Data flow analysis rules help ensuring the correct flow of data and identifying the related state issues within the model-based testing practices. Though, studies have produced encouraging results for detecting data flow and states-oriented faults and enormous research work has been carried out in this direction, still optimal results from multiple criteria have not been achieved simultaneously. We are aiming to come up with a comprehensive approach that is able to (1): find the define-usage errors, (2): automatically generate the test data for UML state machines (3): mutate the states and flow with different level of complexity achieving efficient mutation score (4): provide the optimal def-use path complete coverage (5): applicable to all UML diagrams. This work in progress is a first step in this direction and we have validated our approach through an implementation which can run against state diagrams. Results have been tested based on a case study with one of the latest approaches and have proven to be promising and effective.