This paper introduced and studied a new extension of inverse Weibull distribution with three-parameter named as the X-Gamma inverse Weibull (XGIW) distribution. Reliability and hazard rate properties of this distribution are discussed. Maximum likelihood estimation (MLE), maximum product spacing (MPS) Method of the XGIW distribution parameters are discussed. A numerical study using real data analysis and Monte-Carlo simulation are performed to compare between MLE and MPS methods of estimation. The flexibility and potentiality of the XGIW distribution are examined using two real data sets. The cancer data represents remission times (in months) of a random sample of 128 bladder cancer patients and the second data set of leukemia represents 40 patient suffering from leukemia from one of the Ministry of Health Hospitals in Saudi Arabia. The XGIW model can produce better fits than some well-known distributions as generalized inverse Weibull, Kumaraswamy–inverse Weibull, exponentiated generalized inverse Weibull distribution and inverse Weibull distributions.