Development of an Indoor Particulate Matter (PM2.5) Prediction Model for Improving School Indoor Air Quality Environment

Proposal of a Methodology for Prediction of Indoor PM2.5 Concentration Using Sensor-Based Residential Environments Monitoring Data and Time-Divided Multiple Linear Regression Model

“…The RMSE and MAE are used to measure the difference between the measured and predicted values. The equations for these performance indicators are given in Equations ( 8)- (10), respectively [34]:…”

“…Meanwhile, since indoor air quality is affected by various factors, it is necessary to determine the relative contribution between variables affecting indoor PM 2.5 concentration and determine input variables to accurately predict indoor air quality. Thus, previous studies have developed prediction models by choosing ventilation conditions, indoor pollutants, indoor airflow, and pressure that can be directly measured or validated indoors as input variables to increase predictive performance [ 10 , 11 ]. Using input variables obtained through sampling or surveys, such as ventilation rate, indoor pollutant concentration, pressure, and airflow, activity pattern information, may result in high performance [ 12 ].…”

“…Thus, previous studies have developed prediction models by choosing ventilation conditions, indoor pollutants, indoor airflow, and pressure that can be directly measured or validated indoors as input variables to increase predictive performance [ 10 , 11 ]. Using input variables obtained through sampling or surveys, such as ventilation rate, indoor pollutant concentration, pressure, and airflow, activity pattern information, may result in high performance [ 12 ]. However, these variables are difficult to obtain in real-time, making it even more challenging to predict indoor air quality accurately.…”