Visual Six Sigma: Making Data Analysis Lean

6.7. Revising Knowledge

In the Visual Six Sigma Data Analysis Process, the Model Relationships step is followed by Revise Knowledge (see Exhibit 3.29). This is where we identify the best settings for the Hot Xs, visualize the effect of variation in the settings of the Hot Xs, and grapple with the extent to which our conclusions generalize. Having developed models for the four responses and, in the process, identified the Hot Xs, Sean and his team now proceed to the Revise Knowledge step.

6.7.1. Determining Optimal Factor Level Settings

Sean and his teammates are pleased with their four different models for the responses. At this point, their intent is to identify settings of the five Xs that optimize these four Ys. Sean knows that factor settings that optimize one response may, in fact, degrade performance with respect to another response. For this reason, it is important that simultaneous optimization be conducted relative to a sound measure of desirable performance.

In the Analyze Phase, the team defined target values and specification limits for the four Ys, hoping to guarantee acceptable color quality for the anodized parts. Using these targets and specifications as a basis for optimization, Sean guides the team in performing multiple optimization in JMP.

JMP bases multiple optimization on a desirability function. Recall that when Sean entered his responses in the custom design dialog he noted that the goal for each response was to Match Target, and he entered the specification ...