Anatoly Levenchuk's Runglish Blog

My slides for second session of systems engineering track of Ontology Summit 2012 (it was published as additional slides here — http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2012_02_02 ):

My remark in this session was necessity to add methodologists (situational method engieers) to systems engineers and ontologists. Ontologists are not methodologists and thus need professional help to understand what is systems engieering method. We heavily use situational method engineering standards (e.g. ISO 24744) to understand what is systems engineering as a method of work. And after that we can start ontology work with systems engineering.

Today I deliver a talk “Ontology Engineering for Systems Engineering” at one of the session of Ontology Summit — http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2012_01_26

My slides:

This talk was about 20 minutes in English, but yesterday I presented the same topic at INCOSE Russian Chapter meeting about 3 hours (in Russian, with video: http://incose-ru.livejournal.com/32094.html).

The feature and usability release 0.87 at http://techinvestlab.ru/files/dot15926Editor/dot15926Editor087.rar Some features are really nice, please update!

New in version 0.87:
* Capability to add and edit template instances in preset Tablan.
* Correct visualization of Same As data attribute.
* Data source indication for data items viewed.
* Reload of entities for interrupted SPARQL request.
* Process interruption with window closure (mistake correction capability).
* Capability to access password protected SPARQL endpoints.
* URI dereferencing and web access for data model items and all data items.
* Some usability tips and tricks in documentation.

Next we plan feature release with template-based search to have possibility extract sub-graph from stored semantic network (we call it “scanner” because search engine will scan overall semantic network with template-based pattern matching).

Then we will add support of reporting (rendering, writing) this searched sub-graph in different formats (ISO 15926 part 8, .xls table, HTML, diagrams in GraphViz, etc.).

Next we will support OIMs and reference data library maintenance.

Then we open source code (we hope that to up to that moment architecture of .15926 programming platform for different plug-ins will be stable). This will be final release of .15926 version 1.0

Next really big thing will be support of DSLs with language workbench (projectional technology). This ISO 15926 ontology-based language workbench will be .15926 version 2.0.

The bugfix and usability release of .15926 Editor (version 0.83) can be downloaded from http://techinvestlab.ru/dot15926Editor

Some bugs were quite severe, including incorrect namespace identification. Please update!

New features: Part 2 membership rules and URI resolvability check.

Thank you all for valuable feedback! We expect more usability features in January 2012. Then we will move to development of template-based search and OIM support.

TechInvestLab.ru made the second public release in .15926 series. The .15926 Editor is a freeware and can be downloaded from http://techinvestlab.ru/dot15926Editor

Feature list of current version (0.81):

• The .15926 Editor is designed to search, navigate, create and edit data compliant to the ISO 15926 «Industrial automation systems and integration — Integration of life-cycle data for process plants including oil and gas production facilities» standard.
• With .15926 Editor you can search, navigate, create and edit data for all objects defined by the standard: individuals, relationships, classes, templates and template instances. You can work with reference and project data in the same environment.
• The .15926 Editor visualizes data according to ISO 15926-2 and ISO 15926-7,8 semantic models, showing item types, relationship and template roles and meta-data. The software allows customization of meta-data defined by a particular data-source.
• The .15926 Editor supports work with templates: new template creation, specialization from existing template, template signature editing and role restriction.
• The .15926 Editor can access and edit data in local RDF and OWL/RDF files, adding, changing and deleting data and meta-data in user-friendly multi-window environment with drag-and-drop and undo/redo capabilities.
• The .15926 Editor is preconfigured to search, navigate and edit popular reference data files distributed publicly, including ISO 15926-4, PCA RDL and ISO 15926-8 templates. The .15926 Editor easily handles local copy of PCA RDL (almost 3 ml. triples in memory, load time about 1 minute on average PC).
• The .15926 Editor can access data of a SPARQL-endpoint (provided it doesn’t require authorization) giving access to data compliant to ISO 15926-8.
• The .15926 Editor is preconfigured to search and navigate five popular public ISO 15926 SPARQL-endpoints.
• The .15926 Editor allows to save data downloaded from SPARQL-endpoints to local files, merge and split local data sources.
• The .15926 Editor can simultaneously visualize, navigate and edit data from several interrelated data sources located in local files and SPARQL-endpoints.
• The .15926 Editor works with three different namespaces for ISO 15926-2 (PCA RDL, RDS/WIP or Part8 URI’s) and can be customized to work with interrelated data sources with different namespaces.
• The .15926 Editor can visualize data in ISO 15926-8 formats, and in some legacy formats used by POSC Caesar Association.
• The .15926 Editor is configured to make а conversion of reference and project data from TabLan.15926 table language (represented as EXEL tables) to ISO 15926. The .15926 Editor can be configured to make a conversion of any structure table data to items defined by the standard: individuals, instances of relationships, classes and template instances. The mapping is defined in the Turing-complete domain-specific language (built on Python).
• The data prepared with.15926 Editor can be directly loaded to triple-store exposing ISO 15926-8 compliant data, allowing you to have sandbox RDL up and running in no time.

