, illustrated in a single document.
Deliverables

Our results will be documented in the combined testbed report by September 30, 2003.