Due to the success of Six Sigma, its methodology quickly spread backwards through the manufacturing chain into the design phase, which became known as Design for Six Sigma. One of the most well-known tools of Design for Six Sigma is robust design, which is a powerful and cost-effective quality improvement methodology. Although several approaches to robust design have been proposed in the literature, little attention has been given to its use when experiments are subject to constraints. It is often the case that today's complex manufacturing processes exhibit non-standard experimental characteristics. When such situations arise, experimental design techniques traditionally used in robust design may no longer be appropriate, and a logical alternative is the use of computer-generated designs, specifically D-optimal designs. In this paper, we propose an extension to the traditional robust design methodology within the Design for Six Sigma framework that incorporates D-optimal design techniques to facilitate the application of robust design to real-world situations. A numerical example is used to show our proposed approach, and the results determined from our proposed model are compared with that of the traditional robust design approach.