Optimizing air pollution sensing for social and environmental justice

Abstract

Low-cost sensors have emerged as a new urban technology to provide localized air pollution sensing data. However, common approaches to sensor deployment, whether market-driven or crowdsourced, often reinforce existing data gaps and perpetuate social and environmental injustices. To address this, this paper develops a new location modeling framework that integrates environmental and social justice goals for equitable sensor placement. We propose a gradual covering location model (GCLM) to optimize sensor distribution, considering data for both environmental exposure and sociodemographic vulnerability. Our application to air quality sensing in Chicago (United States) demonstrates the effectiveness of the proposed framework, showing that sensors are suggested to distribute across high-traffic downtown areas and vulnerable communities, providing more equitable coverage compared to existing public, participatory or crowdsourced sensor networks.