Estimation of Reliability Measures for the Hjorth Model Based on Constant-Stress Accelerated Life Tests with Type-II Censoring

Abstract

Accelerated life testing is widely used to assess product reliability under time and cost constraints, particularly when lifetime data are subject to censoring. This paper investigates statistical inference for the Hjorth distribution under constant-stress accelerated life tests with Type-II censoring. The model parameters are estimated using maximum likelihood and maximum product of spacings methods, and asymptotic confidence intervals are constructed. The scale parameter is linked to the stress level through a log-linear relationship, enabling reliability assessment under normal operating conditions. A comprehensive Monte Carlo simulation study is conducted to evaluate the performance of the proposed estimators, and the methodology is illustrated using a real engineering dataset. The results indicate that maximum likelihood estimators provide higher efficiency for parameter estimation, whereas the maximum product of spacings method offers improved performance for reliability measures under normal use conditions. These findings highlight the complementary roles of the two estimation approaches in accelerated life testing analysis.