Understanding of systems engineering ontology

[Lan]

Dear all,

I am always confused when I try to think about the following question by myself. So I am hoping to hear your opinions rather than figure it out alone. You may find what I am going to ask is very naive. But I am ready for a criticism. 

This question is - when we say to develop an ontology for systems engineering, what do we mean by this “ontology” for "systems engineering" here?

To me, systems engineering is a discipline, which is really knowledge-intensive. This discipline provides people with approaches to successfully solve complex problems. The lifecycle processes are one of them and also the one that I am familiar with the most. So an ontology for lifecycle processes is not the same as an ontology for systems engineering, obviously. The latter is much bigger and broader. 

Moreover, I am really inspired by what Richard Martin said in another post. He pointed out that systems engineering need an ontology to unify the languages. That's exactly what I expect from the ontology for systems engineering. So does this mean that the ontology for systems engineering is a knowledge representation which contains all the terminologies (a.k.a. concepts) in systems engineering discipline and their relationships? Just like what the famous Gene Ontology brings to biomedical engineering - it includes all the terms related to biological process, molecular function and cellular component. Then, since this is the ontology (& knowledge management) working group, do we try to provide this ontology as deliverables or not?

As for the application of this ontology, it can be used as a big big "dictionary", or a standardization for the current and future standards of systems engineering, or an optional visualization of the knowledge in systems engineering, or a foundation for capturing the conceptual model of systems engineering, or a support for model-based systems engineering (I don't know how), or a medium for systems engineering knowledge management...etc.

I hope my understanding does not go against the original intention of this working group. 

I am eager to hear any feedback about how you think about this question, whether what I discribe here has already been realised, any suggestions on articles I should read to catch up.

I am also looking forward to tomorrow‘s WebEx meeting.

Best Regards,

Lan

[MASSON Davy (SAFRAN AIRCRAFT ENGINES)]

Hi All.

 

Like Lan, I’m a bit confused about the perimeter and the deliverables of this WG but I think it is mainly due to that fact that I’ve just arrived in this group… but not only if some many of you are asking these kind of questions.

What about defining all of this with a 5W&H approach ?

 

By the way, I’m also interested to know who you are. Are you postgraduate teachers, researchers, manufacturers, … ? Is there a presentation somewhere on the Google group ?

I’m asking this because it will be very difficult to converge on the definition of an ontology if there are 10 people with 10 different fields of expertise. TRC always says that lawyers have a completely different use of the ontology than us at Safran Aircraft Engines which is not absurd.

 

I do agree with Lan and Richard to say that ontology is here to unify the languages. The aim here is to share the same definitions or concepts behind a word. This is a pillar to develop complex systems, in particular when you are designing the functional architecture and writing the specification of your system. My little experience on it is that you cannot arrive with complete ontologies dedicated to every specific field and say : “Hey guys, take this ontology, it has been designed by specialists, trust me it’s perfect for your job”.

Take some stupid but real examples.

Here at Safran AE, I’m working with several guys who are debating on the function “To lubricate sump A” : for some people the aim of this function is just lubricate rotating parts; for others, it is “to lubricate + to evacuate heat”.

Another one : we have tried to define our dictionary based on bond graph theory so as to unify terms. But it is a bit of a failure because it has lead to generalize too much (one term for several synonyms). Sometimes, you have to admit that some terms are strongly rooted in some fields. Ex : a guy who designs engine compressor is not ready to abandon “to bleed” instead of “to discharge” or “to depressurize” even if “to bleed” is not natively dedicated to engines…

To sum up, there is change management to do because it is a brand new approach for a lot of manufacturers, most of whom do not understand the point of spend too much time on writing good specification…L

 

So, are our deliverables “generic” ontologies or shall we produce processes and methods to build an ontology ?

