This thesis aims to analyze energy savings in environments with unpredictable occupancy rates, presenting a comprehensive analysis of thermal modeling for an office space. This goal is pursued through the use of people-counting sensors, integrated with a control system designed to manage the HVAC terminals. Initially, the current context is analyzed, providing an overview of the main occupancy-based control systems and detection technologies, including commonly used people-counting sensors. Following that, the startup Dropper is introduced along with its innovative solutions for people counting, which play a crucial role in data collection for the developed model. The next phase focuses on the thermal modeling of the building.