Thesis defense by Widad ES-SOUFI

The industry is currently undergoing rapid development following the implementation of the Industry 4.0 concept, which aims to make factories smart entities by combining industrial processes and practices with the latest information and communication technologies, such as cyber-physical systems and the Internet of Things (IoT), among others. This industrial development, along with the need to innovate in order to achieve and maintain competitiveness, is driving a phenomenal increase in data volume (known as Big Data), making system design and supervision processes increasingly complex, data within organizations increasingly difficult to exploit, and decision-making increasingly challenging. Indeed, several problems have been identified in industrial environments, which can be classified into three categories: difficulties in searching for, visualizing, and exchanging information; difficulties in decision-making; and difficulties in managing changes in context. Through this work, we propose a modular decision support system, each of whose four modules solves one of the identified problems. The process modeling and traceability modules enable the modeling of processes and their execution. The decision support module proposes the most appropriate pattern for the decision context as well as the most significant parameters. The contextual change management module continuously updates the decision-making module, enabling it to take into account the dynamic nature of the context data. The proposed system is illustrated and validated in the context of the Gontrand project, which aims to provide real-time supervision of gas networks promoting the injection of green gas.