Title: The development of specifications-based N-type robust design
Authors: Hsin-Li Chan; Byung Rae Cho
Addresses: Department of Industrial Engineering, Clemson University, Clemson, SC 296234, USA ' Department of Industrial Engineering, Clemson University, Clemson, SC 296234, USA
Abstract: Robust design (RD), one of the most frequently used engineering methodologies as a set of process improvement techniques, is devoted to improve product quality, to promote cost-effective strategies, and to reduce product development intervals. Given the fact that most products are subject to screening tests implemented by pre-defined specification limits such as lower and upper specification limits for a nominal-the-best quality characteristic or N-type characteristic, their resulting distributions become truncated. However, this truncated distribution concept with the specification limits has not been fully incorporated into the RD models which are currently available. To fill this research gap, this paper proposes the utilisation of a truncated distribution as a basis to modify an RD model using response surface methodology, referred to as specifications-based N-type RD in this paper. In particular, many pharmaceutical quality characteristics are required to meet the specifications imposed by the US Food and Drug Administration (FDA). Our pharmaceutical manufacturing example clearly demonstrates that the proposed N-type RD models with the concept of the truncated distribution are beneficial, as compared with the other existing RD models.
Keywords: nominal-the-best quality; N-type robust design; truncated normal distribution; truncated distribution; response surface methodology; RSM; optimisation; pharmaceutical industry; specifications.
DOI: 10.1504/IJEDPO.2013.055744
International Journal of Experimental Design and Process Optimisation, 2013 Vol.3 No.3, pp.217 - 244
Received: 06 Mar 2013
Accepted: 24 Mar 2013
Published online: 02 Jul 2014 *