The power consumption in buildings represent a 30–40% of the final energy usage, which is caused by: HVAC (Heating, Ventilation and Air Conditioning), lighting and appliances with any connection to the power grid. The major challenge is to minimize the power consumption by optimizing the operation of several loads without impact in the customer's comfort. For this purpose, the design of an Intelligent Demand Management using Intelligent Systems is presented in this paper. Furthermore a comparative analysis is carried out to evaluate the power consumption performance of some Demand Side Management (DSM) techniques. In this case Direct Load Control (DLC), Load Priority (LP) and Scheduled Programming (SP) are compared with the proposed approach based on Artificial Neural Networks (ANNs). Experimental testing is performed with the consumption data base. The testing results show that energy savings can be achieved through control of the states of various loads.