Semantic Technologies for Assisted Decision-Making in Industrial Maintenance


The growing use and popularity of sensors and monitoring devices in the context of industries has paved the way not only for new analysis approaches but also for new insights in the production chains. Nevertheless, the potential of this data, especially for optimizing the maintenance process, has not yet been fully used.
We present an integration and platform approach for optimizing industrial maintenance, based on Semantic Web Technologies, where both physical and digital components are encapsulated in Smart Services and are available via consistent, industry 4.0 compliant interfaces.
Via RDF annotations on service input and output, technical characteristics and life-cycle information we enable an automated establishment of execution pipelines with a rule-based engine. The resulting self-organizing network of heterogeneous services supports the changing nature of production environments, via dynamic adjustments to changing conditions without direct human involvement. The thereby established flexibility allows for new data-driven business cases, and risk analysis based on automated and real-time decision-making.
Hereby we present the next generation for optimizing the schedule of field technicians. Due to the heterogeneous data formats, closed systems and missing insights, the process of aligning maintenance personal with customer demands is currently still dominated by inefficient short-term and partially improvised planning. The architecture that we present connects all available components in a modularized architecture and, therefore, allows for proactive instead of reactive maintenance.