Title: Projected mixture designs based on weighing matrices

Authors: M.L. Aggarwal; Poonam Singh; Vandana Sarin; Rashmi Goel

Addresses: Department of Mathematical Sciences, The University of Memphis, Memphis, TN 38152, USA ' Department of Statistics, University of Delhi, Delhi 110007, India ' Department of Statistics, Kirori Mal College, University of Delhi, Delhi 110007, India ' Department of Statistics, Kirori Mal College, University of Delhi, Delhi 110007, India

Abstract: Prescott (2000) has used augmented pair designs for the projection of response surface designs onto the mixture space and obtained orthogonally blocked designs. Singh (2003) used the Darroch and Waller quadratic model to construct orthogonal designs in two blocks for three and four components. In this paper, we have used quantitative screening designs based on weighing matrices to obtain mixture designs by the method of projection. These mixture designs have been further compared on the basis of uniformity and optimality criteria. Orthogonal blocking of the projected mixture designs has also been presented.

Keywords: mixture experiments; Darroch and Waller model; weighing matrices; projection designs; quantitative screening designs; QSDs; uniformity; optimality; orthogonal blocking.

DOI: 10.1504/IJEDPO.2017.091859

International Journal of Experimental Design and Process Optimisation, 2017 Vol.5 No.4, pp.285 - 300

Received: 30 Oct 2017
Accepted: 19 Mar 2018

Published online: 18 May 2018 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article