DogOnt as a viable seed for semantic modeling of AEC/FM

Tracking #: 1622-2834

Dario Bonino
Luigi De Russis

Responsible editor: 
Guest Editors ST Built Environment 2017

Submission type: 
Ontology Description
Energy consumption and performance assessment of Smart Cities must consider different levels, various sub-domains, and several stakeholders. A comprehensive energy profile of a city, in fact, should work at the city, district, and building levels. At the same time and for each level, it should take into account both electrical and thermal consumptions, and gather these information from a plethora of different stakeholders (i.e., citizens, utilities, policy makers, and energy providers) and sensors. Current modeling approaches for this context address each level and domain separately, thus preventing a structured and comprehensive approach to a unified energy representation. Moreover, current approaches make difficult to keep the consistency between the energetic data through levels, sub-domains and across stakeholders. Starting from an analysis of ontologies at the state-of-the-art, this paper shows how DogOnt can be used as a foundation towards a shared and unified model for such a context. DogOnt was firstly developed in 2008 and withstand over 8 years of usage without major failures and shortcomings. We discuss successful design choices and adaptations, which kept the model up-to-date and increasingly adopted in domains ranging from home automation to energy representation in Smart Cities.
Full PDF Version: 

Major Revision

Solicited Reviews:
Click to Expand/Collapse
Review #1
Anonymous submitted on 05/Jun/2017
Major Revision
Review Comment:

The paper describes DogOnt for supporting AEC/FM processes, in particular, energy profiles for both electrical and thermal consumption. It is difficult to determine from the conclusion what the contribution is though beyond what was presented in 2008 (one of the few papers referenced). It highlights changes to DogOnt, but it is not exactly clear what these are. The paper mentions many times changes to the original 2008 version of DogOnt, but no references are provided.

The SoA is also lacking. There is barely any mention of existing standards for BIM, like IFC (mentioned only in reference to SAREF). While this in itself is not necessarily a criticism, the paper does refer to other ontologies, like SAREF, but doesn’t really explain what the differences are between it and DogOnt, why DogOnt is a better choice, or why DogOnt has not examined the ontologies that were examined in the development of SAREF. Overall, this leaves the reader with the sense that the ‘Overview’ of related work is more of a snapshot, and is incomplete.

The description of DogOnt is good, and explains well its capabilities. There are some more specific comments (see below) that could be further explored, or explained, if one is to back the claim that DogOnt can be applied beyond just the smart home. Also, many uses of DogOnt are referred to, changes made over the years. Without references though these claims cannot be substantiated. For these to be convincing, examples of the use of DogOnt should be provided.

The paper does also provide a simple methodology for using DogOnt. I would like this explored in greater detail. This could be very useful for persons who wish to use DogOnt. It would need to be tied to specific domains and stakeholders though, as I can imagine the range of tasks will vary depending on these. This highlights another concern in the paper. A clear definition of the target users is never given here (perhaps this is available in the 2008 paper?)

There are multiple spelling and grammar mistakes in the paper. I have listed a few at the bottom.

Specific Comments
-> Abstract mentions stakeholders...are sensors considered stakeholders?

Section 1 Introduction
Far too few references!
 ‘At the same time and for each level, it should take into account both electrical and thermal consumptions, and gather these information from a plethora of different stakeholders (i.e., citizens, utilities, policy makers, and energy providers) and heterogeneous sensors.’
o Are sensors stakeholders? Please provide a reference.
 ‘Among the available initiatives, the most renowned encompass the Linked Open Data (LOD) initiative, acting at the application level, the Semantic Sensor Web initiative, which aims at addressing, at least partially, the diversity of sensors and sensors data, and the Semantic Big Data research field aiming at tackling the data cardinality and heterogeneity issue.’
o Hard to parse this sentence. Please rework, and provide some references. SSW is abbreviated later.
 ‘Linked Open Data (LOD) provides machine understandable, shared and open semantics for representing a wide set of knowledge domains in the world. Rather than focusing on a single, rigid and practically not scalable representation model, the LOD approach integrates more than 295 datasets1 with over 31 billions of triples representing real data, from personal e-mail contacts to world nations, from medical topics to plane parts.’
o How is the LOD approach different from singular representation models? I think I understand what is trying to be said here, but the sentence could be worded better. Please clarify.
 ‘One of the most important results of the SSW initiative is the Semantic Sensor Network (SSN) ontology, defined by the W3C SSN XG group which was active from 2009 to 2011 [1]’
o I would update your reference…why refer to the working group? -
 ‘The Semantic Big Data initiative’
o Reference!
 ‘Current modeling approaches and initiatives, however, are too general for the energy context (e.g., SSW) or too specific, since they aim at modeling each level and domain separately.’
o References!, which approaches/initiatives are too specific?
 ‘In the past eight years, it evolved to tackle representation issues emerging from residential, building, and factory automation solutions. Lately, it included primitives for dealing with distributed networks of sensors deployed as part of smart buildings. Nowadays, DogOnt empowers several research projects needing uniform, semantic access to environment sensors and actuators and successfully supports abstraction of several standards including both Internet of Things (e.g., ZigBee) and non-IoT (e.g., Modbus) technologies.’
o Sounds great, but please reference each piece of work, along with projects DogOnt is being used in.

