In modern building management systems (BMS), thermal comfort and energy efficiency are critical considerations. Automated Heating, Ventilation, and Air Conditioning (HVAC) systems are often utilized to achieve these objectives. However, such systems’ effectiveness depends on the accuracy of temperature readings from the thermostat and AHU sensors. Unfortunately, these sensors do not consider several factors that can impact thermal comfort, such as changes in heat loads or occupancy.

To address these limitations, BMS systems are often augmented with sensors on equipment and sun position monitoring capabilities. However, occupancy detection remains a challenge. Commercial solutions, such as human counting sensors at the entrance, only provide an overall headcount and do not capture occupant positions or activity levels. As a result, HVAC systems may over-cool certain areas or underperform in others, leading to poor energy efficiency.

To overcome these limitations, an AI-based computer vision solution can be developed to detect occupant positioning and activity. By analyzing CCTV footage, this technology can provide real-time data on occupant positions and activities, which can be fed into a digital twin application using APS Viewer and its digital twin extension. This approach can enable HVAC systems to adjust airflow and temperature distribution based on occupant positions, leading to optimal thermal comfort and energy efficiency.

In summary, while modern BMS systems have made significant strides in automating HVAC systems, there is still room for improvement. Incorporating AI-based computer vision to sense occupant positioning and activity levels can lead to more sustainable and comfortable built spaces. This technology has the potential to revolutionize building management systems and may become a critical component of future building automation systems.