This paper presents a new finite mixture model using the normal mean-variance
mixture of Birnbaum-Saunders distribution. The proposed model is multimodal with wider
range of skewness and kurtosis, and is beneficial for modelling data with large asymmetric
properties in the various theoretic and applied statistical problems. The maximum likelihood
estimates of the model's parameters are computed iteratively by simple EM-type algorithm.
Proposing a simulation study to illustrate finite sample properties and performance of the obtained
estimators, we demonstrate the usefulness of the new model by analysing a real data example.