A Multi-Agent Framework for a Hadoop Based Air Quality Decision Support System

Abdelaziz El Fazziki, Abderrahmane Sadiq, Jamal Ouarzazi, Mohammed Sadgal
2015 International Conference on Advanced Information Systems Engineering  
Tropospheric pollution is controlled by various factors such as the distribution of pollutant sources, the nature and amount of energy, as well as the land use and meteorological parameters. These factors must be taken into account in the management of the air quality. Thus, a development of an air quality decision support system able to manage these factors and to answer the questions of environmental managers in real-time is imperative. Such system requires an advanced modeling and
more » ... analyzing and processing techniques that should take into account some aspects, such as the integration of a large amount of data, the behavior of the system environment, the available data sources and the emerging paradigm related to the intelligent systems. To this end, we propose an approach based on the use of the agent technology and big data concept. For the air quality data collection and analysis, we use a Hadoop framework: HBase for data storage and a MapReduce based forecasting process; artificial neural network (ANN) based prediction and K-means as clustering algorithm. Finally, the approach is validated by a case study in which an air quality management support system for the Marrakech city is presented.
dblp:conf/caise/FazzikiSOS15 fatcat:kyerz6wxofadhbhtivxwre2zq4