*Future Directions*

The presented work has raised some interesting questions, such as the importance of occupant comfort and its associated economic cost of loss productivity versus utility cost. Further study into this relationship is necessary and can include additional economic and psychological measures, such as the impact of productivity in specific working environments, as well as investigating if integrating user feedback into the control would be beneficial or not. In addition, while the simulations showed the proposed steady-state controller to be successful in reducing the overall cost, verification on the actual building is needed. It would also be of interest to investigate the performance of steady-state predictions versus dynamic predictions and determine exactly how much benefit there is from implementing the more computationally difficult dynamic models of fully implemented MPC.
