Big-Data-Analytics in Environmental and Structural Monitoring (BAM)

Environmental sensors are carried by millions of people through their personal lives. Multisensor systems autonomously track potentially dangerous environmental changes. Current visions of the future suggest that we will increasingly not only perceive environmental changes, but also interpret, evaluate and vividly communicate them in real time. An interdisciplinary team of researchers at Mainz University of Applied Sciences is dedicated to these goals, taking up Big Data methods as they are currently emerging in the world of rapidly growing and increasingly heterogeneous mass data. In close cooperation with the Department of Geoinformatics in the Faculty of Engineering and the i3mainz and Big Data Analytics Institute in the Faculty of Economics, promising methods are being further developed with regard to the monitoring of natural and anthropogenic environmental changes. The project explores the potentials of current data mining and machine learning methods for questions with space-time relevance. With the development of a meta-learning system and new visualization methods, the number of potential users of complex analyses is to be greatly increased. In order to enable the transfer of the researched fundamentals, they will be prototypically implemented and evaluated in complementary application fields. For example, a Big Data analytics system is being developed for issues in the smart city field, focusing on analyses of various sensor data on environmental and health issues. Furthermore, the degree of autonomy of optical monitoring systems for the precision monitoring of large structures, such as wind turbines or bridges, is increased on the basis of image analysis with the help of deep learning systems, examined for reliability and trimmed for practical suitability.