Prophet model for forecasting occupancy presence in indoor spaces using non-intrusive sensors

The Internet of Things is a multi-sensor technology with the unique advantage of supporting non-intrusive and non-device occupancy detection, while also allowing us to explore new forecasting occupancy models. However, forecasting occupancy presence is not a trivial task, since it is still unknown the main criteria in selecting a forecasting modelling approach according to a non-intrusive sensing strategy. Towards this challenge, this paper proposes an analytical workflow developed to support the Prophet model for forecasting occupancy presence in indoor spaces throughout the tasks of sensing, processing, and analysing event triggered data generated from ten nonintrusive sensors, including motion, temperature, luminosity, CO2, TVOC, sound, pressure, accelerometer, gyroscope, and humidity sensors. The usefulness of this analytical workflow is demonstrated with the implementation of an IoT platform for an experiment operating non-intrusive sensing in a classroom. The assessment is made at different time intervals and the results confirm that there is a relationship between the event-count and occupancy presence in such a way that the larger the number of events triggered in an indoor space, the higher the probability of an indoor space being occupied.

[1]  P. Smyth,et al.  Modeling Count Data from Multiple Sensors: A Building Occupancy Model , 2007, 2007 2nd IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing.

[2]  H. Burak Gunay,et al.  Opportunistic occupancy-count estimation using sensor fusion: A case study , 2019, Building and Environment.

[3]  Naveen Kumar Chikkakrishna,et al.  Short-Term Traffic Prediction Using Sarima and FbPROPHET , 2019, 2019 IEEE 16th India Council International Conference (INDICON).

[4]  Sousso Kelouwani,et al.  A comprehensive review of approaches to building occupancy detection , 2020 .

[5]  Benjamin Letham,et al.  Forecasting at Scale , 2018, PeerJ Prepr..

[6]  Lihua Xie,et al.  Building occupancy estimation and detection: A review , 2018, Energy and Buildings.

[7]  Weixi Wang,et al.  A Self-Adaptive AP Selection Algorithm Based on Multiobjective Optimization for Indoor WiFi Positioning , 2021, IEEE Internet of Things Journal.

[8]  Ely Salwana,et al.  Design and Testing of an Epidermal RFID Mechanism in a Smart Indoor Human Tracking System , 2021, IEEE Sensors Journal.

[9]  Günther Retscher,et al.  Location determination using WiFi fingerprinting versus WiFi trilateration , 2007, J. Locat. Based Serv..

[10]  Aris A. Syntetos,et al.  Distributional Assumptions for Parametric Forecasting of Intermittent Demand , 2011 .

[11]  Paul Davidsson,et al.  A comparison of machine learning algorithms for forecasting indoor temperature in smart buildings , 2020, Energy Systems.

[12]  Venkataramana Badarla,et al.  Occupancy detection systems for indoor environments: A survey of approaches and methods , 2020, Indoor and Built Environment.

[13]  Fabio Ciravegna,et al.  A Low Cost Indoor Positioning System Using Bluetooth Low Energy , 2020, IEEE Access.

[14]  A. Florita,et al.  Occupancy sensing in buildings: A review of data analytics approaches , 2019, Energy and Buildings.

[15]  Korra Sathya Babu,et al.  Time Series based Air Pollution Forecasting using SARIMA and Prophet Model , 2019, Proceedings of the 2019 International Conference on Information Technology and Computer Communications - ITCC 2019.

[16]  Hyeun Jun Moon,et al.  Development of an occupancy prediction model using indoor environmental data based on machine learning techniques , 2016 .

[17]  Ravi S. Srinivasan,et al.  From occupancy to occupant behavior: An analytical survey of data acquisition technologies, modeling methodologies and simulation coupling mechanisms for building energy efficiency , 2017 .

[18]  Gierad Laput,et al.  Synthetic Sensors: Towards General-Purpose Sensing , 2017, CHI.

[19]  Hao Jiang,et al.  Non-intrusive occupancy sensing in commercial buildings , 2017 .