A Flexible Odd Exponentiated Half-Logistic Exponentiated Exponential Model for Lifetime and Reliability Data: Applications to Real-Life Datasets

Flexible probability distributions are important for modelling real-life data exhibiting skewness, kurtosis and diverse distributional characteristics. This study investigates the applicability of the Odd Exponentiated Half-Logistic Exponentiated Exponential (OEHL-EE) distribution, a four-parameter extension of the Exponentiated Exponential distribution constructed using the Odd Exponentiated Half-Logistic-G generator. Although the theoretical properties of the OEHL-EE distribution have been established, its practical performance is examined in this study through applications to four real-life datasets. The datasets comprise monthly inflation rates in Nigeria from January 2003 to June 2023, fatigue-fracture lifetimes of Kevlar 373/epoxy specimens, the Gross and Clark dataset, and turbocharger failure times. The proposed model is compared with several competing distributions using maximum likelihood estimation and a collection of model-selection and goodness-of-fit criteria, including the negative log-likelihood, Akaike Information Criterion, Consistent Akaike Information Criterion, Hannan–Quinn Information Criterion, Bayesian Information Criterion, Anderson–Darling, Cramér–von Mises and Kolmogorov–Smirnov statistics. The results reported in the study show that the OEHL-EE model provides a close fit to the four datasets and generally records favourable goodness-of-fit measures relative to the competing models considered. Graphical comparisons based on fitted density and cumulative distribution functions further demonstrate the ability of the proposed model to describe the empirical behaviour of the datasets. The findings illustrate the practical usefulness of the OEHL-EE distribution for modelling skewed lifetime, reliability and other positive-valued data.