Michael Pirker

Towards Semantics-based Monitoring of Large-Scale Industrial Systems

Monitoring todays industrial systems steadily grows in terms of complexity. Examples can be found in industry automation, building management, the medical and the transportation domain. Increasing numbers of components, increasing numbers of sensors producing monitoring data about these components, and increasing numbers of stakeholders managing this data lead to suboptimal exploitation of the actually available monitoring data. What is required, is a semantically meaningful way of representing, integrating and querying monitoring data from different sources. To this end, we propose to apply Semantic Web technologies to the monitoring of large-scale systems. This creates new challenges such as incorporating dynamic real-world data and supporting scalable reasoning over this data. As an approach to this problem, we present a generic architecture with different functional layers . We go into more detail about the distributed reasoning layer, where we present the idea of loosely coupled collaboration of several reasoners. Finally, we describe the current state of our prototype implementation, where we adopt a scenario from the railway monitoring domain, and present first evaluation results.

Semantic data integration and monitoring in the railway domain

Information integration is a key for further growth of efficiency in management decisions for the railway domain. In the context of the EU project InteGRail (funded in the 6th Framework Programme) an integration approach leveraged by ontologies known from the Semantic Web and logic-based reasoning mechanisms has been successfully demonstrated. To this effect existing heterogeneous monitoring data acquired across the European railways (in the context of rolling stock, infrastructure, operations and traffic management) is logically integrated according to a formal information model. Based on distributed reasoning mechanism decentralized data and inferred knowledge does not have to be aggregated in a central repository but can be transparently accessed by applications independently from where it is acquire. We explain how the proposed techniques facilitate integration, analysis and interpretation of distributed observation data in the railway domain. Finally the implementation of the presented approach is presented by a demonstration scenario, which integrates existing real-world data for symptom identification and fault detection.

Evaluation Model for Assessment of Cyber-Physical Production Systems

Cyber-physical production systems based on technologies such as machine to machine communication, the Internet of Things and other cutting edge technologies are going to advance manufacturing automation and industrial production. Information technology seems once again to be the driving force for change in manufacturing automation. But what are the characteristics of such systems in comparison to the existing approaches? In this article we recommend an evaluation model for cyber-physical production systems is proposed based on a set of system characteristics, which defines specific abilities and performance indicators. Furthermore, an analysis and verification of that model is presented sketching the typical pattern and impact of cyber-physical production systems. As a result a refined evaluation model is available, suitable for the characterization of cyber-physical technologies and thereby enabling a technological assessment.

From Personal Assistance to Industrial Solutions – A Generic Ambient Intelligence Architecture and Framework (Von persönlicher Assistenz zu industriellen Lösungen – eine generische Architektur und ein generisches Framework für Ambient Intelligence)

Summary Ambient Intelligence refers to the vision of computationally-enriched environments that are sensitive and responsive to the presence, intention and activity of people. We extend this vision from personal to the business and industrial domain and describe a technological framework that can be leveraged in both worlds. Zusammenfassung Ambient Intelligence repräsentiert die Vision von elektronisch angereicherten Umgebungen, die sensitiv auf die Präsenz, Absichten und Aktivitäten von Personen reagieren. Wir erweitern diese Vision von der privaten in die geschäftliche und industrielle Domäne und beschreiben ein Technologie-Framework, das in beiden Domänen eingesetzt werden kann.