Model clone detection has predominantly gained momentum in the field of software development process and model driven engineering. Reuse and mutations of the existing models will increase the complexity of managing the software’s organisation’s internal repositories. Detecting semantic and structural similarities among the models finds various uses like fault prediction, estimation of maintenance and refactoring. Literature in the domain witnesses very few works focussing on model clone detection. This work proposes a model clone detection technique by leveraging the statistical and lexical properties of the UML diagrams. The primary contribution of the work is the construction of Similarity Measure (SM) by analysing the design metrics from different perspectives namelyi)measuring the statistical variability between the models and ii) estimating the lexical similarity among design metrics. The results of the model indicates that the proposed method was able to detect the semantically similar model clones of the banking use case. Also, the method is very robust and computationally inexpensive, so that it could find its applicability in all fields where model driven engineering is used.