The Delhi government is exploring a collaboration with IIT Kanpur to develop an AI-enabled decision support system (DSS) to tackle the capital’s persistent air pollution problem through real-time data and precise source identification.
The proposed system will focus on hyperlocal source apportionment, sensor-based monitoring and real-time analytics to identify pollution sources at a granular level and enable targeted interventions, moving away from blanket bans and reactive measures.
“Under the leadership of Chief Minister Rekha Gupta, Delhi’s fight against pollution is being made scientific, sustained and strategic. Decisions will be driven by real-time data, source identification and measurable outcomes, rather than emergency responses,” Delhi Environment Minister Manjinder Singh Sirsa said in a press conference. He added that the emphasis is on targeted action at pollution hotspots rather than on city-wide restrictions.
A key feature of the proposed system is dynamic source apportionment, which would help authorities scientifically determine the contributions of various sources such as road dust, vehicular emissions, industrial activity, biomass burning and regional factors to air pollution levels. Officials said this evidence-based approach would allow enforcement agencies to act directly at the source of pollution.
“Pollution control cannot be seasonal. Delhi needs a 365-day action framework that integrates technology, governance and enforcement, backed by data-driven decision-making,” Sirsa said.
Currently, Delhi relies on a decision support system operated by the Indian Institute of Tropical Meteorology (IITM), Pune, and the India Meteorological Department (IMD). While the Air Quality Early Warning System (AQEWS), launched in 2018, has shown over 80% accuracy in forecasting high-pollution days, according to the Council on Energy, Environment and Water, experts have raised concerns about its reliance on outdated emission inventories and its tendency to underpredict pollutant levels.
Earlier, AIM published a story on how AI can connect emissions, traffic and weather data. AI-powered computer vision systems can identify vehicle types at junctions and flag those contributing disproportionately to particulate spikes. When combined with traffic-flow analysis, such systems can recommend dynamic rerouting in polluted corridors before exposure levels become hazardous, said Prof Damodaran from IIM.
The post Delhi govt plans AI-enabled pollution control system with IIT Kanpur appeared first on Analytics India Magazine.
