DocumentCode
1856393
Title
Analysis of parameter-degradation data using life-data analysis programs
Author
Chan, Chun Kin ; Boulanger, Michele ; Tortorella, Michael
Author_Institution
AT&T Bell Labs., Holmdel, NJ, USA
fYear
1994
fDate
24-27Jan 1994
Firstpage
288
Lastpage
291
Abstract
A new application of existing reliability software packages is proposed to analyze parameter-degradation data. This new implementation of existing tools is applicable to a large class of degradation phenomena described by the power law or the stretched exponential law. When these laws are used to model parameter-degradation paths, the parametric drift can be treated as an explanatory variable, on equal footing with conventional explanatory variables such as temperature, electrical stress, or humidity. Thin film integrated circuit resistor degradation data and the STAR (Statistical Analysis of Reliability) software package are used to illustrate the authors´ approach. Other degradation phenomena that can be studied using the proposed method include hot carrier degradation in metal-oxide-silicon (MOS) transistors, laser degradation, and degradation of light emitting diodes
Keywords
circuit reliability; failure analysis; maximum likelihood estimation; parameter estimation; reliability theory; software packages; statistical analysis; thin film circuits; thin film resistors; MOS transistors; STAR software package; explanatory variable; hot carrier degradation; laser degradation; life data analysis programs; light emitting diodes; metal-oxide-silicon transistors; parameter degradation data; parametric drift; power law; reliability software packages; statistical analysis of reliability software package; stretched exponential law; thin film integrated circuit resistor degradation data; Application software; Data analysis; Degradation; Humidity; Resistors; Software packages; Statistical analysis; Stress; Temperature; Thin film circuits;
fLanguage
English
Publisher
ieee
Conference_Titel
Reliability and Maintainability Symposium, 1994. Proceedings., Annual
Conference_Location
Anaheim, CA
Print_ISBN
0-7803-1786-6
Type
conf
DOI
10.1109/RAMS.1994.291122
Filename
291122
Link To Document