 

Etablissement de Villaroche Rond point René Ravaud     BP 42 F-77551  MOISSY CRAMAYEL CEDEX

Safran Aircraft Engines www.safran-aircraft-engines.com

 

 

De : ont...@googlegroups.com [mailto:ont...@googlegroups.com] De la part de Lan
Envoyé : jeudi 22 février 2018 19:43
À : Onto4SE <ont...@googlegroups.com>
Objet : [onto4se] Understanding of systems engineering ontology

--
You can see our wiki site for the Ontology WG at: https://sites.google.com/site/onto4syseng/
---
You received this message because you are subscribed to the Google Groups "Onto4SE" group.
To unsubscribe from this group and stop receiving emails from it, send an email to onto4se+u...@googlegroups.com.
To post to this group, send email to ont...@googlegroups.com.
Visit this group at https://groups.google.com/group/onto4se.
To view this discussion on the web visit https://groups.google.com/d/msgid/onto4se/98ae20bb-42bd-456a-b743-a157703f6e3e%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

#
" Ce courriel et les documents qui lui sont joints peuvent contenir des informations confidentielles, être soumis aux règlementations relatives au contrôle des exportations ou ayant un caractère privé. S'ils ne vous sont pas destinés, nous vous signalons qu'il est strictement interdit de les divulguer, de les reproduire ou d'en utiliser de quelque manière que ce soit le contenu. Toute exportation ou réexportation non autorisée est interdite Si ce message vous a été transmis par erreur, merci d'en informer l'expéditeur et de supprimer immédiatement de votre système informatique ce courriel ainsi que tous les documents qui y sont attachés."
******
" This e-mail and any attached documents may contain confidential or proprietary information and may be subject to export control laws and regulations. If you are not the intended recipient, you are notified that any dissemination, copying of this e-mail and any attachments thereto or use of their contents by any means whatsoever is strictly prohibited. Unauthorized export or re-export is prohibited. If you have received this e-mail in error, please advise the sender immediately and delete this e-mail and all attached documents from your computer system."
#

[Paola Di Maio]

Lan

The broad topic of my PhD research is 'knowledge systems', so I am glad this is resonating both with your thread and Swami's talk.   Thanks for bringing up the issue of scoping the effort.

. 

This question is - when we say to develop an ontology for systems engineering, what do we mean by this “ontology” for "systems engineering" here?

its always good to ask 'the capital question' (Gangemi) 

To me, systems engineering is a discipline, which is really knowledge-intensive. This discipline provides people with approaches to successfully solve complex problems. The lifecycle processes are one of them and also the one that I am familiar with the most. So an ontology for lifecycle processes is not the same as an ontology for systems engineering, obviously. The latter is much bigger and broader. 

SE, from what I have observed, is knowledge intensive, but the knowledge is 

often tacit, this was the main challenge of SeBOK  which when I came across it was

a first effort of its kind.  I invite everyone to take a look and give feedback.

Moreover, I am really inspired by what Richard Martin said in another post. He pointed out that systems engineering need an ontology to unify the languages. That's exactly what I expect from the ontology for systems engineering. So does this mean that the ontology for systems engineering is a knowledge representation which contains all the terminologies (a.k.a. concepts) in systems engineering discipline and their relationships?

Terminology and concepts are not (necessarily) the same. For me a concept is abstract and the

term describes the concept. Concepts can be communicated with drawings, art all sorts of non verbal representations. We can have multiple terms for one concept and viceversa.

 

Just like what the famous Gene Ontology brings to biomedical engineering - it includes all the terms related to biological process, molecular function and cellular component. Then, since this is the ontology (& knowledge management) working group, do we try to provide this ontology as deliverables or not?

As for the application of this ontology, it can be used as a big big "dictionary", or a standardization for the current and future standards of systems engineering, or an optional visualization of the knowledge in systems engineering, or a foundation for capturing the conceptual model of systems engineering, or a support for model-based systems engineering (I don't know how), or a medium for systems engineering knowledge management...etc.

For me, an ontology of SE  should be generalised enough to serve all of the above.

A unified language is found on UML and other unified approaches (which I am interested in and work on) it necessarily needs to be at the highest possible level of abstraction, so UML is very abstract.

Have you seen SysML?  Does that do anything that you require?

PDM

 

I hope my understanding does not go against the original intention of this working group. 