Section 2 Overview on AEC/FM modelling
 ‘This need is currently acknowledged by several research efforts, both industrial and academic, which aim at building domain ontologies to model energy consumption and performance. In the energy domain, ontologies are employed to define shared and common inter-language for performance evaluation, energy rating, device consumption profiling, etc. Approaches present in the literature, typically, address the energy domain by splitting the analysis along different forms of energy, i.e., electrical and thermal. On the one hand, this division permits to tackle the specificity of the single energy form and the related engineering domains. On the other hand, it prevents a structured and comprehensive approach to energy representation, at higher levels of detail, like at the district level.’
o References

Section 2.1 Electrical Sub-domain
 PowerONT
o This was developed by the author. Maybe this should be made explicit.
o This is built upon the analysis of over 23 assets, which include PowerONT and DogOnt. I see MIRABEL and also ThinkHOME mentioned. Would be good to see some awareness of the other ontologies analysed by SAREF, and their relation to DogOnt.

Section 2.3 City and District
 No mention of CityGML for district modelling? Again, feeling that this section is incomplete as an ‘overview’, more a snapshot.

Section 3 DogOnt
3.1 Overview
 ‘In the past years, it underwent several reviews and amendments, and its scope was widened to include devices and technologies typically part of an indoor IoT network.’
o Would be good to again see references, evidence of these extensions

3.2 Device, Environment Modelling
 These sections are well explained, reflecting the capabilities of the ontology
 For environment modelling though, is there any way of locating devices other than the room they are in, and can rooms be broken into zones, or sub spaces? (Unlike homes, some rooms can be fairly large in offices and factories, what if a sensor is on the ceiling in a corner of one of these spaces?).
3.3 Modelling Walk through
 This methodology is important. Giving people clear guidelines how to use an ontology, would really help. Could this process be captured graphically though, perhaps as a BPMN or activity diagram? Might help to make decision points clearer.
 “As a general hint, in this phase, the more quick approach to browsing is “reasoning” by systems: the plug is part of a general electric system /plant, and it is something that can be controlled.”
o Could you explain what is meant by reasoning in this context? Is this automated (or intended to be)?
 ‘The same simplification is applied to graphical representation where the containing room is represented by omitting related concepts, such as walls, floor and ceiling, etc.’
o How are the rooms represented graphically using contains relationships? Is there a coordinate system? I may have missed something here.

Section 5 Conclusion
 ‘'last edition of DogOnt'
o Should this read 'latest', or 'most recent'? Could explicitly say what the 'latest modification' to DogOnt is in conclusion.
 'Proposed mappings' –
o perhaps reword as 'The proposed mappings'
 'Additional efforts are needed to explicitly address policy regulations for the energy market, as this aspect is crucial for the successful exploitation of ontology based energy profiling.'
o This is the first time policy regulations are mentioned (other than as users in the abstract). This needs to be addressed in the paper, if it is important with respect to the work presented.

There are quite a few grammar and spelling mistakes. Here are a few:
‘Differently from the initial modeling approach’
‘The modeling approach to follow, which finally leads to the result reported in Figure’ (figure number not given)
‘we finally obatin the result in Figure 9’
‘The istantiation process’
‘The Artificial Intelligence’ – remove ‘the’
‘SEMANDO HEAD’ – Fig 13.

Review #2
By Laura Daniele submitted on 09/Jun/2017
Review Comment:

This paper presents the DogOnt ontology and relates it to the most relevant models and standards in the Architecture/Engineering/Construction (AEC) and Facilities Management (FM) domain. The first publication of DogOnt is dated 2008 and this paper reflects its latest evolutions and presents perspectives for its adoption.

I recommend the paper for publication once my main remark is addressed (see below).

The paper is very well written and easy to read. It provides an excellent survey of current energy modeling efforts and clearly shows how DogOnt relates to these efforts, also through the explicit high-level mappings presented in Section 4. The ontology is well described, providing a clear overview of the main classes and also a useful modeling walk-through that concretely shows the modeler how to create instances based on the schema.

DogOnt is a sound and solid ontology, designed according to high quality standards, and it is a relevant piece of work in the energy domain, as correctly claimed by the authors. Although it is an extensive ontology with a considerable number of classes and axioms, it is well structured and documented, making it understandable and usable by others. Its relevance is proved by the fact that important standardization activities supported by the European Commission and Standard Developing Organizations (SDOs), such as SAREF from ETSI, are based on DogOnt, and other relevant ontology efforts can be directly related to it, as shown in Section 4. It could be further stressed in the paper (footnote 3) that DogOnt was among the most important input sources used in the creation of SAREF.

