A Systematic Approach to Planning for a Designed Industrial

The use of a typical ie mid level operator and the

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: The use of a “typical” (i.e., “mid-level”) operator and the blocking of the blank forgings by lot number were decisions made for experimental insurance, although neither variable was expected to have important effects. Not that it is not practical or desirable to hold some factors constant. For example, although it might be ideal to have experimental material from only one titanium forging, there may not be enough material within one forging, and forg- INDUSTRIAL ing may interact with experimental variables. The operator’s role in this highly automated process is small, and material properties of the blank titanium forgings are carefully controlled because of the criticality of the part. 6. Measurement precision-how known? Type of cutting fluid Standard Temperature of cutting fluid (degrees F.) Operator lOO-110°F. when machine is warmed up Several operators normally work in the process Material properties may vary from unit to unit Not sure, but thought to be adequate 1-2” F. (estimate) Titanium forgings FACTORS Factors Desired experimental level and allowable range type NUISANCE Processes vary over time. Experimental conditions vary over time. “Identical samples” differ. Some variations are innocuous, some are pernicious. Examples include contamination of process fluids over time, equipment wear, build-up of oxides on tools, and so forth. Some of these can be measured and monitored to at least ensure that they are within limits; others must be assessed subjectively by experts; still others are unmeasured. Nuisance factors are not controlled, and are not of primary interest in this experiment. They differ from held-constant factors in that they cannot be deliberately set to a constant level for all experimental units. If the level can be selected for any experimental unit, however, blocking or randomization might be appropriate. If levels cannot be selected (i.e., the levels of the factor are unpredictable, perhaps continuous), then the nuisance factor becomes a covariate in t...
View Full Document

Ask a homework question - tutors are online