Correlation effects of ATE transfers: An empirical study

Abstract

This study presents an empirical analysis of the effects on correlation of the site-to-site transport of ATEs. Four correlation scenarios are presented with permutations of calibration/non-calibration of the testers. The results show that the least number of uncorrelated tests are encountered when there is calibration at the reference tester site, with or without calibration on the target site. Also, there is no significant shift in the distribution of tests after the physical transport and the calibration of the testers. Any shift encountered in these cases can be attributed to a number of factors, such as device-specific sensitivity, vulnerability of related tester resources, or degradation of the representative units used per device. Overall, this study demonstrates that non-calibration at the target site, provided there is calibration at the reference tester site, is a viable and credible option for a less time-consuming and more cost-efficient correlation procedure, especially in massive transfer of ATEs.