The focus of this thesis is the performance modelling of shading devices with regard to daylight illumination and thermal effect in non-domestic buildings - offices, classrooms based on real time-varying conditions. The research is centred in quantifying the impact of external shading devices on energy demand for electric lighting, heating and cooling. The Unix-based Radiance lighting simulation program as the engine, and a nodal thermal model processing irradiation inputs, have been used. An adaptation of suitable calculation approaches and the development of custom-written data analysis programs have been also performed on this behalf. The thesis begins with a literature. review of current practice in modelling buildings particularly examining the daylight and thermal modelling approaches used. Daylight assessment tools and shading systems are also looked at. A formulation of a novel hybrid dynamic lighting thermal model using the daylight coefficient approach and Radiance, combined with a simple thermal model has been developed. A range of external shading devices are reviewed and computer models developed in order to simulate their lighting and thermal performance. These models use, as input, real time-varying meteorological data and hence predict the internal illuminance distribution and incoming irradiation through the window-shading device system. Automated models obtaining heating and cooling loads from possible scenarios where thermal loads are linked to daylight-linked electric controls are used. The validity of the use of these modelling programs in combination is compared against an existing validated computer model. The results of energy consumption for electric lighting, heating and cooling are organized and presented to easily compare and rate the performance of different shading options, facade orientation and climates. The results show a significant influence of shading device design upon some visual environment comfort parameters as well as upon energy consumption for electric lighting. As regards of thermal loads, it is shown that the influence of shading can be considerable if coupled with control switching strategies. It is expected that this study and the methodologies proposed should be useful to architects, building designers and developers, particularly those requiring research based on precise modelling techniques or parametric studies. It may also be of use to developers of such as shading devices, since it can assist them in their research on improving the characteristics of their products. Abstract