03 Jul
03/07/2020 10:00

Sciences & Société

Soutenance de thèse en visioconférence : Minhu LYU

Towards Control as a Service models and architecture for the Industry 4.0 »

Doctorant : Minhu LYU

Laboratoire INSA : LIRIS

Ecole doctorale : ED512 : Informatique et Mathématiques de Lyon

To fit the renewed globalized economic environment, enterprises, and mostly SMEs, have to develop new networked and collaborative strategies, focusing on networked value creation (instead of the classical value chain vision), fitting the blue ocean context for innovative products and service development. Even if collaborative organizations have been studied for decades, the closer connection of information systems involved by the so-called “Industry 4.0” developed by leading industries in Europe, US and Asia requires to set new IT models to support agile and evolving collaborative Business Process (BP) enactment, integrating both traditional Information Systems (IS) and production control processes. By now, these product/service ecosystems are mostly supported by software services, which span multiple organizations and providers, and on multiple cloud-based execution environments, increasing the call for openness, agility, interoperability and trust for both production and Information System organization. These requirements are well supported by SOA, Web 2.0 and XaaS technologies for Information Systems. Taking advantage of IoT, services and Cloud technologies, the development of Cloud of Things (CoT) changes the way control application are engineered and developed moving from a dedicated design and development of control applications to a Control as a Service vision. This vision requires developing a new architecture to connect physical and logical objects as well as integrating basic control patterns to organize a consistent control service orchestration. To fit this challenge, we propose a multi-layer Control as a Service architecture to describe control systems in a holistic way. Our Control service model is built according to an event-driven orchestration strategy. Thanks to the integration of a context manager, analyzing continuously the system environment as well as the control system behavior, these context-aware control services can be deployed