Semantic Web and IoT
11
ligence (AI) capabilities are being placed in objects, robots and spaces, enabling them
to comprehend their environment and reason, interpret and learn. As the number and
intelligence of “things” increases, there is a need to shift from statically interconnected IoT nodes to autonomous and collaborative entities in Industry to harness
intelligence and support dynamic connectivity, interactivity and decision making
augmenting the operational value of the industry.
The combination of AI and Semantic Web technologies will lead to a solution
for a number of complicated problems related to interoperability, automated and
self-configurable systems such as those from Industry 4.0. Through this combination it can be achieved an holistic view of a Factory of the Future (FoF) enabling
better decision making across different management layers to reduce overall complexity. This holistic approach includes the interconnection of heterogeneous data
sources, the production chain, business processes and so on. Another example for the
integration of Semantic technologies is on the Autonomous systems. Machines can
communicate and exchange information with other machines under the same vocabulary, in order to succeed a common target. New machines can participate easily in
the production chain without the necessity delegate a heavy work force on this task,
by simply creating interoperable services. Faulty devices are easily being substituted
by discovering new devices with similar functionality to prevent downtime during
the production process.
The semantic technologies in an Industry 4.0 platform can be integrated at the edge
or at the cloud layer, depending on the use case application. The semantic knowledge
layer receives data after a middleware uses standard protocols like MQTT, OPC UA
or HTTP(S) and formatting the data using open standards, like OPC UA, PPMP,
PackML. Then, it employs defining and sharing of semantic information to allow for
easier analysis across different systems [65].
There are many studies which [66, 67] model industrial products and services.
In [68] an approach is presented for integrating IoT to a MAS (Multi Agent Systems) based manufacturing environment, semantically enriched the relevant ontology and its validation through a Hardware-in-a-Loop simulation utilizing a gamification system. In [69] the SAREF ontology was extended with the creation of
the SAREF4INMA ontology for describing the Smart Industry and Manufacturing
domain. SAREF4INMA is based on several standards and IoT initiatives, as well as
on real use cases, and includes classes, properties and instances specifically created
to cover the industry and manufacturing domain. Alvanou et al. [70] proposes the
implementation of MTConnect as machine-interpretable ontology (OWL) to achieve
two things: Firstly, to preserve the semantics of the reference within the model and
secondly, to enable its interlinking with other datasets to form the basis of the Industry
4.0 vision. MTConnect defines specific data patterns to facilitate healthcare monitoring of machine tools. Thus, it provides the foundations for predictive maintenance
to reduce the possibly premature exchange of expensive machine parts or to prevent
entire machine outages due to ruptured parts based on sensor data. Authors utilized
the existing ontologies such SSN, SAREF and SEM for their ontology. Some queries
in SPARQL as a proof of concept for ontology completion were also mentioned.
Industrial robots used in manufacturing kitting stations are modelled in ontologies
11
ligence (AI) capabilities are being placed in objects, robots and spaces, enabling them
to comprehend their environment and reason, interpret and learn. As the number and
intelligence of “things” increases, there is a need to shift from statically interconnected IoT nodes to autonomous and collaborative entities in Industry to harness
intelligence and support dynamic connectivity, interactivity and decision making
augmenting the operational value of the industry.
The combination of AI and Semantic Web technologies will lead to a solution
for a number of complicated problems related to interoperability, automated and
self-configurable systems such as those from Industry 4.0. Through this combination it can be achieved an holistic view of a Factory of the Future (FoF) enabling
better decision making across different management layers to reduce overall complexity. This holistic approach includes the interconnection of heterogeneous data
sources, the production chain, business processes and so on. Another example for the
integration of Semantic technologies is on the Autonomous systems. Machines can
communicate and exchange information with other machines under the same vocabulary, in order to succeed a common target. New machines can participate easily in
the production chain without the necessity delegate a heavy work force on this task,
by simply creating interoperable services. Faulty devices are easily being substituted
by discovering new devices with similar functionality to prevent downtime during
the production process.
The semantic technologies in an Industry 4.0 platform can be integrated at the edge
or at the cloud layer, depending on the use case application. The semantic knowledge
layer receives data after a middleware uses standard protocols like MQTT, OPC UA
or HTTP(S) and formatting the data using open standards, like OPC UA, PPMP,
PackML. Then, it employs defining and sharing of semantic information to allow for
easier analysis across different systems [65].
There are many studies which [66, 67] model industrial products and services.
In [68] an approach is presented for integrating IoT to a MAS (Multi Agent Systems) based manufacturing environment, semantically enriched the relevant ontology and its validation through a Hardware-in-a-Loop simulation utilizing a gamification system. In [69] the SAREF ontology was extended with the creation of
the SAREF4INMA ontology for describing the Smart Industry and Manufacturing
domain. SAREF4INMA is based on several standards and IoT initiatives, as well as
on real use cases, and includes classes, properties and instances specifically created
to cover the industry and manufacturing domain. Alvanou et al. [70] proposes the
implementation of MTConnect as machine-interpretable ontology (OWL) to achieve
two things: Firstly, to preserve the semantics of the reference within the model and
secondly, to enable its interlinking with other datasets to form the basis of the Industry
4.0 vision. MTConnect defines specific data patterns to facilitate healthcare monitoring of machine tools. Thus, it provides the foundations for predictive maintenance
to reduce the possibly premature exchange of expensive machine parts or to prevent
entire machine outages due to ruptured parts based on sensor data. Authors utilized
the existing ontologies such SSN, SAREF and SEM for their ontology. Some queries
in SPARQL as a proof of concept for ontology completion were also mentioned.
Industrial robots used in manufacturing kitting stations are modelled in ontologies