Research Article Published online 24 May 2010 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/qre.1113 The Mean Time Between Failures for an LCD Panel Fu-Kwun Wang a,b∗ † and Tao-Peng Chu b The calculation of mean time between failures is very important in reliability life data analysis. For different distri- butions, the values of mean time between failures are always different. The two-parameter Weibull distribution is widely used in reliability engineering. However, some distributions may offer a better fit of data. This paper aims to develop an algorithm for determining the best-fitted distribution of a liquid crystal display panel based on the field return data. The two-parameter and three-parameter Weibull distributions and other distributions such as the Burr XII distribution, the Pareto distribution and the Log-logistic distribution are compared to provide a better characterization of the life data which is based on the maximum value of all log-likelihood functions. We also provide a goodness-of-fit test for the best-fitted distribution. It is recommended that the Burr XII distribution could be used to characterize the reliability life of a liquid crystal display panel. Copyright © 2010 John Wiley & Sons, Ltd. Keywords: mean time between failures; LCD Panel; Weibull distribution; Burr XII distribution 1. Introduction T hin film transistor-liquid crystal display (TFT-LCD) has been widely used in a wide variety of consumer electronic devices such as cellular phone, digital camera, LCD-TV, monitor and personal digital assistant (PDA). The development of LCDs began in 1964. A major milestone in the development of LCD products was made in 1988 when an active-matrix (AM) TFT display was demonstrated for application to the display. Compared with conventional displays such as cathode-ray tube displays, LCDs have prominent strengths, including thin thickness, low power consumption and low weight for the same display size. In the 1990s, Japan and Korea began to produce the large sized TFT-LCD. Their global market shares of TFT-LCD reached over 70% in 2000. Taiwan has since then become a rising production source. Since the Asian financial crisis in 1997, Taiwanese companies such as AUO, CMO, CPT and Hannstar have produced more than 40% of the world’s display products. In 2007, the global market of the LCD-monitor exceeded $100 billion US dollars. Korea, Taiwan and Japan now manufacture the majority of TFT-LCD products. The calculation of mean time between failures (MTBF) for an LCD panel is an important task. This kind of analysis can offer customers’ information about the product’s reliability life. In general, the reliability life analysis of an LCD panel can be obtained by two different approaches which are based on an accelerated life testing and censored data from the field return data. In this study, we focus on the field return data which is collected from the vendor’s report of about 2 years. The MTBF of an LCD panel is about 100 000 h at normal condition from an accelerated life test. However, this claim should be verified by the actual life data which is based on the field return data. From the literature of the lifetime distributions, the Weibull distribution has been widely used in reliability engineering 1 . The Burr XII distribution offers a more flexible alternative to the lognormal and the log-logistic distributions 2 . Rodriguez 3 demonstrated that the Burr XII distribution could cover many different distributions such as Weibull, lognormal and logistic. Zimmer et al. 4 showed that the Burr XII distribution could allow for a wide variety of shapes in its density distribution making it useful for fitting many types of data and for approximating many different distributions. Watkins 5 provided an algorithm to exploit the link between the three-parameter Burr XII distribution and the two-parameter Weibull distribution, which emerged as a limiting case of the Burr XII distribution, and also calculated a discriminating value to determine which distribution could provide a better fit to data. Furthermore, Shao 6 observed that the finite maximum likelihood estimates for the three-parameter Burr XII distribution do not always exist, because the distribution tends towards non-degenerated limiting forms as parameters tend to the boundary. For different distributions, the values of MTBF are always different. This paper aims to develop an algorithm for determining the best-fitted distribution of an LCD panel based on the field return data. The remainder of this paper is organized as follows. a Department of Industrial Management, National Taiwan University of Science and Technology, 43 Keelung Road Section 4, Taipei 106, Taiwan b Graduate Institute of Management, National Taiwan University of Science and Technology, 43 Keelung Road Section 4, Taipei 106, Taiwan ∗ Correspondence to: Fu-Kwun Wang, Department of Industrial Management, National Taiwan University of Science and Technology, 43 Keelung Road Section 4, Taipei 106, Taiwan. † E-mail: fukwun@mail.ntust.edu.tw Copyright © 2010 John Wiley & Sons, Ltd. Qual. Reliab. Engng. Int. 2011, 27 203--208 203