Deep Reinforcement Learning for Urban Air Quality Management: Multi-Objective Optimization of Pollution Mitigation Booth Placement in Metropolitan Environments

TL;DR

The study addresses urban air pollution by optimizing the placement of air-purifying booths through a deep reinforcement learning framework based on Proximal Policy Optimization (PPO). It constructs a high-resolution, multi-channel grid (e.g., $50 \times 50 \times 6$) that encodes AQI and contextual urban features, and defines a multi-objective reward that balances local and global AQI gains with population, traffic, and industrial considerations. Compared with baseline random and greedy placements, the PPO-driven strategy achieves about $25.40\%$ improvements in overall AQI while delivering more balanced spatial coverage and higher population-impact effectiveness, as demonstrated through multi-dimensional metrics and visualizations. The framework also highlights methodological challenges such as dispersion modeling simplifications and data quality, and proposes future work to incorporate meteorological factors and scalable, multi-agent approaches. Overall, the work demonstrates the viability of DRL for data-driven, multi-criteria urban air quality management and provides a blueprint for deployment in other metropolitan environments.

Abstract

This is the preprint version of the article published in IEEE Access vol. 13, pp. 146503--146526, 2025, doi:10.1109/ACCESS.2025.3599541. Please cite the published version. Urban air pollution remains a pressing global concern, particularly in densely populated and traffic-intensive metropolitan areas like Delhi, where exposure to harmful pollutants severely impacts public health. Delhi, being one of the most polluted cities globally, experiences chronic air quality issues due to vehicular emissions, industrial activities, and construction dust, which exacerbate its already fragile atmospheric conditions. Traditional pollution mitigation strategies, such as static air purifying installations, often fail to maximize their impact due to suboptimal placement and limited adaptability to dynamic urban environments. This study presents a novel deep reinforcement learning (DRL) framework to optimize the placement of air purification booths to improve the air quality index (AQI) in the city of Delhi. We employ Proximal Policy Optimization (PPO), a state-of-the-art reinforcement learning algorithm, to iteratively learn and identify high-impact locations based on multiple spatial and environmental factors, including population density, traffic patterns, industrial influence, and green space constraints. Our approach is benchmarked against conventional placement strategies, including random and greedy AQI-based methods, using multi-dimensional performance evaluation metrics such as AQI improvement, spatial coverage, population and traffic impact, and spatial entropy.

Figures (14)

• Figure 1: Overall Architecture of the Booth Placement Framework.
• Figure 2: Multi-Channel grid layers: AQI, Population, Traffic, Industrial, Green Space, and Booth Placement.
• Figure 3: Reward Components
• Figure 4: CNN Architecture for PPO-based Booth Placement Optimization
• Figure 5: Booth Locations - Random Placement Strategy