Air pollution is a problem that causes adverse effects, which tends to interfere with human comfort, health or well-being, and that may cause serious environmental damage. In this regard, this study aims to analyze large data sets generated by real-time wireless sensor networks that determine different air pollutants. Business Intelligence and Data Mining techniques have been applied in order to support subsequent decision-making strategies. For normalization and modeling, we applied the CRISP-DM methodology using the Pentaho Data Integration. Then, the Sap Lumira has been applied in order to acquire models of tables and views. For the data analysis, R-Studio has been used. For validation, Clustering has been applied using the k-means algorithm by the Jambu method, where it has been proceeded to check the consistency of these, being later stored and debugged in PostgreSQL. Results demonstrate that the increase in air pollutants is directly related to the traffic hours, which may cause an increase of asthma or sick related syndrome in the population. This analysis may also serve as a source of information to authorities for improving public policies in such matter.