Abstract
Smart buildings enabled by sensing and control technologies offer opportunities to optimize indoor environmental quality (IEQ) while reducing energy consumption. However, most existing optimization frameworks neglect occupancy dynamics which leads to unnecessary energy consumption and inconsistent comfort. This study proposed a Multi-Trial Vector based African Vulture Optimization Algorithm (MTV-AVOA) with occupancy driven constraint handling mechanism, designed to jointly optimize temperature, humidity, illumination, and air quality. A 528-hour smart office dataset was used to benchmark MTV-AVOA against four state of the art algorithms. The obtained results show that MTV-AVOA delivers average comfort index value during occupied hours is 0.8026. while during non-occupied hours it is 0.750. It also significantly improves from the non-optimized case (0.665) and yields the most balanced overall outcome. The proposed algorithm achieves lower energy consumption during both occupied (638.10 kWh) and non-occupied (528.20 kWh) compared to benchmark algorithms. These evaluation demonstrate that by linking occupancy with IEQ optimization, MTV-AVOA offers an effective balance between comfort and energy efficiency than existing approaches, thus contributing to the development of sustainable smart buildings.