Fig. 2

Two factor optimization and exploring the design space. Each axis represents a factor to be optimized for a given process whose output is shown as a contour plot, where each contour line represents a constant response (a, b). Determining the optimum using one-factor-at-a-time (OFAT) method first requires varying one factor (triangles) along the first axis to locate the maximum (red triangle). Once this is determined, a second set of experiments (stars) is performed by varying the second factor until its maximum is reached (red star). While the output has been increased, the true optimum in the space has not been reached (a). Conversely, starting with a widely spaced factorial experimental design allows for the evaluation of both variables in the first experiment (triangles) and statistical analysis can determine the path of steepest ascent (arrow). This is followed by a second set of experiments (stars) which can better locate the maximum output in the design space (red star) (b)