Smart grid aim to ensure transferring of the future ‎networks into intelligent ones through the promotion of bi-directional information and active participation from whole inter-connected subsystems.‎ Demand Side Management (DSM) programs play a key role in the future smart grid through intelligently managing the loads. These programs are implemented via residential load management systems for smart cities. In which the power consumption pattern of the household appliances is scheduled to deliver desired benefits i.e. optimizing the ON-OFF cycles of appliances while minimizing end-user electricity costs, reducing the Peak to Average Ratio (PAR), and increasing user comfort. In this study, for fulfilling the previous features, optimized DSMs based on evolutionary techniques, that are genetic algorithm and binary particle swarm, have been proposed for scheduling residential users' appliances. The effectiveness of the suggested DSMs has been verified utilizing MATLAB simulator. According to the obtained results, the introduced methodologies optimally schedule the appliances, resulting in lower electricity bills and PAR