Perceptual Robust Design

The research presented in this PhD thesis has focused on a perceptual approach to robust design. The results of the research and the original contribution to knowledge is a preliminary framework for understanding, positioning, and applying perceptual robust design. Product quality is a topic that have received much attention from the literature, and for good reason.

Defining quality as the ability to fulfil product requirements, the cost of non-quality does not only cause poor product performance, it also imposes huge direct costs for companies. Prevention, appraisal, and product failures are just a few examples of cost drivers. In general, the earlier quality issues are addressed the less costs they impose.

Robust design methodology seeks to anticipate many of these quality issues by making product designs less sensitive to variation. The approach was first introduced by Genichi Taguchi in the 1980s and has since been expanded and refined. In more recent contributions, the notion of visual robustness has been introduced to the field of design research.

However, contributions have only addressed the visual domain and no underlying theory on which to position or understand these studies have been presented. Therefore, this study set out to contribute to the understanding and application of perceptual robust design. To achieve this, a state-of-the-art and current practice review was performed.

From the review two main research problems were identified. Firstly, a lack of tools for effectively communicating robustness information as part of product requirements. And secondly, the need for a framework to understand, position, and apply perceptual robust design. The first research problem was addressed with the introduction of the robust design requirements specification method.

The method merges quality loss functions, a well-established robust design tool, with requirements development. For preliminary validation of the applicability and usefulness of the method three case study examples were presented revealing a promising potential. The second research problem was addressed with the introduction of the perceptual robust design framework that merged robust design methodology with Psychophysics theory.

To evaluate the applicability and usefulness of the framework a case study was performed showing that product requirements could be loosened by up to 14.74%. However, the optimum for perceptual robustness was found to overlap with the optimum for functional robustness and at most approximately 2.2% out of the 14.74% could be ascribed solely to the perceptual robustness optimisation.

In conclusion, the thesis have offered a new perspective on robust design by merging robust design methodology with theory from relevant scientific fields. Furthermore, this new perspective has been operationalised through a preliminary framework for understanding, positioning, and applying perceptual robust design.