Search

A variety of factors can affect a person's perception of their environment and health, but one factor that is often overlooked in indoor settings is the air quality. To address this gap, we develop and evaluate four Machine Learning (ML) models on two disparate datasets using Indoor Air Quality (IAQ) parameters as primary features and components of self-reported IAQ satisfaction and sleep quality as target variables. In each case, we compare models to each other as well as to a simple model that always predicts the majority outcome. In the first analysis, we use open-source data collected from 93 California residences to predict occupant's satisfaction with their indoor environment. Results indicate building ventilation rate, Relative Humidity (RH), and formaldehyde are most influential when predicting IAQ perception and do so with an accuracy greater than the simplified model. The second analysis uses IAQ data gathered from a field study we conducted with 20 participants over 11 weeks to train similar models. We obtain accuracy and F1 scores similar to the simplified model where PM2.5 and TVOCs represent the most important predictors. Our results underscore the ability of IAQ to affect a person's perception of their built environment and health and highlight the utility of ML models to explore the strength of these relationships. Implications: The results from this study show that two outcome variables – occupant's indoor air quality (IAQ) satisfaction and perceived sleep quality – are related to the measured IAQ parameters but not heavily influenced by typical values measured in apartments and homes. This study highlights the ability of machine learning models as exploratory analysis tools to determine underlying relationships within and across datasets in addition to understanding the importance of certain features on the outcome variable. We compare four different models and find that the random forest classifier has the best performance in both analysis on IAQ satisfaction and perceived sleep quality. It is a suitable model for predicting IAQ-related subjective metrics and also provides value insight into the feature importance of the IAQ parameters. The accuracy of any of these machine learning models in predicting occupants' comfort or sleep quality is limited by the dataset size, how data is collected, and range of data. This study identifies the factors that are important to IAQ perception: ventilation rate, relative humidity, and concentrations of formaldehyde, NO2, and particulate matter. It indicates that sensors that can measure these variables are necessary for future, related studies that model occupants' IAQ satisfaction. However, this study does not find strong relationships between any of the IAQ parameters measured in this study and perceived sleep quality despite the logical pathway between these many pollutants and respiratory issues. A prediction model of IAQ perception or sleep quality can be integrated into home management systems to automatically adjust building operations such as ventilation rates in smart buildings. Once buildings are equipped with a network of low-cost sensors that measure concentrations of pollutants and operating conditions of the ventilation system, the prediction model can be used to predict the occupants' comfort and facilitate the control of the ventilation system. [ABSTRACT FROM AUTHOR]

Wildfires have become an important source of particulate matter (PM2.5 < 2.5-µm diameter), leading to unhealthy air quality index occurrences in the western United States. Since people mainly shelter indoors during wildfire smoke events, the infiltration of wildfire PM2.5 into indoor environments is a key determinant of human exposure and is potentially controllable with appropriate awareness, infrastructure investment, and public education. Using time-resolved observations outside and inside more than 1,400 buildings from the crowdsourced PurpleAir sensor network in California, we found that the geometric mean infiltration ratios (indoor PM2.5 of outdoor origin/outdoor PM2.5) were reduced from 0.4 during non-fire days to 0.2 during wildfire days. Even with reduced infiltration, the mean indoor concentration of PM2.5 nearly tripled during wildfire events, with a lower infiltration in newer buildings and those utilizing air conditioning or filtration. [ABSTRACT FROM AUTHOR]

Exposure to cleaning products has been associated with harm to the respiratory system, neurotoxicity, harm to the reproductive system, and elevated risk of cancer, with greatest adverse impacts for workers exposed in an occupational setting. Social and consumer interest in cleaning products that are safer for health created a market category of "green" products defined here as products advertised as healthier, non-toxic, or free from harmful chemicals as well as products with a third-party certification for safety or environmental features. In the present study we examined the air quality impacts of cleaning products and air fresheners, measuring the number, concentrations, and emission factors of volatile organic compounds (VOCs) in an air chamber following product application. Across seven common product categories, 30 products were tested overall including 14 conventional, 9 identified as "green" with fragrance, and 7 identified as "green" and fragrance-free. A total of 530 unique VOCs were quantified with 205 additional VOCs detected below the limits of quantification. Of the quantifiable VOCs, 193 were considered hazardous according to either the California's Department of Toxic Substances Control Candidate Chemicals List or the European Chemical Agency's Classification and Labeling Inventory. The total concentration of VOCs and total emission factors across all products with detections ranged from below limits of detection to 18,708 μg/m3, 38,035 μg/g product and 3803 μg/application. Greater total concentration, total emission factors, and numbers of VOCs were generally observed in conventional cleaning products compared to products identified as "green", particularly compared to fragrance-free products. A hazard index approach was utilized to assess relative risk from measured VOC emissions. The five products with the highest hazard indices were conventional products with emissions of 2-butoxyethanol, isopropanol, toluene and chloroform. Overall, this analysis suggests that the use of "green" cleaning products, especially fragrance-free products, may reduce exposure to VOC emissions. [Display omitted] • Air chamber testing of 28 cleaning products and 2 air fresheners detected 530 VOCs. • The number and concentrations of VOCs were lower in "green" products. • Fragrance-free products had significantly lower numbers and concentrations of VOCs. • Exposure limits were used to calculate relative risk for each product. [ABSTRACT FROM AUTHOR]

