|
Abstract:
|
A mixture experiment (ME) involves mixing components to form an end product and measuring a response, which depends on the proportions of the components and maybe other factors (e.g., total amount, process variables). Methods for ME design and analysis have been developed over 50 years. However, challenges often exist for ME problems. One challenge is to decide which of several approaches is best for a given ME problem. Another challenge might be a "twist" in the ME or its data analysis not addressed by existing methods or software. Roundtable participants should come prepared to briefly describe one or two challenges they have faced in ME problems, what they did to address the challenges, and whether new methods or software capabilities are needed. The opportunity for discussion and feedback from other participants should make this an interesting and useful roundtable.
|
