RePlanIT Ontology for FAIR Digital Product Passports of ICT: Laptops and Data Servers

This paper presents an ontology of ICT Digital Products Passport called RePlanIt, which captures various aspects of such products that are useful to characterize their life-cycle, with the aim of adhering to FAIR principles. The ontology is presented through a use case and its purpose is to categorize the indicators that are used to measure the sustainability of a device, towards circular economy.

The ontology is fairly well structured and reuse several concepts taken from other existing ontologies. Given the complexity of the domain to be represented, in which many different subdomains are intertwined. RePlanIT does a good job in showing the interconnection between different aspects of DPPs.

The data file is well organized and provided within GitHub, with an accompanying README file. The documentation adheres to the FAIR principles.

DETAILED COMMENTS

*Section 2*
The related work section is well written and reasonably exhaustive, highlighting where existing ontologies may be reused, and the reasons that hinder their reuse when they may not (interconnections with other modules, too coarse-grained granularity etc.). It thus constitutes a good comparative framework.

*Section 3*
The methodology section is very clear and well explained.

*Section 4*
I appreciate very much the effort the authors have made to reuse as much as possible concepts derived from already available ontologies. However, in order to grant that such concepts have been used correctly and in a meaningful way, the authors should provide the link of such reused ontologies to top level ontologies whenever they are available, otherwise they should provide a brief explanation of why a certain concept is adequate to be reused, especially when concepts are named with ambiguous terms (see, for instance, Icc:location, matonto:Material, prov:Agent, dcat:Role, time:Temporal Entity, sosa:Result). Such terms are so general that we can hardly grant they are used with the same meaning in different ontologies.

On the other hand, when introducing new concepts, whose function is central in the ontology (see for instance Switch, Data Server, ICT Device and, especially, Indicator), they should provide a brief natural language definition, as their subclasses are not sufficient to explain which are the essential characteristics that are being taken into account when using such concepts.

Another issue that I see is that many data properties cannot be absolutely ascribed to devices, as they change in time, so a temporal index should be provided for such properties. This could be especially helpful when trying to support predictive maintenance processes (as they allow to foresee how such properties could change in the future, based on how they have changed in the past).

The data property replanit:MaterialRecyclability is a very important one, but I don’t understand the choice of making it a Boolean, as there are materials which are only partially recyclable. Wouldn’t it be useful to indicate a value such as a percentage? Are these data available?

The class replainit:Reason should be better explained, due to its centrality and the vagueness of the term. Also in such case, linking to a top level ontology would enormously ease the task.

I found the category replanit:CircularActivityCost very interesting, but I would encourage the authors to consider not only the economic cost, but also the environmental one, given the specific context and purpose of RePlanIt. At least, they should highlight at this point the indicators that connect these two aspects (economic and environmental cost).

Also, some of the mentioned CE strategies (e.g. remanufacturing, refurbishment, recovery, repir) are distinguished by subtle nuances, thus I encourage the authors to provide at least a short natural language description.

I found it a bit counterintuitive to assume that replanit:Activity refers only to processes supported by a software, as there are many other activities of interest in the domain which are not ICT-supported (or on which the software part is not the fundamental one). Wouldn’t be worth providing a more specific name?

Figg. 3, 4, 5, 6, 7, 8 and 9 are hardly readable, especially in the blue boxes.

*Section 5*
Fig. 10 is hardly readable, the authors should either provide a higher definition image or remove it. This also holds for Fig. 11 and Fig. 12. Also, on Fig. 12, there seems to be a typo Climat vs. Climate.

*Typos*

p. 3 such as our -- > such as ours
p. 6, footnote 7: the final dot is missing and I would remove the reference to footnote 2, I’ve never encountered a footnote of a footnote before.
p. 7 as a standalone classes -- > as standalone classes
p. 8 in the ontology, is available -- > in the ontology is available (remove comma)
p. 8 expends -- > expands
p. 8 replant: -- > replanit: (check throughout the paper, it looks like an automatic correction)
p. 9 manufacturering -- > manufacturing
p. 10 GassEmissions -- > GasEmissions
p. 12 graphs, store -- > graphs store (remove comma)
p. 12 in according -- > according
p. 15 in the numbered list, there is i) instead of ii)
p. 15 but it is also -- > but is also
p. 15 ontology, which -- >ontology which,
p. 16 utilisation of the proposed in this paper knowledge graph-based DPPs -- > utilisation of the knowledge graph-based DPPs proposed in this paper

To sum up, I enjoyed reading the paper, which I believe is well written and well structured. I would have liked more foundational issues to be addressed (which I specified in the detailed comments), but overall I believe it is a good contribution.

