Some results on the construction of sliced orthogonal arrays of parallel-flats type
Journal
Communications in Statistics - Theory and Methods
Date Issued
2020
Author(s)
Abstract
Sliced orthogonal arrays are useful combinatorial tools that can be used to construct sliced space-filling designs for planning computer experiments with both quantitative and qualitative input variables. A new class of sliced orthogonal arrays called the sliced orthogonal arrays of parallel-flats type is proposed in this paper to address this practical issue. Specifically, a set of sufficient and necessary conditions is first presented to characterize the required sliced orthogonal arrays. Based on the obtained conditions, some construction methods are then developed to generate the proposed designs.
SDGs
Type
journal article