Optimum design method for a steel frame considering prior information on parameters using Bayesian Information Criterion

Makoto Yamakawa, Yoshikazu Araki, Yoshihisa Nakagawa, Koji Uetani

Research output: Contribution to journalArticle

Abstract

In this paper, we present a method for controlling the distribution of optimal member sections for a steel frame. The smoothing parameter is selected by using Bayesian Information Criterion, which is called ABIC. The checkerboard problem refers to the formulation of regions of alternating large and small elements ordered in a checkerboard-like fashion. The checkerboard problem can be avoided with the presented method. The validity of the method is demonstrated by numerical examples of the plane steel frame model.

Original languageEnglish
Pages (from-to)2021-2028
Number of pages8
JournalJournal of Structural and Construction Engineering
Volume74
Issue number645
DOIs
Publication statusPublished - Nov 2009

    Fingerprint

Keywords

  • Bayesian estimation
  • Information Criterion
  • Optimum design
  • Smoothing
  • Steel frame

Cite this