< Back to previous page

Project

MARS Designs: Enumerating and Exploring A New Family of Response Surface Designs:

Gaining competitive advantage requires today’s businesses and industries to innovate and improve at an increasing speed. Modern products, production processes and services are more complex than ever, so optimizing product formulations/compositions, production process settings and services requires solving high-dimensional decision problems. An important management science tool to cope with this task is design of experiments. A design is a plan to systematically explore alternative product formulations/compositions, process settings or service offerings, identify the factors that drive their performance/quality, and determine the optimal settings for these factors. Response surface designs are commonly used to study the impact of quantitative factors in such scenarios. However, for larger numbers of factors, these designs involve too many experimental tests and therefore are too costly. For this reason, we developed a new family of response surface designs, involving fewer runs, but with the same kind of attractive statistical properties. We call the new designs minimally aliased response surface (MARS) designs. In this project, we will extend our current proof-of-concept catalog of MARS designs, such that we can tackle the most common complications in experimentation for innovating products, processes and services. Success in this project will advance the state of the art in design of experiments and fast-track innovation and quality improvement through better experiments.

Date:1 Jan 2019 →  31 Dec 2022
Keywords:Management science
Disciplines:Biostatistics