Modeling of information, knowledge and requirements

The Gellish Semantic Modeling Methodology is systematic approach that enables you to develop high quality information models of products and processes as well as of knowledge, requirements and rules and to store and exchange all of that in a computer interpretable and system independent way. The methodology applies generally applicable semantic principles of natural languages. Modeling conform those principles leads to a universal and extensible human as well as computer interpretable information models that can be stored and exchanged in universal semantic databases and data exchange messages. The methodology is independent of the used language and application domain.

The methodology provides a structured approach to the semantic modeling of information, knowledge and requirements. For various application domains provides a quick start by the availability of a basic dictionary-taxonomy and by guidance on how to express various kinds of facts that are typical for the application domain. The methodology also includes a systematic guide for adding definitions of domain specific concepts and terminology. The methodology guides you in the development of your proprietary domain dictionaries-taxonomies, knowledge bases and ontologies, object libraries, product catalogs, Facility Information Models, Building Information Models (BIMs), etc. It enables smooth integration of any kind of model by common use of a formal language, such as Formal English.

The methodology is based on universal semantic principles. The universal semantic concepts that resulted from a semantic analysis of natural languages are standardized and defined in the Gellish Formal English Dictionary-Taxonomy, together with their synonyms and abbreviations. Gellish Formal English is thus a formalization of natural languages and it is also based on a number of International ISO and IEC Standards. The English version of Gellish (Gellish Formal English) includes among others ordinary English terms, synonyms and abbreviations as well as phrases for kinds of relations that are used to create expressions. For example, the phrase <is a part of> is a standard (Gellish) Formal English phrase for a composition relation. It can be used to create expressions, such as "New York is a part of the United States".

Gellish English expressions can be stored and exchanged in the form of Data Tables, XML messages, RDF/OWL format and even in Excel spreadsheets.

The Gellish Semantic Modeling Methodology is documented in the following volumes that can be purchased via the webshop and that are free accessible for licensees:

  • Part 1: Architecture
  • Part 1A: General Principles and Guidelines
  • Part 2: Creation of Domain Dictionaries & Taxonomies
  • Part 3A: Knowledge Modeling
  • Part 3B: Specification of Requirements and Verification of Deliverables
  • Part 4: Managing Facility Information
  • Part 5: Modeling of Activities and Processes
  • Part 6: Development of Facility Models and Product Models (incl. BIM)
  • Part 7: Modeling of Rules and Regulations (in preparation)
  • Part 8: Integration of Product Models in 1D, 2D and 3D
  • Part 9: Modeling of Measurements and Observation

 

Types of Organizations that may benefit most

  • Organizations that produce knowledge and information expressed as documents and data files and that support the implementation of such knowledge in application systems in their branch and knowledge domains.
  • Engineers and contractors companies that deliver Facility Information for the facilities they design and construct.
  • Parts of organizations that are responsible for documenting and recording knowledge and standards and for optimizing the access and use of it.
  • Developers of products and installations, being modelers or users of knowledge.
  • Managers of documents and data about installations, plants, ships, etcetera, and the users of that information.
  • Database designers, data architects, application domain engineers.
For all those users it’s important that there is a clear and simple method to record and communicate knowledge and information in their branch and to integrate and smoothly exchange such information. The Gellish Modeling Method is the smart solution to make this possible.

 

Key characteristics of the Gellish Semantic Modeling Methodology

  • The method is based on a "common language", being a formal version of natural languages, such as English, Dutch (Nederlands,) etc.,  and includes a grammar and an electronic smart Dictionary-Taxonomy.
  • The method is based on universal semantic principles.
  • The language is extensible with domain specific concepts and terminology as well as proprietary concepts and terminology.
  • The method is universally applicable for any kind of data (models), including vocabularies, dictionary definitions of concepts (classes), taxonomies, knowledge models, requirements models, as well as product models, Facility Information Models, BIMs, including individual things)
  • The method applies very powerful semantic expression capabilities, exceeding current database capabilities and formal language capabilities.
  • The method enables the implementation of a powerful search method, without the need for a separate query language, and supports the simultaneous query of (multiple) databases.
  • The method is based on a development and implementation experience of over 15 years.
  • It is a proven technology.

The uniqueness of the Gellish Semantic Modeling Methodology

  • First operational method for semantically rich models (see best practices).
  • Very rich language with standard grammar (semantic definition of standard relation types).
  • Includes extensive domain Dictionaries / Taxonomies
  • Parts are standardized on ISO level.
  • Proven technology.
  • Quality assurance: Gellish@Work provides Semantic Verification and Certification services of models.

Benefits

  • Reduction of rework, because use of a common language with controlled consistency and unambiguous definitions.
  • Enables data integration from different sources without costly data conversions.
  • Enables data exchange between multiple application systems without costly data conversions.
  • Cost reduction of engineering by reuse of knowledge, reduction of data conversions and less ambiguity.
  • Computer system independent long term archiving capability.

Best Practice Application Examples

  • Harmonization of many SAP implementations world wide.
  • Facility Information Model of a Process Plant (including over 3000 pieces of equipment and 50.000 documents).
  • Specification of a Renovation Project of a Government building in the form of a Project model  (improved requirements management).
  • Project Models of of various Tunnels (e.g. Westerscheldetunnel, Coentunnel); avoiding problems as with the A67 tunnel.
  • National Knowledge Model for sewer systems, also including data requirements.

Other Examples of Applications

  • Shell: Facility Information Models, Knowledge Models and Requirements Models.
  • CROW and customers: CHEOBS Knowledge Models (Dutch Civil Industry).
  • RGD, Heijmans: IFD Knowledge Model (Dutch Building Industry).
  • International Compressor Manufacturers Association (Siemens, General Electric, etc.): Common Compressor Model.
  • Breda, Almere, etcetera.: City Knowledge Model, including also Sewer systems.
  • RWS, Cities, Provinces: OKSTRA operational in the Netherlands.

References

  • Keywords: Vocabulary, Dictionary, Taxonomy, Ontology, Object Library, Reference Data Library, Semantic Wave, Ontology Language, Web 3.0, Project Information Model, Building Information Model, BIM, Facility Information Model.
  • ISO 16354, Guidelines on Knowledge Libraries and Object Libraries
  • ISO 15926-2
  • ISO 15926-4
  • ISO 15926-11
  • ISO 12006-3
  • ISO 10303-221
  • PhD thesis Dr. Ir. A.S.H.P. van Renssen: Gellish, A Generic Extensible Ontological Language.