Non-Linear Constraints in
Cornerstone Design of Experiments

The constraint editor of Cornerstone Experimental Design only allows for linear constraints: the involved factors may get a pre factor (coecient) and the resulting weighted sum can be constrained by a certain limit. This can be further restricted to certain levels of categorical factors. General
constraints involving nonlinear functions like log(), exp(), sin(), tan() and even simpler ones like the product of 2 factors are not supported by the editor. As an example we take three factors named d, h2 and alpha. From engineering considerations the term h2 + 3.5 * tan(alpha) + d / (2 * cos(alpha)) has to be restricted to a range between 15 and 35. This is a clear nonlinear constraint that cannot be imposed with the standard editor.