Designing for Robustness

Today many companies are struggling with issues related to unwanted variation. Even though Robust Design Methodology (RDM) and Design for Six Sigma (DFSS) are suitable approaches of counteracting such issues there are few industrial applications documented. Previous research indicates that this phenomenon might be due to a hitherto existing focus on appropriate statistical tools for achieving robustness in the sense of insensitivity to noise factors. This focus may have led to a negligence of the overarching frameworks of RDM and DFSS with their underlying principles. Instead of focusing on the essential principles, there has been a focus on the statistical details of associated tools, such as Design of Experiments. Despite the fact that tools associated with RDM and DFSS are spread in industry there is limited use of these methodologies. This means that the right tools are applied but with no full utilization of their possible capabilities in terms of achieving robustness. The purpose of this thesis is to contribute to the practical realization of the aforementioned frameworks and their underlying principles. This is accomplished by presenting designing for robustness efforts from a broad perspective that is based on principles, practices and tools and by providing some concrete guidelines and examples concerning their application. Concerning the designing of robust products three fundamental principles are emphasized: awareness of variation, insensitivity to noise factors and continuous applicability meaning that efforts are applicable in all phases of design activity. Practices are identified that describe on an activity-based level what needs to be done to put those principles into action. The practices, in turn, are supported by a wider array of tools and concrete guidelines are provided for where and how these tools can be applied for the best results in terms of robustness. A case study from SKF is used to show why and how to introduce DFSS in a company setting. Another case study illustrates the contributions and usefulness of Variation Mode and Effect Analysis as a tool to put the above mentioned principles into action