1997 年 5 巻 3 号 p. 63-70
A study with the same objective contained in this paper was reported in 1994 (Journal of Quality Engineering Forum, Vol. 2, No. 4). ln the study, parameter design was conducted using sensory-type graded data as a means of evaluation. ln this paper, Mahalanobis Distance was used to analyze the same data collected in the previous study. From this study, it was found that the trend of some factors affecting legibility were quite similar in both methods. For example, the trend of two distributions showing legibility and illegibility based on the significant factors using Mahalanobis Distance is fairly similar to the ones obtained from graded data.