Improved efficient, nearly orthogonal, nearly balanced mixed designs

Designed experiments are powerful ways to gain insights into the behavior of complex simulation models. In recent years, many new designs have been created to address the large number of factors and complex response surfaces that often arise in simulation studies, but handling discrete-valued or qualitative factors remains problematic. We proposed a framework for generating, with a (given) limited number of design points n, a design which is nearly orthogonal and also nearly balanced for any mix of factor types (categorical, numerical discrete, and numerical continuous) and/or mix of factor levels. Our approach can be used to create designs with low maximum absolute pairwise correlation, low imbalance level, and high D-optimality for simulation problems with mixed factor types. Our mixed designs are much more efficient than existing alternatives.