Current version work only in Windows. Screenshots and full docs you can see online at http://www.slideshare.net/ailev/15926-editor.

All docs and menus of .15926 software is in English, but discussions about development of .15926 software is mostly in Russian and can be found at http://dot15926.livejournal.com

Current roadmap lead to next version of .15926 Editor with support of OIMs and template-based search.

Yesterday I delivered a key talk “Future of Engineering” at CAE/PLM Multi-D Forum organized by NIAEP in Nizhny Novgorod (http://niaep.ru/ru/455/). Here is outline of engineering trends that I presented there:

1. Merging with software engineering:

• Software-intensive systems and cyber-physical systems as systems-of-interest
• Model-oriented engineering (virtual systems: testing in computer). System definition is completely done with software tools.
• Generative engineering (generation of engineering solutions by software: knowledge acquisition of engineering knowledge that suited not for humans but for software programs.
• According to BKCASE and GRCSE projects, systems engineering education should include full-fledged education of software engineer. Soon it will be the same for every engineering specialties.
• Conclusion: generation change of engineering technologies will be with the same impetuosity as in case of software technologies: one in five years.

2. Merging of engineering and engineering management.

• ISO 15288 have only one technical (i.e. “true systems engineering”) process from four. Other three are belong to engineering management disciplines.
• According to ASEM Handbook technology management and systems engineering are part of engineering management.  Also ASEM perform survey that show 50% of engineers obtain some of management functions during first 5 years of their work.
• Lean, agile, collaborative engineering, concurrent engineering and other non-cascade life cycle kinds requiring different organizations of engineers. Choosing one of life cycle kinds is management or engineering task?
• engineering thinking in management: factory physics, enterprise engineering.

Conclusion: engineering and management merged now de facto, in nearest future they tend to be joint discipline.

3. Explicating of engineering knowledge.

• System thinking and philosophical logic as a base discipline (not only mathematics).
• Engineering method research (engineering heuristics vs scientific theories), non-classical logics.
• Situational method engineering.
• Pattern languages, architectural patterns.
• Reference processes frameworks, bodies of knowledge, method components catalogs, methodologies.
• Generative design and generative manufacturing: computer should invent, not only simulate.
• Knowledge management and capturing of tacit knowledge.

Conclusion: art and mastery of today will be in curriculum tomorrow, level of art and mastery will be significantly higher.

4. New methods of systems definition.

• GORE
• Architectural languages (SysML and AADL is only first, ontology-based languages will be next).
• TRIZ, DSM.
• Multi-physic simulations and behavioral modeling.
• Model-oriented engineering and virtual mock-ups.
• Data-centric engineering and data logistics (data-centric information management).
• Collaborative design and concurrent engineering.
• Generative design.

Conclusion: old engineers needs retraining to perform system definition of nearest future.

5. New methods of system realization.

• New materials (composites, nano-, meta-).
• Additive manufacturing, 3D printing.
• Generative manufacturing.
• Robotics.
• New modularity, new logistics, new construction and assembly (Broad Group from China construct 15 store building in 6 days, preparing to construct 666 meters hight skyscraper in 4 months).

Conclusion: competition of engineering companies in nearest years will be more sever than you can imaging now.

6. Additional requirement from market and regulations.

• Safety greater than in nature environment.
• Environment cleaner than in raw nature.
• Non-defectedness (6 sigma: 3.4 defects for 1mln. products).
• De facto lessening role of intellectual property regulations.

Conclusion: changing requirement change the concept of “impossible”.

7. Expansion to non-traditional material.

• Software engineering (this is past trend but still important).
• Gene and bio-engineering.
• Molecular and nano-machines, meta-materials, drugs engineering.
• Enterprise engineering.
• Social engineering (ignoring all the critics).
• Underwater mining engineering.
• New physics (quantum devices, plasma, doubtful LENR and Mach effect devices).
• Systems of systems (including engineering of standard based extra large distributed systems, e.g. internet).

Conclusion: improbable earlier diversification of engineering and engineering specialties.

April 2011 was very busy at INCOSE Russian chapter. Usualy we have bi-weekly regular meetings. But in April 2011 there was two bigger activities: second Systems Engineering Challenges Workshop of INCOSE Russian chapter 7-10 of April 2011 at Bekasovo (near Moscow) and Complex technical systems: evolution of contemporary engineering disciplines conference that was organized 27 of April 2011 at Moscow by MISiS University together with several engineering holdings.

