Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Computational Science, Engineering & Technology Series
COMPUTATIONAL METHODS FOR ENGINEERING TECHNOLOGY
Edited by: B.H.V. Topping and P. Iványi
Optimized Design of Computer Experiments: A Review
E. Mysakova, A. Pospisilova and M. Leps
Department of Mechanics, Faculty of Civil Engineering, Czech Technical University in Prague, Czech Republic
E. Mysakova, A. Pospisilova, M. Leps, "Optimized Design of Computer Experiments: A Review", in B.H.V. Topping and P. Iványi, (Editor), "Computational Methods for Engineering Technology", Saxe-Coburg Publications, Stirlingshire, UK, Chapter 13, pp 325-344, 2014. doi:10.4203/csets.35.13
Keywords: design of experiments, space-filling, constrained design spaces, updating, reliability-based design optimization, surrogate models.
This chapter presents the current history of the design of experiments for computer simulations, the so-called design of computer experiments. The methods for designsí optimization with respect to some metrics particularly will be the main topics of this chapter. The starting point is the description of several criteria that are commonly used to compare and optimize design of experiments. Attention is placed on minimax metrics, which describe the largest unsampled space. Then the advances in the design of regular and constrained (mixture) design of experiments are presented. Many methods are used within both research areas, however they can differ in application details and performances obtained. Finally, the design of experiments for adaptive and/or reliability purposes will be investigated employing methodologies from other areas.
purchase the full-text of this chapter (price £20)