Chinmayi Kanthila optimizes building occupancy to reduce energy consumption by 40%

Chinmayi Kanthila, from the National Polytechnic Institute of Brittany, has developed machine learning methods to optimize building occupancy, a crucial issue in reducing global energy consumption, which accounts for 40%. This thesis, defended in 2024, proposes innovative predictive models to improve energy efficiency in the context of climate change.

Chinmayi Kanthila's research at the National Polytechnic Institute of Brittany focuses on optimizing building occupancy, a major societal and environmental issue. Effective occupancy management is essential to reduce energy consumption and improve user comfort, particularly in a context of limited resources and climate change.

By integrating machine learning techniques, including LSTM and Bi-LSTM neural networks, the thesis has modeled and predicted occupancy behaviors. These methods, adapted to the fluctuating dynamics of occupancy, offer solutions to better manage occupancy variations influenced by various factors such as work schedules and weather conditions.

The implications of this research are significant for decision-makers and building managers. By adopting these predictive models, it is possible to improve energy efficiency and reduce costs while meeting the growing demand for technological solutions in the building sector, which represents a strategic investment opportunity.

« « The energy consumption of buildings accounts for about 40% of global consumption, highlighting the urgency to act to optimize their occupancy, » says Chinmayi Kanthila. »

« « Access to innovative solutions like those developed by Chinmayi Kanthila is essential to inform decision-makers and promote concrete actions on energy efficiency, » emphasizes the Lektaris team. »