Performance and applicability of low-cost PM sensors to assess global pollution variability through machine learning techniques

Air quality monitoring and analyses became easy and affordable due to emergence of low-cost sensors. Their efforts to improve the monitoring and understanding of region-specific air pollution events have attracted immense global attention. Nevertheless, they show applicability issues due to data reliability and inconsistency caused by reserve testing of performance parameters for better accuracy, selection and deployment of sensors without considering their fitness for the purpose, and area-specific requirements. This paper analyses and evaluates low-cost sensor deployments across lower, middle, and higher-income group of countries, emphasizing variations in pollutant sources, performance parameters, and machine learning approaches for local source categorization. The 3 Key focuses are defined by performance parameters 1) the Performance Index, 2) Sector Sensitivity Ratio, and 3) Data Reliability Indicator, that provide a comprehensive understanding of sensor efficiency in diverse environments. Our findings reveal distinct trends among income group countries. Higher income group countries exhibit the highest Performance Index (0.35), followed by middle (0.33) and lower income group countries (0.27). However, lower income group countries show the highest data reliability indicator for maximum sector contribution (14.26), surpassing high- (11.74) and middle income (10.71) countries. Sector wise, transport (higher income), industry (middle income), and power (low income) demonstrate the highest data reliability based on its indicator. Additionally, it is observed that advanced machine learning algorithms help improve performance parameters, particularly in middle and lower income group countries where pollution variability is higher. These findings underscored disparities in sensor performance and data reliability across diverse income groups. • A methodological approach was developed to evaluate and compare performance parameters of low-cost sensors (LCS) using Performance Index, Sector Sensitivity Ratio, and Data Reliability Indicator. • Particulate matter (PM) from local sources and particle size variations were found to differ across the globe, affecting LCS calibration and data reliability. • Analysis of deployment, measurement techniques and results associated with various low-cost PM sensors across the globe, was performed. • Various machine learning (ML) methods for sensor calibration were evaluated and compared, revealing gaps in current practices regarding LCSs performance and model evaluation.