Analysis of Indoor Air Quality in Inpatient Wards using IoT Sensor Data and Machine Learning

Sejong University, Architectural Engineering, 05006 Seoul, Republic of Korea

^* Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

Since COVID-19, interest in Indoor Air Quality (IAQ) has increased and research has been conducted on Airborne Infection Isolation Rooms (AIIRs), but more research is needed on long-term IAQ analysis in hospitals. IoT sensors were used to acquire data from Patient Room (PR) and AIIR, and long- term data is used to analyse IAQ. Clustering was performed on target space. Unlike PR and AIIR, Nurse Station (NS) has a full-time occupancy and CO2 concentrations above 1,000 ppm occurred frequently. For temperature, NS, PR, and AIIR all remained within the 23 to 26 °C range, although AIIR occasionally exceeded 30 °C. For PM2.5, the overall average remains below 20 μg/m³ in NS, PR, and AIIR. However, it remained largely unchanged in NS, but sharp events with concentrations above 100 μg/m³ occurred frequently in PR and AIIR. The overall PM2.5 concentrations were well maintained, but the frequent PM2.5 peak concentrations were very high for short periods of time and need to be further managed. IoT sensors can be utilized to acquire continuous IAQ data in wards. IAQ management is required through continuous measurement using IoT sensors. The large amount of IAQ data acquired can be utilized in various ways by applying machine learning.