A model predictive control application for air quality management

This study introduces a model predictive control methodology to determine optimal measures for mitigating air pollution, assisting Local Authorities in policy development. Anchored in an auto-regressive model, it analyzes dynamic air quality patterns over a defined timeframe using daily observed pollutant concentration, meteorological variables, and estimated emission data. Employing model predictive control methodology, the approach aims to optimize daily emission reductions. Evaluated in Milan, a heavily polluted European city, the findings highlight the methodology's potential as a robust tool for Local Authorities, enabling informed decisions in crafting efficient air quality management strategies, in the specific context of NO2.