SS06 Industrial Automation Systems in the Industrial Internet (of Things)
Special Session Organizers:Kristian Sandstrom, ABB, Sweden
Moris Behnam, Mälardalen University, Sweden
Aim: There is a push in the market for making devices smarter and to generate new services by exploiting the increased connectivity and bandwidth, the computational capabilities and power efficiency of new hardware, the scalability and cost advantages of cloud computing, and the collaboration be-tween these devices, i.e., the Internet of Things (IoT). This is also true in the industrial domain, and although there are similarities between IoT for general systems and industrial systems, e.g., with respect to scalability, there are also significant differences such as constraints on low latencies, criticality of the systems, requirements for predictability and resilience to failures in the system. Hence, there is justification for specifically consider Industrial IoT systems (industrial Internet, Industry 4.0).
There are many potential applications for the industrial internet, e.g., collecting large quantities of device data for subsequent (big) data analytics to support preventive maintenance and remote service. It is also possible to envision the analytics as part of control loops, and act on the information in real-time. Clearly, there are challenges that need to be addressed to support critical industrial applications in an industrial internet scenario, e.g. Managing latencies, Supporting the required predictability, Resilience to failures and non-determinism in the infrastructure, Dealing with mixed criticality.
Topics: This special session will be focusing on (but not limited to) the following topics:
- Resource management related to edge computations, e.g., allocation of computations at different infrastructure levels ranging from edge to cloud.
- Architectures for robust and predictable control in an Industrial IoT setting.
- Supporting virtualization for industrial IoT context using servers, gateways or other devices.
- Efficient design and implementation of Virtual Machine Monitor VMM in terms of the timing constraint, QoS assurance, runtime overhead and energy consumption requirements.
- Methods for achieving advanced features found in server virtualization, such as live migration, with predictable real-time performance.
- Supporting service level virtualization for distributed real time systems.
- Supporting heterogeneous network technologies including Ethernet based field busses through virtualization techniques.
- Case studies of industrial IoT applications.
- Analysis framework for servers, communication, distributed systems for applications spanning in several levels of the IoT infrastructure.
- Runtime adaptable mechanisms for optimizing resource usage or for fault tolerance management in the IoT infrastructure.