IoT-Based Intelligent System for Monitoring and Predicting Urban Air Pollution: Spatial and Temporal Analysis in Seoul

Mina-Rios A.; Albornoz-Yanac S.; Lucich-Muñoz S.; Gamboa-Cruzado J.; Salazar-Marzal A.; Chávez-Herrera C.; Del-Valle-Jurado C.; Pichardo-Lagunas O. · Computacion y Sistemas (2026) · DOI: 10.13053/cys-30-1-6346

Resumen

Urban air pollution constitutes a critical challenge for public health and environmental sustainability, while traditional monitoring approaches present limitations in coverage, timeliness, and data quality. In this context, the objective of this research was to develop and implement an IoT-based intelligent system for monitoring and analyzing air pollution in the city of Seoul, integrating data acquisition, cloud storage, analytical visualization, and experimental evaluation of its performance. Methodologically, an applied approach with a pure experimental design was adopted, and the Scrum methodology was employed for the development of the solution. The analysis was conducted using the public dataset Air Pollution in Seoul, and a control group with manual processes was compared with an experimental group using automated processing. The results show a consistent classification performance of the system, with values of Accuracy = 0.822, Precision = 0.868, Recall = 0.822, and F1-Score = 0.836, as well as a substantial reduction in report preparation time, in the percentage of incorrect data, and in missing data. Likewise, the spatial and temporal analysis of PM2.5 made it possible to identify persistent hotspots and relevant variation patterns at the urban scale. In conclusion, the proposed system improves operational efficiency, data reliability, and the analytical capacity of environmental monitoring, representing a robust alternative to support decision-making in smart urban environments.

Instituciones: Universidad Nacional Mayor de San Marcos; Universidad Nacional Federico Villarreal