Systems Engineering Challenges Workshop is second such a meeting (first Workshop was situated the same venue one year ago) and gather INCOSE Russian chapter members from Moscow, Saint-Petersburg, Kiev. Main topic of workshop discussions this year was engineering breakdowns structures such as functional, system, work, location etc.. Term “breakdown structure” comes from project management (engineering management disciplines), engineering disciplines more often speaks about “decompositions” and “classifications”. We at INCOSE Russian chapter have apply ontology of ISO 15926 to establish relations types of different “breakdowns”, “decompositions”, “classifications”, etc.. There were four presentations about engineering and engineering management breakdowns with examples from construction industry, nuclear power plant engineering, shipbuilding.

Other topics included usage of ISO 42010 for engineering information systems architecture descriptions, applications of ISO 15926 for life cycle data integration of industrial equipment catalogs and building of formal requirements models, enterprise engineering with usage of DEMO method.

Russian language report about this Workshop you can find here: http://incose-ru.livejournal.com/24388.html

During this year Bekasovo Workshop it was not discussed only international systems engineering community experience. It was dramatic shift to our own systems engineering application projects that INCOSE members perform in everyday work.

MISiS Conference was devoted mainly to life cycle management talks that perfectly fit between engineering and engineering management fields. This conference was attended by many members of INCOSE Russian chapter. Moreover, at this Conference there was three presentations delivered by INCOSE Russian chapter members.

I  delivered at this conference presentation “Evolution of engineering practice of life cycle data integration” that was devoted to evolution from almost 20 years old life cycle data integration standards like STEP (ISO 10303) to contemporary  ISO 15926 and thus shifting of engineering data modeling community from object-oriented “close world” paradigm data integration to fact-oriented “open world” ontology-based contemporary paradigm.

It was a couple of hundreds representatives of Russian engineering companies and universities participated in this conference. Communiqué of this Conference directly addressed systems engineering as leading practice. This is mean recognition of systems engineering within Russian engineering community. Not bad result of INCOSE Russian chapter activity in recent couple years.

Russian language report about this Workshop you can find here: http://incose-ru.livejournal.com/24748.html

Next big event (not counting of regular Chapter meetings that we expect in May) with INCOSE Russian chapter active participation will be international forum “Intellectual design. Complex engineering objects life cycle management” that will be in Nizhny Novgorod 15-16 of June 2010 (Russian language info: http://niaep.ru/ru/455/). I will give there a key talk “Engineering management and systems engineering”.

Today TechInvestLab released second “ontology engine”: .15926 browser 0.4 alpha 4 (http://community.livejournal.com/dot15926/21692.html).

First “ontology engine” of  TechInvestLab (Communiware, first release was in 1998) still alive, and have usage to produce websites. Ten years ago it was a dream “rewrite Communiware for OWL”, it seems incredible cool in 2001 to have something related to OWL. Now .15926 engine eats OWL files in several varieties, but semantics of its ontology is not from OWL. “Semantic web” have only “transport format” usage for much more expressive semantic of ISO 15926. Discussion of developers also not about “semantics”. We already discuss “formal pragmatics” with ISO 15926 as А дискуссии разработчиков вообще проходят уже не на тему семантики, а на тему формальной прагматики, где ISO 15926 will be only one of many available semantics. This “pragmatics over semantics” version of .15926 is not for near future implementation, but wait 10 years as with Communiware and OWL…

This is my third ontology engine. Before Communiware it was Acquisitor (knowledge acquisition tool) that I was released 25 years ago not only as manager but also as Pascal programmer.

Start of .15926 software development was 1st of October, 2o10. There was thrown three consecutive version of code (in release 0.4 means fours rewriting from scratch). When started, we expected to release in March  2011 not only Browser, but Editor. But there was many unexpected difficulties: support of triple store for 2.7mln triples (this is current size of PCA RDS reference data library), show on screen 5 kinds of data from 8 exotic file formats, provide “bridge” for two of “almost right” RDL, have extensive research about metadata representation (and have extensive discussions with ISO 15926 community about metadata, still we have problems with it), etc.

Therefore we released only Browser (five times more advanced then we assume when start coding). Editor is scheduled for release later.

This is international release: help is in English, users expected from ISO 15926 community not only from Russia but all over the world.
So, currently I am not software engineering theorist but actual software project manager.

Separate story is about example reference data that we bundle to .15926 browser distribution kit. This is Rosatom RDL, and test project data of Rosatom (publicly available at http://iatom.ru/documents/others/). TechInvestLab was participant of ontology engineering project that lead to creation of this RDL. Thus I now have not only software development experience, but ontology development experience too.

