Evaluating the effects of indoor air quality on teacher performance using artificial neural network

Abstract

Purpose - Indoor Air Quality has a direct impact on occupant health and productivity. Understanding the effect of Indoor Air Quality (IAQ) in educational buildings is essential in both the design and construction phases for decisionmakers. Hence, it is equally important to recognise and appreciate the influence of design judgements on occupants' performance, especially on teacher and students. Design - This study aims to evaluate the effect of IAQ on teachers' performance. This study would deliver air quality requirements to BIM-led school projects. the methodology of the research approach uses quasi-experiment using questionnaire surveys and physical measurements of indoor air parameters to associate correlation and deduction. A technical college building in Saudi Arabia was used for the case study. The study developed an Artificial Neural Network model to define and predict relationships between teachers' performance and indoor air quality. Findings - This paper highlights a detailed investigation into the impact of indoor air quality via direct parameters (relative humidity, ventilation rates and carbon dioxide) on teacher performance. Research findings also indicate an optimal relative humidity with 65%, ranging between 650 ppm to 750 ppm of CO2, and 0.4m/s ventilation rate. This ratio considered optimum records for both comfort and performance. Originality – This paper focused on teachers’ performance in Saudi Arabia and used Artificial Neural Networks to define and predict the relationship between performance and indoor air quality. There are few studies focusing on teachers’ performance in Saudi Arabia and very few that uses ANN in data analysis.