This article introduces the Marshall-Olkin extended power function (MOEPF) distribution as a generalization of the standard power function distribution. The new distribution has a bathtub shaped hazard rate function. The MOEPF distribution have the beta and power function distribution as special cases. Some statistical and reliability properties of the new distribution were given and the method of maximum likelihood estimates was used to estimate the model parameters. The relevance and exibility of the MOEPF distribution was demonstrated with two di erent real and uncensored lifetime data sets. The goodness of ts of the distribution was assessed via the p − value criterion. All the three parameters of tted MOEPF distribution were found statistically signi cant based on their corresponding p − values (p − value = 2.20 × 10−16 for each of the three parameters). The MOEPF distribution is therefore recommended for e ective modelling of lifetime data.