Bayesian Model Assessment: Methods and Case Studies
A Bayesian framework for model assessment, including an analysis of uncertainties due to small sample size, model inexactness and measurement error, was described in a paper by the first author in the previous IFIP Conference . In this paper, we present computational methods for numerical evaluation of the statistical properties of the model parameters. These include efficient integration and simulation methods aimed at computing the marginal distributions and marginal and joint moments of the uncertain parameters. In the second part of the paper, the proposed methods are employed in evaluating several models recommended by the American Concrete Institute for design of concrete structures. To the authors’ knowledge, these applications are first attempts at rigorous evaluation of some widely used models in structural engineering.
Unable to display preview. Download preview PDF.
- Der Kiureghian A. Bayesian analysis of model uncertainty in structural reliability. In Der Kiureghian A. and Thoft-Christensen P., editors, Reliability and Optimization of Structural Systems ‘80, Proceedings of the 3rd IFIP WG7.5 Conference, Berkeley, CA, pages 211–221, Berlin, Germany, 1990. Springer-Verlag.Google Scholar
- Monteiro P.J.M., Moehle J.P., and Tang II.T. Aging effects on mechanical properties of power plant concretes. Technical report, Department of Civil Engineering, University of California at Berkeley, Berkeley, California, USA, 1990Google Scholar
- Der Kiureghian A. Structural component reliability. Lecture notes for: Structural reliability - methods applications, Department of Civil Engineering, University of California at Berkeley, Berkeley, California, USA, 1989.Google Scholar
- Ang A-S. and Tang W.H. Probability Concepts in Engineering Planning and Design. Volume I: Basic Principles. John Wiley Sons, New York, USA, 1975.Google Scholar