Learning curve analysis in total productive maintenance

The continuous improvement concepts such as total quality management, just-in-time and total productive maintenance have been widely recognized as a strategic weapon and successfully implemented in many organizations. In this paper, we focus on the application of total productive maintenance (TPM). A random e3ect non-linear regression model called the Time Constant Model was used to formulate a prediction model for the learning rate in terms of company size, sales, ISO 9000 certi6cation and TPM award year. A two-stage analysis was employed to estimate the parameters. Using the approach of this study, one can determine the appropriate time for checking the performance of implementing total productive maintenance. By comparing the expected overall equipment e3ectiveness (OEE), one can improve the maintenance policy and monitor the progress of OEE. ? 2001 Elsevier Science Ltd. All rights reserved.