Minimum Life-Cycle Cost Structural Design Against Natural Hazards

Kang, Y-J.; Wen, Y.K.

Permalink

https://hdl.handle.net/2142/14245 Copy

Description

Title

Minimum Life-Cycle Cost Structural Design Against Natural Hazards

Author(s)

• Kang, Y-J.
• Wen, Y.K.

Issue Date

2000-01

Keyword(s)

• Natural hazards -- Design criteria.
• Life-cycle cost
• Optimal design
• Expected Failure Cost
• Wind load
• Earthquake motions

Abstract

A methodology is developed for determination of design criteria for structures under single and multiple hazards. Emphasis is on consideration of the uncertainties in the hazards and structural capacity, and costs, i.e., initial (construction) cost, costs of consequences of various structural limit-states including deaths and injuries, and discounting of cost over time. Optimal structural strength and target reliability are obtained by minimizing the expected life-cycle cost. Sensitivity of optimal design to important loading and structural parameters is also investigated. It is found that, for a single hazard, the optimal design depends primarily on the limit state consequences (costs), and to a lesser extent on the structural life. For multiple hazards, the optimal design is controlled by the hazard with large uncertainty and severe failure consequences. The methodology is then applied to the design of a 9-story office building against earthquakes in Los Angeles. The seismic hazards are based on USGS data and FEMA 273 provisions. The structural response is evaluated via an equivalent nonlinear SDOF system. Cost estimates are based on findings in recent FEMA reports. The optimal structural strength is found to be higher than the strength according to current design. The application is further extended to design against both winds and earthquakes in Seattle, Charleston and Boston. Both structural and building envelope (glass window) limit states are considered for winds. It is found that one hazard often dominates in the optimal design and uniform reliability against different hazards, a commonly accepted notion, is not required. The method and results are useful in developing next generation codes and standards.

Publisher

University of Illinois Engineering Experiment Station. College of Engineering. University of Illinois at Urbana-Champaign.

Series/Report Name or Number

Civil Engineering Studies SRS-629

Type of Resource

text

Language

en

Permalink

http://hdl.handle.net/2142/14245

Sponsor(s)/Grant Number(s)

National Science Foundation Grant CMS 95-10243

Owning Collections