I am eager to hear any feedback about how you think about this question, whether what I discribe here has already been realised, any suggestions on articles I should read to catch up.

I am also looking forward to tomorrow‘s WebEx meeting.

Best Regards,

Lan

--

You can see our wiki site for the Ontology WG at: https://sites.google.com/site/onto4syseng/
---
You received this message because you are subscribed to the Google Groups "Onto4SE" group.

To unsubscribe from this group and stop receiving emails from it, send an email to onto4se+unsubscribe@googlegroups.com.

To post to this group, send email to ont...@googlegroups.com.
Visit this group at https://groups.google.com/group/onto4se.
To view this discussion on the web visit https://groups.google.com/d/msgid/onto4se/98ae20bb-42bd-456a-b743-a157703f6e3e%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

A bit about me

[Paola Di Maio]

Lan,

let me add

system ontology may be described as a set of concepts

engineering however may be a set of processes

A lot of interesting  relevant resources come up

with a search for

process ontology

Let us know if you find anything that helps

P

--

You can see our wiki site for the Ontology WG at: https://sites.google.com/site/onto4syseng/
---
You received this message because you are subscribed to the Google Groups "Onto4SE" group.
To unsubscribe from this group and stop receiving emails from it, send an email to onto4se+unsubscribe@googlegroups.com.
To post to this group, send email to ont...@googlegroups.com.
Visit this group at https://groups.google.com/group/onto4se.
To view this discussion on the web visit https://groups.google.com/d/msgid/onto4se/98ae20bb-42bd-456a-b743-a157703f6e3e%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Lan]

Hi Davy,

Many thanks for the reply.

I'm a second year PhD candidate from National University of Ireland, Galway. But the majority of the members in this group are not postgraduate students. I enclose the initial list of members members.xlsx of the first kickoff meeting at the INCOSE Symposium in July 2016 in Edinburgh. However, the list need an update now.

Thank you very much for sharing the interesting insights and examples in Safran Aircraft Engines. They remind me of an article written by Lin (2004) which also addressed the same problem. In the paper, it says "Commonly, people working within a particular company or group will develop their own vocabulary or common terms for particular issues, elements or activities with which they often work. Hence, when people are brought together from different groups or companies, two common types of problem in communication can occur. First, that the same term is being applied to different concepts (semantic problem). Second, that different terms may be used to denote the same entity (syntax problem)." 

To answer your last question, my humble opinion is that they can both become deliverables of this WG, 1) ontology the artefact and 2) the methodology to develop it. Together with my original question, I believe that the scope of the ontology is really important. "Ontology for systems engineering" seems to be too abstract and broad. Unless we can define a very clear boundary of the ontology - what we intend to include and what we don't.

Best Regards,

Lan

[Lan]

Dear Paola,

Many thanks for the comments. I’m going to respond the two of them in one post together.

Firstly, I very agree with you that terminologies and concepts are not the same. I also agree the relations you described between them. So, before starting the discussion, we are clarified with the languages between us, aren't we? ;)

Secondly, as for the topic of unified language, I have some experience of using UML in my previous studies. It is very commonly accepted and a huge success. SysML is what less

 familiar to me. I have heard of it many many times but haven‘t use it in my project. I also found that some authors are interested in the comparison of UML, SysML and OWL, as well as the transformation between them. I haven't drawn a deep look into the literature about this yet, but it is possibly a different perspective from which we study about ontology or knowledge systems for SE.

Next, about the SEBoK, I can imagine that it is not easy to summarise such a knowledge-intensive discipline into a well-structured knowledge resource. As Juan mentioned, the previous knowledge management group is responsible for the production of the SEBoK. Were you a participant? I can't stop thinking about the knowledge areas, topics and glossary of terms provided by the SEBoK, when I firstly come across them. I always consider that there should be some ways to re-organise the relations between them via an ontology, which possibly can add some value to the current version, but I am not sure how.

Finally, thank you for the suggestion on a search of "process ontology". Will definitely do that.

