Methodology to Monitor and Estimate Occupancy in Enclosed Spaces Based on Indirect Methods and Artificial Intelligence: A University Classroom as a Case Study

Despite the indoor occupancy monitoring has been studied over the years to reduce energy and to improve users’ comfort, the authors in their works only present a brief explanation of the sensors, algorithms, and models implemented, as well as the results obtained, leaving aside the preview steps related to the sensor selection and deployment. Therefore, this paper proposes a methodology for indoor occupancy monitoring in real-life scenarios in a non-intrusive manner, including data collection aspects, data preprocessing, and models selection. Besides, to evaluate the effectiveness of the methodology proposed, a case study is presented. The experiment was conducted in a classroom at the University of the West of England, Bristol UK deploying four Internet of Things (IoT) environmental sensors to collect the data. The case study showed that the methodology can guide researchers interested in monitoring occupancy in enclosed spaces through non-intrusive sensors and using Artificial Intelligence.