IEEE 2012 Prognostics and System Health Management Conference (PHM-2012 Beijing), Beijing , pp. 1-7, May 23-25,2012

Comparison of Statistical Models for the Lumen Lifetime Distribution of High Power White LEDs

Jiajie Fan, K.C.Yung, Michael Pecht
Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hong Kong, China
Centre for Prognostics and System Health Management, City University of Hong Kong, Hong Kong, China



Compared to conventional light sources, high power white LED (HPWLED) possesses superior benefits in terms of efficiency, power consumption, environmental friendliness, and lifetime. Therefore, the market of HPWLED is growing rapidly in the application of general lighting, LCD- TVs back-lighting, motor vehicle lighting. However, traditional reliability assessment techniques have several limitations on this highly reliable electronic device with little failure during life test. This paper uses the general degradation path model to analyse the lumen maintenance data of HPWLED with two approaches (Approximation approach and Analytical approach). And three statistical models (Weibull, Lognormal, and Normal) were utilized to predict the lu men lifetime of HPWLED and finally the prediction results were verified by the Akaike Information Criterion (AIC). Results show that Weibull model is the best­ fitting one to the "pseudo failure time" data in the approximate approach, however, Lognormal is the most suitable fitting model for the random effect parameter, fI, in analytical approach.


Keywords- High Power White LEDs, Lumen Lifetime Distribution, Statistical Models, Weibull Distribution, Lognormal Distribution, Normal Distribution, Akaike Information Criterion (AIC)

Complete Article available from the publisher and to CALCE consortium members.

© IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

[Home Page] [Articles Page]
Copyright © 2014 by CALCE and the University of Maryland, All Rights Reserved