Best Regards

Lan

在 2018年2月24日星期六 UTC下午5:08:33，Paola Di Maio写道：

To unsubscribe from this group and stop receiving emails from it, send an email to onto4se+u...@googlegroups.com.

To post to this group, send email to ont...@googlegroups.com.
Visit this group at https://groups.google.com/group/onto4se.
To view this discussion on the web visit https://groups.google.com/d/msgid/onto4se/98ae20bb-42bd-456a-b743-a157703f6e3e%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Paola Di Maio]

Davy

I am not on Lan's list. Cannot remember the date of the Edinburgh event.

 I am based i Edin but away most of the time.

I am a researcher, over the years have become 'an academic' but my first experiences 

were in the real world- working in various roles in different sectors in many countries

mostly journalism and publishing. The real world upset me a lot so I spent a lot of time in the library.

Now I work mainly as lecturer, consultant and author/editor on various subjects

I got into knowledge systems as I followed my main life quests: understanding everything.

As you can imagine, my mind was blown up by this quest fairly early, and eventually realised understanding everything was a  non trivial cognitive challenge and not so much about knowledge itself which can be easily acquired and almost as easily forgotten bur rather as to our

ability to perceive and figure out  new things is constrained by other knowledge (what we were taught in school). That bothered me.

Later found that there is an entire knowledge domain devoted to 'enlightenment', where different rules apply.   That's what I am at more or less, right now. 

Long story. You asked.  :-)

----

Now to the question of ontology - uhm

In your view, in relation to the purpose of the ontology,

 is 'writing a good technical specification' a generic or a specific task?

I am asking this aiming to transcend dualistic view of reality

(which is somewhat related to UML)

On Fri, Feb 23, 2018 at 3:25 PM, MASSON Davy (SAFRAN AIRCRAFT ENGINES) <davy....@safrangroup.com> wrote:

Hi All.

 

Like Lan, I’m a bit confused about the perimeter and the deliverables of this WG but I think it is mainly due to that fact that I’ve just arrived in this group… but not only if some many of you are asking these kind of questions.

What about defining all of this with a 5W&H approach ?

 

By the way, I’m also interested to know who you are. Are you postgraduate teachers, researchers, manufacturers, … ? Is there a presentation somewhere on the Google group ?

I’m asking this because it will be very difficult to converge on the definition of an ontology if there are 10 people with 10 different fields of expertise. TRC always says that lawyers have a completely different use of the ontology than us at Safran Aircraft Engines which is not absurd.

 

I do agree with Lan and Richard to say that ontology is here to unify the languages. The aim here is to share the same definitions or concepts behind a word. This is a pillar to develop complex systems, in particular when you are designing the functional architecture and writing the specification of your system. My little experience on it is that you cannot arrive with complete ontologies dedicated to every specific field and say : “Hey guys, take this ontology, it has been designed by specialists, trust me it’s perfect for your job”.

Take some stupid but real examples.

Here at Safran AE, I’m working with several guys who are debating on the function “To lubricate sump A” : for some people the aim of this function is just lubricate rotating parts; for others, it is “to lubricate + to evacuate heat”.

Another one : we have tried to define our dictionary based on bond graph theory so as to unify terms. But it is a bit of a failure because it has lead to generalize too much (one term for several synonyms). Sometimes, you have to admit that some terms are strongly rooted in some fields. Ex : a guy who designs engine compressor is not ready to abandon “to bleed” instead of “to discharge” or “to depressurize” even if “to bleed” is not natively dedicated to engines…

To sum up, there is change management to do because it is a brand new approach for a lot of manufacturers, most of whom do not understand the point of spend too much time on writing good specification…L

 

So, are our deliverables “generic” ontologies or shall we produce processes and methods to build an ontology ?

 

Etablissement de Villaroche Rond point René Ravaud     BP 42 F-77551  MOISSY CRAMAYEL CEDEX

Safran Aircraft Engines www.safran-aircraft-engines.com

 

 

De : ont...@googlegroups.com [mailto:onto4se@googlegroups.com] De la part de Lan

Envoyé : jeudi 22 février 2018 19:43
À : Onto4SE <ont...@googlegroups.com>
Objet : [onto4se] Understanding of systems engineering ontology

 

Dear all,

 

I am always confused when I try to think about the following question by myself. So I am hoping to hear your opinions rather than figure it out alone. You may find what I am going to ask is very naive. But I am ready for a criticism. 