Now we are waiting publishing of Shipbuilding RDL of Sudoexport. This is another engineering ontology that we participated in development.

TechInvestLab just published “Reference data engineering. ISO 15926 outside method description” document — http://techinvestlab.ru/files/RefDataEng/RefDataEngr_ver_2_25feb11.doc (19 pages in Russian, version 2.0).

This is first description that has not only new drafts of ISO 15926 parts in it base but also ISO 15288 system development terminology (with “reference data” as system-of-interest) and ISO 24744 as reference data life cycle description framework that defined content and structure of method description.

First part of “ISO 15926 outside” description consist of purpose and value validation for suggested method.

Then presented description of work products. After clearing the difference between information, data and information objects, we have a long list of clauses devoted to many of used artifacts: reference data of several kinds, project data, data model, full plant life cycle data pyramid, reference data libraries of several levels, mappings, data warehouses. All this with 4 pictures take 11 pages.

Next in document presented incremental reference data life cycle that consist of 6 stages:

1. Identification of project data sets.

2. Identification of data store definition items.

3. Data store definition items set characterization.

4. Mapping development.

5. Actual data transfer.

6. Verification of data transfer.

Each stage have a couple of pages description of practices that it used.

Then listed data modeling languages that support defined reference data life cycle: language of Part 2, template language, RDF and OWL.

Then method declare capabilities needed for people that fitting reference data engineer (data modeler) professional role.

Last section of text is about tools: reference data editor, mapping editor, data warehouse-specific adapter.

We have plans to translate this method description to English.

While this text devoted to “ISO 15926 outside” method, main focus of .15926 project still is in realm of “ISO 15926 inside”  approach and far outreached reference data engineering.

There are not too many methodology (i.e. way of work) literature about ISO 15926 implementation in real engineering environment. This is due extremely complex nature of ontology-based standard. Like software with at least computer science, software engineering as a separate disciplines, ontology have  “ontology science” and ontology engineering as distinct area of knowledge. Even management information systems as application-specific discipline can be reflected in it’s ontology-based mirror. Programming, modeling (as in MDA) and “ontologiing” is a very similar human activities and they are converging now. Look e.g. DDD (domain driven development) software engineering methodology.

We  already have several hundreds of pages in typical “Ontology handbooks”. Now we should rewrite these to very specific ISO 15926 methodology with “science”, “engineering” and “application-specific” parts. I expect that .15926M will be no less than several hundreds pages, that is comparable to any “general” ontology handbook.

We will structure content of .15926M according to several methodology description-related international standards:

ISO 24744 as a metamodel for methodology description. This standard require depict all kinds of work products (artifacts, including models and documents), works (processes, life cycle models etc.), organization (people and tools) and languages with their notations. This standard will be accompanied by ISO 24774 that prescribe hierarchy and content of work description (ISO 15288 conform to this standard).

ISO 42010 require present all models in descriptions grouped by needed to stakeholders views that conform to their explicitly shown viewpoints.

ISO 15288 is about lifecycle processes of systems engineering. We will refer to this standard in application part of our methodology (we consider that in ISO 15926 we have not management domain up to date but systems engineering due to life cycle integration nature and process plant engineering as application field. This will be expanded in future, but “engineering accent” will rest even in management applications. Think about “organization engineering”).

“Ontology science” part of methodology will cover 4D extensionalism, class-of-class classifications, work with information representations entities, template lifting-lowering and template characterization methodology, logic constraints (rules), different diagramming techniques and text notations, difference between true “upper ontology”-based standards like ISO 15926 and  other similar initiatives (like super-light ISO 12006-3). All implementation computer science specialties (EXPRESS, OWL, algebraic type representations, FOL and it’s syntax variants, queries and pattern matching, validation algorithms) will be here.

Ontology engineering (like any engineering) will be about ISO 15926-compliant schema development practice. Needs, requirements, architecture, coding, tools, development lifecycle, verification and validation, professional roles and overall ISO 15926 usage effort.

Application-related part of .15926M will be traditionally focused at process plant engineering domain (equipment catalogs, instrumentation templates, modified 4-level architecture of tag life cycle etc.). We hope also add to this some organizational engineering topics like presentation of methodologies (with .15926M presented in ISO 15926-compliant form as ultimate example). This part of methodology also will explain usage of .15926O (system thinking and methodology reference data that we developing in .15926 project — https://levenchuk.com/2010/11/14/dot15926-iso-15926-implementation-project/).

We are developing .15926M in Russian now but have plans to translate it to English.