The Evaluation of Median-Rank Regression and Maximum Likelihood Estimation Techniques for a Two-Parameter Weibull Distribution

Abstract

Practitioners frequently model failure times in reliability analysis via the Weibull distribution. Often risk managers must make decisions after only a few failures. Thus, an important question is how to estimate the parameters of this distribution for extremely small sample sizes and=or highly censored data. This study evaluates two methods: maximum likelihood estimation (MLE) and median-rank regression (MRR). Asymptotically, we know that MLE has superior properties; however, this study seeks to evaluate these two methods for small numbers of failures and high degrees of censoring, where one cannot depend on the asymptotic properties of maximum likelihood estimation. This research is the direct result of a practitioners question at Pratt & Whitney, where they use both methods of estimation. We evaluate the two estimation methods for extremely high censoring cases, focusing on censoring greater than 99%. Such extreme levels of censoring present certain difficulties. For example, the estimated parameters based on MLE follow extremely skewed distributions, even under a log transformation. This article compares the two estimation methods based on parameter estimation and ability to predict future failures. We provide recommendations on which method to use based on sample size, the parameter values, and the degree of censoring present in the data.

Failure statistics for commercial lithium ion batteries: A study of 24 pouch cells

“…In addition, the MRR technique does not take into account the cycle number where the tests ended (593 in this case) when there are suspended tests. This, in addition to other well-known issues with MRR analysis 56,58 , motivates us to avoid the OLS-based analysis and use the maximum likelihood (ML) technique 48,56 . This involves, in effect, searching for the values of (β, η), (β, η, ), or () that maximize the likelihood of the associated pdf, where likelihood is defined as the probability that a given combination of parameters would produce the observed vector of failure times.…”

Application of the Two-parameter Weibull Distribution Method to Assess the Reliability of Gas Turbine Compressors

“…There are several methods of estimating Weibull parameters, such as the maximum likelihood estimation, method of moment and median rank regression. Olteanu and Freeman [25] studied the performance of maximum likelihood estimation (MLE) and median rank regression (MRR) methods and they concluded that the median rank regression method offered the best combination of accuracy and ease of interpretation when the sample size and number of suspensions are small. This method is very popular in industry because fitting can be easily visualized.…”

Joint economic design of production, continuous sampling inspection and preventive maintenance of a deteriorating production system

“…The parameters of Eqs. 14and (15) can be derived from historical data records using techniques such as the maximum likelihood estimation and the median-rank regression methods (Soliman et al, 2006;Olteanu and Freeman, 2010).…”