Commercial buildings are characterized by high operational costs and high CO2 emission levels due to the high demand for energy. The management of energy demand, also known as Demand-side Management (DSM), allows buildings to make informed decisions about their energy consumption. However, current building management systems control heating, ventilation and air conditioning, lighting and other aspects only by basic scheduling functions. At the same time, Internet of Things (IoT) devices have the capability to transform building management systems into intelligent tools for DSM. This means that buildings equipped with a large variety of sensors and actuators have the potential to become intelligent and significantly reduce energy consumption and, as a consequence, operational cost. However, a small amount of installed suitable infrastructure in offices and a lack of understanding of the benefits of DSM hinders this development. This calls for developing approaches that can efficiently find sequences of actions which, upon execution, reduce energy consumption in an office building. In this context, Artificial Intelligence (AI) planning provides powerful techniques to intelligently plan an office building’s demand by computing effective plans or schedules of device actions. This thesis presents an approach for demand-side management by using AI planning. It involves defining and modeling scenarios from demand-side management as a domain-independent planning problem using the Planning Domain Definition Language (PDDL) and solving the planning problem using an existing AI planner. Additionally, the approach is evaluated regarding performance and potential for savings, both energy-wise and financially. The evaluation provides evidence in favor of DSM being beneficial for saving energy and operational cost.