2002 Volume 53 Issue 5 Pages 356-362
As a feature of tool lifetime in a wear-out failure mode, it is pointed out that as the cutting speed slows, the lifetime lengthens and the variance becomes larger. In other words, the tool lifetime has a statistical feature that the expectation and variance of the tool lifetime change according to variation in the cutting speed. As a distribution including this feature, the log-normal distribution, which has the correlation of the expectation with the variance, is adopted. In determining whether or not a production lot of tool is accepted based on the mean time to failure (MTTF) of tool lifetime, we wish to provide the most effective sample size and the acceptance criterion for the specified producer and consumer risks. Then, in the design of reliability tests for assuring MTTF in the log-normal distribution, even if the log-nomal distribution is transformed into the normal distribution, the reliability test for assuring MTTF in the log-normal distribution sometimes cannot be designed in the normal distribution. Therefore, the objective of this article is to propose a procedure for designing reliability tests for assuring MTTF in the log-normal distribution employed as the distribution of tool lifetime, by using a chi-square approximation for log-normal distribution. Further, we also consider the optimum cutting speed in testing under the proposed reliability tests as an economical cutting speed in testing.
