Providing an energy efficient environment to the occupants of the residential buildings is an interesting area of research. In the literature a number of techniques have been proposed for energy management, but the trade-off between users comfort index and energy consumption is still a challenge and unsolved. Previously we have proposed PSO based power control methodology. Our technique achieved good performance up-to some extent. In this paper, we propose an improved optimized power control methodology for occupants comfort index, energy saving and energy prediction using genetic algorithm (GA). Our proposed GA based optimized technique improved the occupants comfort index and consumed minimum power as compare to our previous work. Here our focus is to increase occupants comfort index, minimize energy consumption and comparison of power consumption using GA and PSO based predicted systems. GA based predicted system consumed less power as compare to its counterpart PSO based predicted system. The output and comparative results show the efficiency of the proposed method in increasing the occupant’s comfort index and minimizing energy consumption.