Multiobjective residential load scheduling in smart homes using an improved artificial hummingbird optimizer with MCDM-based decision support

The increasing imbalance between electricity demand and available generation capacity has intensified the need for effective demand response (DR) strategies in smart grids, particularly within the residential sector. Load scheduling programs (LSPs) have emerged as a key approach to shift flexible household loads from peak periods to off-peak periods, thereby reducing electricity costs and peak demand while maintaining acceptable levels of user comfort. This article proposes a multiobjective residential load scheduling framework that simultaneously minimizes energy cost, peak load demand, and user inconvenience. The optimization problem is solved using an improved multiobjective adaptive artificial hummingbird algorithm, which incorporates enhanced initialization and local search mechanisms to improve exploration and convergence performance. To objectively determine the importance of the optimization criteria, the Method based on the Removal Effects of Criteria is employed to derive criterion weights, while the Combined Compromise Solution method is used to rank the Pareto-optimal solutions and identify the most balanced scheduling strategy. A time-of-use pricing scheme is considered to evaluate the effectiveness of the proposed framework for residential energy management. Simulation results demonstrate that the proposed approach achieves up to a 62% reduction in electricity cost and a 44% reduction in peak load, while maintaining an acceptable level of user inconvenience. Comparative analysis with existing approaches confirms the effectiveness of the proposed hybrid optimization and decision-making framework. The proposed method provides a practical solution for improving DR performance and supporting efficient energy management in smart residential environments.