Quantitative Selection of Variation Reduction Plans
Abstract Quality has been a rallying call in the design and manufacturing world for the last two decades. One way to improve quality is to reduce the impact of manufacturing variation. Variation risk mitigation is challenging especially when a product has multiple quality characteristics and complex production and assembly. It is common wisdom that companies should identify and mitigate the risk associated with variation throughout the design process. As yield problems are identified, they should be mitigated using the most cost effective approach. One approach to variation risk mitigation is variation reduction (VR). VR targets reduction of variation introduced by existing manufacturing processes using tools such as Design of Experiments (DOE) and robust design. Many companies have specialized groups that specialize in these methods. VR teams have the role of improving manufacturing performance; however, these teams are limited in their resources. In addition, no tools exist to quantitatively determine where a VR team’s efforts are most effectively deployed. This paper provides a mathematical and optimization model to best allocate VR resources in a complex product.