In probability and statistics theory, the two-parameter Weibull distribution has been most commonly used in reliability
and life testing, extreme value theory and many other areas due to its distribution with several desirable properties and nice
physical interpretations. Recently, Prabhakar Murthy et al. (2004) have studied two-parameter Weibull models in detail. But
often, the hazard rate of lifetime data has increasing (decreasing) shape, the bathtub shape or upside-down bathtub shape.
Since two-parameter Weibull distribution has monotone hazard rate, it is not a proper statistical model to accommodate
nonmonotone hazard rates. To this end, this has lead to the need to seek generalizations of the two-parameter Weibull
distribution. The exponentiated Weibull (EW) family of distributions was constructed by Mudholkar and Srivastava (1993)
by using exponentiation on the two-parameter Weibull family of distributions. That is, the EW distribution with three
parameters is a generalization of the commonly known two-parameter Weibull distribution. And the EW distribution is
quite adequate for modeling non-monotone failure rates, including the bathtub shaped hazard rate, which are quite common
in reliability and biological studies. It has been shown in the literature that the EW distribution has significantly better fit
than traditional models based on the exponential, gamma, Weibull and log-normal distributions in many cases.