The main remark to the authors concerns the relation of DogOnt to SAREF. The paper refers to the first release of SAREF, published by ETSI in November 2015. However, in the meantime important updates took place and some extensions were created (see Therefore, it is important that the authors revise the survey in Section 2 to reflect these updates. The new SAREF specifications relevant for the purpose of this paper were published in 2017 and are the following:
• TS 103 264 v2.1.1 (SAREF 2.0) ,
• TR 103 411 SmartM2M; Smart Appliances; SAREF extension investigation
• SAREF for Energy (SAREF4ENER): TS 103 410-1 SmartM2M; Smart Appliances Extension to SAREF; Part 1: Energy Domain
• SAREF for Building (SAREF4BLDG): TS 103 410-3 SmartM2M; Smart Appliances Extension to SAREF; Part 3: Building Domain

Of particular importance in the context of DogOnt are the latest SAREF release together with the SAREF4ENER and SAREF4BLDG extensions. More details on SAREF4ENER can also be found in [1]. Main changes in the latest SAREF release include the following:
• The classes and properties related to how to represent devices in building spaces have been removed from SAREF and incorporated into the SAREF4BLDG extension related to buildings.
• The information specific for energy efficiency has been moved to the SAREF4ENER extension. For example, the “Profile” concept has been redefined to accommodate only the properties that are general enough for any type of profile, not only for energy and power. Details on how to specifically model a power profile can now be found in the SAREF4ENER extension.

Finally, a few editing remarks:
• Page 9: the definitions provided for the two last points of the bulleted list at the end of the page, i.e., “OnOffState” and “SinglePhaseActivePowerMeasurementState” do not explicitly convey the message that these are the states in which a device can be found (e.g., the term used in the current definition of “OnOffState” is “ability” which is suitable to define functionality rather than state). I would suggest to improve the current definitions accordingly, for example:
- “an OnOffState, modeling the ability to be providing power to connected devices, or not;” could be replaced with “an OnOffState, modeling the state assumed by the plug, reflecting its ability to be providing power to connected devices, or not;”
- “a SinglePhaseActivePowerMeasurementState, representing the currently measured consumption value, and the relative unit of measure.” could be replaced with “a SinglePhaseActivePowerMeasurementState, representing the current state in terms of the currently measured consumption value, and the relative unit of measure.”
• Check the acronym AEC/FM domain throughout the paper. For example, in the first paragraph of Section 5 Conclusions (page 14), there are two occurrences where it should be “AEC/FM domain” instead of “AEM/FC domain”. Also at page 12 there are two some wrong occurrences.

[1] L. Daniele, M. Solanki, F. den Hartog, J. Roes: “Interoperability for Smart Appliances in the IoT World”. In Groth P. et al. (eds) Proceedings of the 15th International Semantic Web Conference (ISWC 2016), Kobe, Japan, October 19-21, 2016. Lecture Notes in Computer Science, vol 9982. Springer, Cham.

Review #3
By German Nemirovski submitted on 09/Jun/2017
Review Comment:

This article describes the DogOnt, an ontology for modelling of settings required for estimation of energy profiles (and probably operation and management of devices) at different levels of neighbourhood (city, district, building) while taking into account information collected from different stakeholders (citizens, utilities, policy makers and energy providers).
In general, this sort of papers must be welcome in the community that is still lacking interdisciplinary approached for the tasks mentioned above. The article gives a nice survey of ontologies and initiatives in the related context. It also describes nicely 1) the DogOnt’s key concepts: Building Thing, Functionality and State, 2) the modelling process – however, I think a cross-level modelling example is lacking at this place - and 3) on the rather technical level the possible interconnections between DogOnt and other ”modelling approaches” while illustrating those descriptions with rather sounding examples. Well, may be the only example that does not seem to fit is the one in the chapter 4.2 “Building to city and district mappings”, namely this example is hardly related to the title of the chapter.
However, the paper has two significant weaknesses: 1. In the very last paragraph of the conclusion, authors mention the “computation issues” related to the expressivity of the models. BUT, I did not find in the article a place where the modelling tasks (how the resulted models are applied) are clearly defined or at least listed. Those tasks defined on the application (knowledge domain) level determine the formal inference tasks to be solved, e.g. data querying or consistency checking, and in turn, the choice of the formal modelling instruments, i.e. the DL dialect, which would allow acceptable computability. Therefore, the selection of ALCHIQ(D) does not appear to me well motivated.
The second weakness is the complete lack of any sort of quality evaluation. I am not convinced, that DogOnt is so widely established and used ontology that the debates about its quality are superfluous.
If the time to address my comments can be given to authors before publication of the article, I’d appreciate it. In other case I still would not reject the article in its current state.