Optimal seismic design of steel frame buildings based on life cycle cost considerations

TLDR

The study presents a multi‑objective optimization procedure for designing steel moment‑resisting frame buildings within a performance‑based seismic design framework. The method models life‑cycle cost, construction complexity, and seismic performance as separate objectives, selects wide‑flange steel sections, validates designs against AISC‑LRFD and NEHRP criteria, evaluates drift ratios via a pushover‑derived single‑degree‑of‑freedom model, and uses a genetic algorithm to generate a Pareto‑optimal design set, illustrated on a five‑storey perimeter frame example. The optimization yields a wide range of valid design alternatives, allowing decision makers to choose the most preferred balance of objectives. © 2003 John Wiley & Sons, Ltd.

Abstract

Abstract A multi‐objective optimization procedure is presented for designing steel moment resisting frame buildings within a performance‐based seismic design framework. Life cycle costs are considered by treating the initial material costs and lifetime seismic damage costs as two separate objectives. Practical design/construction complexity, important but difficult to be included in initial cost analysis, is taken into due account by a proposed diversity index as another objective. Structural members are selected from a database of commercially available wide flange steel sections. Current seismic design criteria (AISC‐LRFD seismic provisions and 1997 NEHRP provisions) are used to check the validity of any design alternative. Seismic performance, in terms of the maximum inter‐storey drift ratio, of a code‐verified design is evaluated using an equivalent single‐degree‐of‐freedom system obtained through a static pushover analysis of the original multi‐degree‐of‐freedom frame building. A simple genetic algorithm code is used to find a Pareto optimal design set. A numerical example of designing a five‐storey perimeter steel frame building is provided using the proposed procedure. It is found that a wide range of valid design alternatives exists, from which a decision maker selects the one that balances different objectives in the most preferred way. Copyright © 2003 John Wiley & Sons, Ltd.