Multicriteria steepest ascent in a design space consisting of both mixture and process variables

Steepest ascent is shown to be a feasible method for problems where two or more responses are to be optimized. With the aid of Pareto optimality the (one response) standard method is adapted for the use of more responses. A special kind of steepest ascent problem involves the presence of both mixture and process variables. This problem is solved by the use of projection designs. The combined method is demonstrated with a simulated problem.