品質工学 Vol. 24(2016) No. 5

Improving the machining precision of a machining center calls for minimization of displacements in the machine geometry, but a variety of thermal factors lead to such displacements in a machining center. A study was made of the feasibility of estimating current displacement from the temperature of the parts involved in the displacement. If the current displacement can be estimated, the machining center geometry can be corrected to compensate for the displacement,thereby improving the machining precision. Another object of the study was to find an optimum set of points at which to measure temperature in the machining center, and to determine the feasibility of reducing costs by using a small set of measurement points. The results of the study indicated that displacement could be estimated by use of the T-method.

Minimizing displacements in the geometry of a machining center is essential for improving its machining precision, but a variety of thermal factors lead to such displacements. In part 1 of an ongoing study aimed at reliable displacement compensation in changing environments, it was confirmed that displacement could be estimated by measuring the temperature of the relevant parts and applying the T-method, but there were large variations in the estimated values; more precise estimates were necessary. In part 2, the study was repeated to look for problems when displacements were estimated under different environmental conditions. The estimated compensation results were worse than in part 1. The study is continuing with a reconsideration of the unit space and the items treated as features.

The results obtained from a quality engineering study are not always persuasive. Sometimes people are convinced by an experiment even though it failed,and sometimes they refuse to believe the results even though the gain of the S/N ratio was reproduced. A survey was therefore carried out concerning the extent to which people were persuaded by the results of studies carried out using quality engineering. Questionnaires conceming several papers that had been published in academic journals were sent out and analyzed by use of the Mahalanobis-Taguchi system to find which factors were deeply related to persuasiveness. As a result of the analysis, it became clear that the S/N-centered approach was highly persuasive. Individual differences in persuasiveness ratings were also found, and it became clear that these differences were related to a person's quality engineering career.