Software smells have negative impacts on the reliability and modifiability of software systems. The smells in architecture design can be cascaded down to the implementation level and cause issues that require much effort to fix. Therefore, early detection of the architecture smells can benefit the overall quality of the software system. This paper presents an integration of methods that formally define the software architecture design towards architecture smell detection. Our approach serves as a framework that allows the architectural structures and behaviours to be formally analysed based on a coherent principle. We evaluated the accuracy and performance of our approach with the models generated from open source projects. The results show that our approach is effective and functions well.