Paired PM 10 and fine particulate matter (PM 2.5) samples were collected indoor and outdoor at a fully mechanically ventilated high school in the ultraindustrialized ship channel region of Houston, Texas over a two-month period. The primary foci of this research are to measure the elemental composition of ambient outdoor and indoor coarse particulate matter (PM 10-2.5), characterize aerosol infiltration/ventilation, and obtain clues to their origins. Several metallic hazardous air pollutants were detected inside the school but at concentrations lower than even the most rigorous indoor air quality benchmarks. Only chromium may have potentially posed a hazard based on indoor air quality standards set by the California Environmental Protection Agency. Indoor-to-outdoor (I/O) mass abundance ratios for some metals (Li, Be, Si, Ti, Cr, Co, Ni, Zn, As, Se, Zr, Cd, Sn, Sb, Hf, and Pb) were significantly greater than 1 in PM 10-2.5 , suggesting corresponding indoor sources in the PM 10-2.5 size range. Some other metals (Al, K, Ca, V, Mn, Fe, Cu, Ga, Rb, Sr, Y, Mo, Ce, Pr, Nd, Sm, Gd, Dy, Yb, and U) in PM 10-2.5 had I/O ratios close to one suggesting they had predominantly outdoor origins. Enrichment factors of several metals including light rare earths (e.g. Cu, Sb, Pb, Mo, Cd, La, Ce, Pr, Nd, and Sm) with respect to local soil were overall the highest both indoors and outdoors, illustrating their anthropogenic origins. Several lines of evidence including light lanthanoid ratios, simultaneous La–Ce-Sm variations (ternary diagrams), and correlations with cracking catalysts indicated that petroleum refineries are principally responsible for the elevated lanthanoid concentrations. Rare earth element analysis revealed shifts between transient catalyst releases from petroleum refining operations and windblown soil resuspension as their dominant sources. • Numerous metals measured in paired coarse PM indoors and outdoors of a school. • All indoor metals were at very low levels and below health standards (except Cr). • Metals traced to petroleum refining, oil combustion, soil/road dust, and vehicles. • Most rare earths and many indoor metals infiltrated/ventilated from outside. • Li, Be, Si, Sc, Ti, Cr, Ni, Zn, As, Se, Zr, Cd, Sn, Sb, Hf, W, Pb, Co had indoor sources. [ABSTRACT FROM AUTHOR]

This study examined the use of zoned ventilation systems using a coupled CONTAM/EnergyPlus model for new California dwellings. Several smart control strategies were developed with a target of halving ventilation-related energy use, largely through reducing dwelling ventilation rates based on zone occupancy. The controls were evaluated based on the annual energy consumption relative to continuously operating non-zoned, code-compliant mechanical ventilation systems. The systems were also evaluated from an indoor air quality perspective using the equivalency approach, where the annual personal concentration of a contaminant for a control strategy is compared to the personal concentration that would have occurred using a continuously operating, non-zoned system. Individual occupant personal concentrations were calculated for the following contaminants of concern: moisture, CO2, particles, and a generic contaminant. Zonal controls that saved energy by reducing outside airflow achieved typical reductions in ventilation-related energy of 10% to 30%, compared to the 7% savings from the unzoned control. However, this was at the expense of increased personal concentrations for some contaminants in most cases. In addition, care is required in the design and evaluation of zonal controls, because control strategies may reduce exposure to some contaminants, while increasing exposure to others. [ABSTRACT FROM AUTHOR]