Towards an Open Standards-based Architecture for Condition-based Predictive Maintenance and IIoT

Tracking #: 2083-3296

Karamjit Kaur
Matt Selway
Georg Grossman
Markus Stumptner
Alan Johnston
Pak Wong

Responsible editor: 
Guest Editors Sensors Observations 2018

Submission type: 
Full Paper
In today's swiftly emerging Industry 4.0 environment, sensors/devices, machines and components have digital twins which are connected together enabling them to talk to each other. Factory operators can thus continuously capture machine states/condition and combine it with information from other systems, analyze it and predict the optimal point in time at which to initiate maintenance. This approach for maintenance generally called Condition-based Predictive Maintenance (CBPdM) can pinpoint imminent outages well before they occur, significantly enriching business performance by avoiding lengthy production outages. CBPdM is among the major focus points of the Industry 4.0 and IIoT. Interoperability of the asset management systems is crucial in achieving accurate diagnosis and prognosis as it can highly augment the data received from assets. The foundation of an interoperability architecture are standards. Unfortunately, even after wider adoption of CBPdM in industry, to the best of our knowledge there does not exist any reference architecture for it. This paper contributes by introducing Open Industrial Interoperability Ecosystem (OIIE) architecture which is an outgrowth of several related industry standardization activities for achieving standards-based interoperability. We illustrate how the architecture addresses the requirements of Industry 4.0 and CBPdM with the help of a detailed use case.
Full PDF Version: 


Solicited Reviews:
Click to Expand/Collapse
Review #1
By Simon Cox submitted on 13/Jan/2019
Review Comment:

The paper describes an architecture for a particular aspect of the IoT, related to maintenance.
The architecture appears to be based on consolidation and harmonization of existing technologies.
However, these are based on XML, and the (now largely superseded) SOAP and WSDL stack, and the paper does not make use of or reference to any of the conventionally understood Semantic Web stack. Furthermore, semantic web standards which are potentially directly relevant to the topic (e.g. W3C SSN Ontology) are not mentioned.

As far as I can tell the paper may be a solid contribution in the application area.
But the technical disjuncture makes it inappropriate for SWJ.
I suggest the authors submit it to a more suitable venue, probably in information systems.

Review #2
Anonymous submitted on 13/Feb/2019
Minor Revision
Review Comment:

This is a very interesting paper which clearly fulfills a research gap by introducing Open Industrial Interoperability Ecosystem (OIIE) architecture which is an outgrowth of several related industry standardization activities for achieving standards-based interoperability. I have only one comment: Section 2 introduces the terms CBPdM and prescriptive maintenance, but they do not seem very different. This may cause confusion. I would prefer something like "CBPdM, sometimes referred as prescriptive maintenance (reference), ..." or "There is not a widely agreed term for...".

Review #3
By Maxime Lefrançois submitted on 05/Apr/2019
Review Comment:

This paper describes a novel architecture that leverages a chosen set of standards in order to address interoperability in Industry 4.0 and Industrial Internet of Things settings, specifically targeting condition-based predictive maintenance use cases.

The paper starts with extensive and insightful description of the different strategies for asset maintenance in Industry 4.0. It then provides a lists of chosen relevant international standards for condition-based maintenance (Section 3), then related to the Oil and Gas domain (Section 4). In these sections, the list of domains that are addressed is very big, which leads to the problem that the set of available standards is not described (for examples for units of measures), and the paper basically just lists and describes the set of standards that they promote for addressing interoperability problems.
Section 5 presents the proposed architecture, named Open Industrial Interoperability Ecosystem, that relies on a small set of communication standards, conceptual object models stored in a registry, interoperability registers using SOAP/XML Web services for management and administration of connected systems, and transformations from external standards to the chosen one.
Section 6 shortly describes how the proposed ecosystem answers to the main principles of Industry 4.0.
Section 7 describes a use case for which an pilot *will* be developed to demonstrate the proposed approach. The structure of the XML messages is described, along with the set of verbs that can be used in request/response and publish/subscribe communications. Mapping of this common communication object model to the ISO 13374 Conditional Based Maintenance standard is proposed.

I believe this paper fails at demonstrating it is in the scope of the topics addressed by this special issue of the Semantic Web Journal. The proposed solution addressing Interoperability using a selection of standards has very little relation to the Semantic Interoperability, Semantic Web, Linked Data. With respect to the main selection criteria of the Semantic Web Journal:
(originality): An ecosystem using a selection of standards and translators to/from other standards is not something new.
(significance of the results): This ecosystem is not compared to other proposed approaches, and not (yet) implemented and evaluated.