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. Please also assess the data file provided by the authors under “Long-term stable URL for resources”. In particular, assess (A) whether the data file is well organized and in particular contains a README file which makes it easy for you to assess the data, (B) whether the provided resources appear to be complete for replication of experiments, and if not, why, (C) whether the chosen repository, if it is not GitHub, Figshare or Zenodo, is appropriate for long-term repository discoverability, and (4) whether the provided data artifacts are complete. Please refer to the reviewer instructions and the FAQ for further information.

The paper present an ontology for documenting the sustainability of laptops in the context of a circular economy scenario. The ontology is sound and the paper broadly respects good practices in ontology engineering. Related work is well presented, the authors provide a comparison between this ontology and existing alternative efforts.

The FAIR notion in the title seems overused; there is little discussion of FAIR principles in relation to ontology engineering practice or reference scenario. Of course, ontologies are somehow FAIR because of interoperabilty etc…

There is an accent of cross-domain knowledge exchange in the motivation and a lot of the limits of existing approaches are related to this aspect. However, there is little description of how the presented ontology would solve this problem. Generally, any time. The limit of existing ontologies from the domain perspective is always that either they are too specific, or not enough. It's very hard to assess what is it that makes a domain coverage "good enough". One way to solve this problem is in being very clear about the task(s) that the ontology is supposed to support. The article would be improved by starting with a clear scenario and tasks that are supposed to be supported. How this solution is cross-domain? Table 1 is an interesting starting point to develop a narrative showing the value-to-users of the ontology.

The ontology is paired with a UI -- is this being evaluated with domain experts?

There is a KG applying the ontology to a small set of samples but there is no discussion about knowledge acquisition. Where is this information coming from? How this ontology plugs into the larger IT infrastructures of procurement/management in organisations?

Note 20 on the circularity calculator should be an extensive paragraph linking this effort to the problems and needs of who work for supporting CE.

The ontology seems a mere collection of classes, relationships, properties. No discussion on automated inferencing, what type of description logic is required. Why even making an Ontology, after all? A web database application should do the job fine. Interoperability? Inferencing? A scenario should tell.

Finally, the ontology documentation seems much broader than what described in the article, I wonder that maybe the work progressed since the time of this submission.

The paper presents an ontology for digital product passports for ICT devices. It is well written and presents a good structure. The work reported in the paper appears to be mature with indications provided by the authors of concrete uses. From the references throughout the text, the authors demonstrate that previous phases/parts of the work, such as the systematic literature review, have been published and this article provides the most complete and up-to-date account of the research.

Below i provide a few minor comments that can improve the clarity and readability of the paper. However, my main comment is related with the title, specially the part “FAIR Digital Product Passport”. Besides the abstract and introduction, there is no further mention of the FAIR principles. I’d expect from such a title, a section of the paper discussing which and how the FAIR principles have been covered by the ontology. Without FAIR in the title, the paper maintain its value but mentioning the principles generate expectations that, in my opinion, were not satisfied.
Further, and minor comments are:
- page 4, line 3. The acronym BIM has been introduced without explaining what it stands for. From the context, I could infer that it was about Building Information Modelling, but it is always good to explain the acronym the first time it is mentioned in a text.
- page 5, line 39, the authors state that “ For consistency, an "isA" relationship was followed when defining classes and their sub-classes.”. However, in the diagrams, the predicate rdfs:subclassOf is used instead. By the way, commonly, predicate names have the first letter in lower case.
- In figure 2, the class Certification is introduced. The name suggests and event or activity of being certified but I believe that authors meant the informational artefact that is the product of the certification process, the certificate.
- page 7, line 39. Serial number and model are mentioned as devices characteristics that change over their lifetime. I was curious about examples of devices that change their serial numbers or their models after they are build.
- figure 7, it is not clear which classes are connected from the predicates hasIndicator and hasMeasurementUnit. In this figure the class HardwareComponent seems to be the source of both predicates but figure 9 depicts differently.
- page 16. The last sentence of the Conclusions section would benefit from a rephrasing as it is very long and not very clear.