ExtruOnt: An Ontology for describing a type of manufacturing machine for Industry 4.0 systems

Tracking #: 2217-3430

Victor Ramirez
Idoia Berges
Arantza Illarramendi1

Responsible editor: 
Guest Editors SemWeb of Things for Industry 4.0 - 2019

Submission type: 
Ontology Description
Semantically rich descriptions of manufacturing machines, offered in a machine-interpretable code, can provide interesting benefits in Industry 4.0 scenarios. However, the lack of that type of descriptions is evident. In this paper we present the development effort made to build an ontology, called ExtruOnt, for describing a type of manufacturing machine, more precisely, a type that performs an extrusion process (extruder). Although the scope of the ontology is restricted to a concrete domain, it could be used as a model for the development of other ontologies for describing manufacturing machines in Industry 4.0 scenarios. The terms of the ExtruOnt ontology provide different types of information related with an extruder, which are reflected in distinct modules that constitute the ontology. Thus, it contains classes and properties for expressing descriptions about the components of an extruder, the spatial connections, features, and 3D representations of those components, and finally about sensors used to capture indicators about the performance of this type of machine. The ontology development process has been carried out in close collaboration with experts from a manufacturing company.
Full PDF Version: 

Major Revision

Solicited Reviews:
Click to Expand/Collapse
Review #1
By Marwan Al-Tawil1 submitted on 14/Jul/2019
Major Revision
Review Comment:

The paper proposes an ontology for describing a type of manufacturing machine (extruder) for performing extrusion processes. The paper describes the ontology design process and the five models that constitute the ontology and briefly explains how each module supports three types of stakeholders - novice workers, designers and domain experts. The described process is interesting and combines concepts and features from other existing ontologies. The proposed ontology has been evaluated from three different points of view - ontology pitfall scanner, ontology quality and ontology metrics. The proposed ontology is documented and its components are available online.

In general, the paper is well written and organised. The proposed ontology is original, interesting and relevant. The authors provide a detailed description of the different modules that constitute the ontology. The illustration of the ontology components is readable and easy to follow. However, there are some points require improvement.

The paper is lacking a related work section. Although the paper (at the beginning of section 2) provides a brief summary of similar ontologies and lists some methodologies for ontology development, however, the authors are encouraged to create a separate section of related work to describe in more detail similar ontologies, ontology development methods and methodologies, and ontology evaluation approaches. This will enable the authors to provide better justification why the NEON ontology development methodology and evaluation approach were chosen (some missing founding works about ontology evaluation approaches are listed below).

Section 2 describes the ontology development process using NEON methodology. In order to provide a better illustration of the development process, I suggest restructuring that section into subsections to describe the main phases of NEON (i.e. subsections to describe Initiation, Design, Implementation and Maintenance of the proposed ontology).
One of the contributions of the proposed ontology is reusability. Although the paper says that the extruder component module can be replaced by a module that describes components of another type of manufacturing machine, however the paper does not provide proper descriptions nor examples of how the different modules of the proposed ontology can be re-used/customised to develop other ontologies for describing other types of manufacturing machines. This can be described in the evaluation section (e.g. extend subsection 4.2 – Adaptability).

In some occasions (e.g. Introduction/4th Paragraph and Section 2/ end of 1st paragraph) the paper mentioned that concepts of the proposed ontology align with concepts/terms from other ontologies. However, the description of the component module in Section 2 indicates that non-ontological resources have been also used to define concepts. This brings the confusion if all (or some) concepts are aligned with the ontological recourses.

The ontology evaluation section does not provide proper justification why three different approaches for evaluation and why those approaches were chosen. What about the evaluation approach in [1] which considers aspects such as: a) gold standard (i.e. find similarities and differences between the current ontology and existing ones); b) application-based (i.e. what are the potential uses and applications of the , and c) Human assessment (i.e. to answer questions such as: can humans recognise all/some concepts?, do humans approve the facts driven from the ontology?, and will human trust the encoded knowledge in the ontology?). What about domain coverage? (i.e. to what extent does the proposed ontology cover concepts and properties required in extrusion process?).

Finally, the paper does not talk about future work. I would add a new paragraph to the conclusion and append extra applications of the proposed ontology.

Minor issues:
Abstract: descriptions about the components of an extruder, the spatial connections…  descriptions about components of an extruder, spatial connections… .
Introduction: machine type that performs an extrusion process…  machine type (called extruder) that performs an extrusion process…
Introduction. I assume that the development of a visual query system is one of the contributions of this paper and thus the paragraph “Finally, an ontology …” should be added to the two previous contributions (i.e. 3) the development of an ontology-based query system will ….).
Section 2. P2: we did not found any ontology that described industrial  we did not found any ontology that describes industrial.

Suggested references
[1] Brank et al. A survey of ontology evaluation techniques. SIKDD’05
[2] Sabou & Fernandez, Ontology (network) evaluation. Ontology Engineering in Networked World, 2012.

Review #2
Anonymous submitted on 07/Aug/2019
Major Revision
Review Comment:

This manuscript was submitted as 'Ontology Description' and should be reviewed along the following dimensions: (1) Quality and relevance of the described ontology (convincing evidence must be provided). (2) Illustration, clarity and readability of the describing paper, which shall convey to the reader the key aspects of the described ontology.

The paper presents an ontology, ExtruOnt to describe a type of manufacturing machine, extruder. The paper further describes the process of the development of the ontology, based on different dimensions - components, spatial connections, features of the components, 3D representation and sensors. The authors have motivated the need for the ontology well and the impact that the ontology could have within the manufacturing industry and industry 4.0. Overall, the paper is very well written and describes the process of the development of the ontology very well. Although the paper describes a manufacturing machine and I am not an expert in manufacturing, I very much enjoyed reading the paper. I found the descriptions of the ontology modules easy to follow and well written, and the authors have linked with a variety of (specific as well as external) domain ontologies. The ontology is in a highly specialised and applied domain and developed with experts in the manufacturing domain. Given this highly specialised domain, I think it would help to discuss more about the experts that had participated in the process - the paper discusses that the ontology emerged from the process of creating the Ontology Requirements Specification document. However, I am unclear how many experts were involved in this process, what kind of agreement they had amongst themselves and how representative are the competency questions. This is particularly important as the process of the development of the ontology involved aligning the sections of the ontology with how it could answer the competency questions. As such, I think the evaluation of the ontology would benefit immensely with the involvement of experts outside the project team/consultation team. Although the competency questions appear thorough and robust for the purpose of the development of the ontology, I am not sure if critical questions could have been left out - a team of experts could potentially help in evaluating this aspect. While I very much enjoyed the narrative of the paper, including the descriptions of the different components of the ontology, I think the evaluation section could be stronger. This is more so, given the highly specific and applied nature of the domain - it would be very helpful to see some discussions of the evaluation applied in context of the domain as an application or inspection by experts. However, given that the domain is very specialised, this may not be always feasible as experts are expensive and hard to find.