 

This question is - when we say to develop an ontology for systems engineering, what do we mean by this “ontology” for "systems engineering" here?

 

To me, systems engineering is a discipline, which is really knowledge-intensive. This discipline provides people with approaches to successfully solve complex problems. The lifecycle processes are one of them and also the one that I am familiar with the most. So an ontology for lifecycle processes is not the same as an ontology for systems engineering, obviously. The latter is much bigger and broader. 

 

Moreover, I am really inspired by what Richard Martin said in another post. He pointed out that systems engineering need an ontology to unify the languages. That's exactly what I expect from the ontology for systems engineering. So does this mean that the ontology for systems engineering is a knowledge representation which contains all the terminologies (a.k.a. concepts) in systems engineering discipline and their relationships? Just like what the famous Gene Ontology brings to biomedical engineering - it includes all the terms related to biological process, molecular function and cellular component. Then, since this is the ontology (& knowledge management) working group, do we try to provide this ontology as deliverables or not?

 

As for the application of this ontology, it can be used as a big big "dictionary", or a standardization for the current and future standards of systems engineering, or an optional visualization of the knowledge in systems engineering, or a foundation for capturing the conceptual model of systems engineering, or a support for model-based systems engineering (I don't know how), or a medium for systems engineering knowledge management...etc.

 

I hope my understanding does not go against the original intention of this working group. 

 

I am eager to hear any feedback about how you think about this question, whether what I discribe here has already been realised, any suggestions on articles I should read to catch up.

 

I am also looking forward to tomorrow‘s WebEx meeting.

 

Best Regards,

Lan

--
You can see our wiki site for the Ontology WG at: https://sites.google.com/site/onto4syseng/
---
You received this message because you are subscribed to the Google Groups "Onto4SE" group.

To unsubscribe from this group and stop receiving emails from it, send an email to onto4se+unsubscribe@googlegroups.com.

To post to this group, send email to ont...@googlegroups.com.
Visit this group at https://groups.google.com/group/onto4se.
To view this discussion on the web visit https://groups.google.com/d/msgid/onto4se/98ae20bb-42bd-456a-b743-a157703f6e3e%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

#
" Ce courriel et les documents qui lui sont joints peuvent contenir des informations confidentielles, être soumis aux règlementations relatives au contrôle des exportations ou ayant un caractère privé. S'ils ne vous sont pas destinés, nous vous signalons qu'il est strictement interdit de les divulguer, de les reproduire ou d'en utiliser de quelque manière que ce soit le contenu. Toute exportation ou réexportation non autorisée est interdite Si ce message vous a été transmis par erreur, merci d'en informer l'expéditeur et de supprimer immédiatement de votre système informatique ce courriel ainsi que tous les documents qui y sont attachés."
******
" This e-mail and any attached documents may contain confidential or proprietary information and may be subject to export control laws and regulations. If you are not the intended recipient, you are notified that any dissemination, copying of this e-mail and any attachments thereto or use of their contents by any means whatsoever is strictly prohibited. Unauthorized export or re-export is prohibited. If you have received this e-mail in error, please advise the sender immediately and delete this e-mail and all attached documents from your computer system."
#

--

You can see our wiki site for the Ontology WG at: https://sites.google.com/site/onto4syseng/
---
You received this message because you are subscribed to the Google Groups "Onto4SE" group.

To unsubscribe from this group and stop receiving emails from it, send an email to onto4se+unsubscribe@googlegroups.com.

To post to this group, send email to ont...@googlegroups.com.
Visit this group at https://groups.google.com/group/onto4se.

To view this discussion on the web visit https://groups.google.com/d/msgid/onto4se/826ee3bc5d274a40b97fade532562da0%40Y003OLE.rd1.rf1.

For more options, visit https://groups.google.com/d/optout.