Towards a Healthier Built Environment: Improving Indoor Air Quality through Machine Learning and Life Cycle Analysis

Abstract

Indoor air quality (IAQ) is a critical aspect of human health and comfort within built environments. It refers to the quality of air inside and around buildings, which can be contaminated by a range of pollutants, including particulate matter, volatile organic compounds, and carbon dioxide (CO2), among others. The impact of indoor air pollution on human health is well-established, with research consistently linking poor IAQ to respiratory problems, headaches, and other health issues. Moreover, the levels of some pollutants in indoor air can exceed those in outdoor air, highlighting the need to address IAQ with advanced research. Reducing indoor air pollution and improving IAQ is a key global challenge, and various strategies exist for addressing it. Among these, vertical greenery, particularly the use of active living walls (ALWs), is a promising solution for sustainable indoor living practices. While ALWs have been shown to effectively reduce indoor air pollution in laboratory and real-world settings, there are still unresolved questions regarding their optimization and control in day-to-day life. There is emerging evidence that both outdoor and indoor air pollutants have a strong influence upon IAQ so as the functionality of ALW systems. However, to date it has been challenging to gather robust evidence that portrays the hidden relationship between the ambient air parameters, indoor pollutants, and IAQ. Hence, more fundamental and advanced research can help us better understand the sources and impact of indoor air pollution and identify sustainable solutions for improving the performance of ALW systems. The advancement of miniaturized, discrete, and low-cost sensors alongside advancements in machine learning (ML) offers new promise in uncovering these behaviors. Recent progress in artificial intelligence has motivated various efforts to create algorithms utilizing machine learning techniques to establish the intricate relationship between air parameters and air pollutants. As surface dust emission is considered as one of the major sources of particular matters inside buildings, this study aimed to develop best ML model to visualize the complex relationship between surface dust and several meteorological factors which could in turn influence the architecture of the ALW systems inside building as well as the policies regarding IAQ. This relationship is important to consider when designing control systems for ALWs. Building professionals are striving to minimize the impact of pollutants on indoor environmental health, and as carbon dioxide (CO2) has a strong correlation with human presence, it is considered one of the most significant pollutants. Therefore, forecasting future indoor CO2 using artificial intelligence plays a central role in optimizing the ALW systems to keep CO2 level as low as possible, especially during unexpected events like COVID-19. Therefore, this study suggested a feasible method for forecasting indoor CO2 by utilizing a short period of recent environmental data (such as temperature, humidity, and CO2) to train neural network models. Again, as this study aimed to enhance the IAQ using ALW from a broader overview, it aimed to study the ALW systems from the sustainability viewpoint which rarely been discussed in the building research area. This gap presents an excellent opportunity to assess the different ALW systems thoroughly from life cycle perspective. Life cycle assessment is the prevalent way for quantifying environmental consequences connected with all stages of a product's life from extraction of raw materials through final disposal. Overall, this study aims to enhance IAQ using ALW from a broader perspective, using advanced techniques to better understand the relationship between IAQ, pollutants, and meteorological factors. The study proposes practical solutions for optimizing ALW systems while considering sustainability implications. This study's results form a strong basis for future research and advancement of more sustainable ALW systems, aiming to decrease indoor air pollution.

Date of Award	2023
Original language	American English
Awarding Institution	HBKU College of Science and Engineering