Graphical User Interfaces (GUI) have become very popular and are used as the primary user interface in many of today’s software. However, the complexity and diversity of today’s GUIs introduce new challenges for effective GUI testing. Responsive GUIs, in particular, have different outcomes and event sequences depending on the device, its orientation, and other environmental factors. This thesis proposes an approach that aims to increase the readability and maintainability of responsive GUI tests and to make them fully automated. This approach utilizes concepts from page object pattern, factory method pattern, the simple factory approach, enhanced input from Unreal Engine, and behaviour-driven development (BDD). A case study was done in conjunction with IVU Traffic Technologies to evaluate the proposed approach in a real world setting. The testing tool Squish was used for this purpose. This approach was tested with two kinds of IVU GUIs that have different event sequences in some of their functionalities. The results demonstrate improved test readability and maintainability, reduced test script complexity and the time for creating the tests. The approach could also automatically recognize the GUI and device being tested and could adjust itself to use specific class methods for the particular GUI events and functions.