Business Intelligence Framework for Air Quality: Management A Case Study in the Municipality of Lisbon

Abstract

Air quality is a constant public health issue in urban areas, where traffic density, meteorological conditions, and population growth contribute to high concentration levels. In this research we propose to develop a Business Intelligence solution to monitor and analyze air quality sensor data in the Municipality of Lisbon by integrating environmental and traffic congestion data. The solution followed the Kimball lifecycle methodology was implemented through Microsoft Fabric tools, incorporating a full architecture. We propose a tool with both current air quality data and historical data analysis, to discover hidden patterns in the time of the day, day of the week and months, and identify hotspot areas in Lisbon. Additionally, the incorporation of Key Performance Indicators allows to develop an alert system to inform authorities of health risk implications. A forecasting component using both SARIMA and Prophet models supports short-term monitoring of key pollutants. The results reveal that PM10, PM2.5 and NO2 are the most critical pollutants in Lisbon, with frequent high concentration values in different locations, as well as Prophet outperforming SARIMA across most pollutants. This study contributes to a fully integrated solution by combining Business Intelligence tools and forecasting techniques in an interactive tool, allowing proactive decision-making.