CONNECTING EVERYTHING:  

A NEED IN SYSTEMS ENGINEERING AND A CLOSE FUTURE USING OSLC

Dr. Jose María Alvarez-Rodríguez (speaker) and  Prof. Dr. Juan Llorens

The Swiss Society of Systems Engineering Day

SESE 2014, Zurich,  Switzerland
 

The Initial Joke...

#me

The Problem...

The Problem...

Services (tool) Interoperability

Cross-cutting Services

Knowledge Management

Naming

Quality

Traceability

...

Impedance Mismatch

British to Australian problem..

Interoperability definitions

In healthcare, interoperability is the ability of different information technology systems and software applications to communicate, exchange data, and use the information that has been exchanged.

Source: Healthcare Information and Management Systems Society

The ability of a service to easily interact with other services ( from the same cloud service provder and from other cloud service providers). A KPI indicator

Source: Service Measurement Index by CSMIC consortium

and more and more...ETSI, IEEE SmartGrid, IEEE Intercloud, etc.

Communicate
Exchange
(Re)-use

data, information and knowledge between different agents

Types of Interoperability

 

 

  • Technical
  • Syntactic
  • Semantic
  • Business Process (Organizational)

 

 

  • Knowledge & Services
  • Social networks
  • Electronic identity
  • Cloud
  • Ecosystems

 

 

  • Data
  • Process
  • Rules
  • Objects 
  • Software systems
  • Cultural

...and issues

Interoperability issues

Measuring interoperability

  • The LISI Reference Model (a kind of CMMI)
  • Organizational Interoperability Maturity model
  • Enterprise Interoperability Maturity model
  • The Information Systems Interoperability Maturity Model (ISIMM)
  • ...

The  LISI Reference Model

Procedures, Applications, Infrastructure and Data

in other words....

 
 

Is OSLC & Linked Data a potential solution?

Interoperability

British to Australian solved!

Semantic web in a nutshell

  • Common and shared data model

    • Graph (subject, object, predicate)

    • RDF with different serialization formats

    • Implicit Multilinguism suport

  • Knowledge Representation & Management

    • Ontologies

    • OWL (Ontology Web Language)

    • Logic formalism: DL, F-Logic, etc.

    • Reasoning

    • Expert systems

    • Standards: Query Languages, Vocabularies, Datasets, ...

Graph

jose-foaf: http://www.josemalvarez.es/foaf.rdf#.
diego:foaf: http://berrueta.net/foaf.rdf#.

jose-foaf:me foaf:name “Jose”.

jose-foaf:me foaf:family_name “Alvarez”.

jose-foaf:me foaf:knows diego-foaf:me.

diego-foaf:me foaf:name “Diego”.

 RDF triples

Other example of Linked Data...

  • We are a public body

  • We want to buy "beds" for a new hospital

  • The call is open to any European entity

  • Data

    • Type of object (Common Procurement Vocabulary)

    • Location (Greece)

    • Multilingual environment 

    • Value / Quantity / currency

    • ....

Linked Data Principles

  • Use URIs to name things
  • When someone looks up a URI, provide useful information, using the standards (RDF*, SPARQL).
  • Use HTTP URIs.
  • Include links to other URIs. 

5* model

Linked Open Data Cloud

Database comparison

AS a RDF graph...

As Linked Data...

Extending Linked Data...

Web-based standards


Common and shared data model (RDF)

Open Services for Lifecycle Collaboration (OSLC)

OSLC Principles

  • Build on the WWW

    • HTTP REST

    • CRUD operations

    • Auth methods

  • Keep things simple

    • Consensus

  • Accommodate different schemas

  • Accommodate different representations

    • RDF/XML, JSON, ATOM, etc.

  • Align with the W3C Linked Data initiative

    • Follow the Linked Data principles

Specifications

*OSLC Core-OASIS TC

OSLC-based services

OSLC Core-Overview

OSLC Service Provider

OSLC Specification

Resources

Properties

OSLC-RM

(partial set of properties)

OSLC Preliminary Evaluation

  • A Linked Data approach for a specific domain

    • OSLC Resource shape similar to XML-Schema, SPIN or SheX.

    • How RDF is generated? 

    • Specifications under development

    • Authentication

      • HTTPAuth is recommend

      • OAuth was not created for these purposesEx

Similar standards...

STEP-ISO 10303

The Standard for the Exchange of Product Model Data

  • Manage Technical Product Data
  • Exchange data between tools. E.g. CAx ones
  • Several Application protocols (or specifications)
  • ...

Some initiatives to merge or use both

PLM scenario

The

Systems Engineering

context

Some existing methodologies

  • Requirements-driven engineering ( What? )
    • Focus on  requirements
    • E.g. RDE
  • Model-based systems engineering ( How? )
    • Focus on models
    • E.g. MDA, MDE, MDD, etc.
  • Model-driven requirements engineering ( What + How? )
    • Conjunction of models and requirements
    • E.g. MDRE

Requirements Driven Engineering (RDE)

Continuous review of requirements to fulfill stakeholder needs ...and...

Must be necessarily aligned to the construction of models, etc. to ease 

Validation

Verification

&

Model Driven Engineering

  • Requirements are used to define What?

  • ... and to serve as System Specification

But

They are sometimes isolated... (inception phase)

Approach

Try to elevate the meaning of natural language specs. through models (MDE)

Model Driven Architecture

A software design process providing guidelines for...

 

Structuring specifications through intensive use of

Models

Metrics

Transformations

...
 

Model Driven Development

A software development methodology

 

Create and exploit domain models

 

Application of MDA with focus on models

 

Attention of both industry and academy

Model Based Systems Engineering

...the formalized application of modeling to support system requirements, design, analysis, verification and validation activities beginning in the conceptual design phase and continuing throughout development and later lifecycle phases

  • consistent views

  • executable behavior

  • linked repositories

  • dynamic view generations

  • repeatable process

  • automatic change propagation, etc.

  • ... using models

Model Based Systems Engineering

Encompass system design and specifications

Adding new capabilities to ensure system 

Correctness


Consistency 


Completeness

Model Driven Requirements Engineering

=

MDE

+

RE

Some Open Issues...

  • Tangled environment of tools/techniques/methods
    • Data sources
      • IBM Doors, Reqtify, Rational Rose, RAT, KM etc.
    • Formats
      • XMI, ReqIF, etc.
    • Languages
      • ...
    • Models, services and APIs
      • XML-Schema, SQL, REST, WSDL/SOAP, etc.

...and more

  • Link of requirements to models ( and other artifacts ) to...
    • share a common vocabulary
    • model the X Breakdown Structures (e.g. PBS)
    • ensure quality: understandability, traceability, etc.

Impedance Mismatch?

Requirements to models problem...

Evolutionary requirements

INCOSE

Use of requirements as a basic unit of specification!

(NLP and Knowledge management are required!)

...and the V model?

From natural language

To   requirementsmodels, etc.

3 main challenges and key-enablers

  • Controlled vocabulary and ontology 

  • X Breakdown Structures (e.g. PBS) 

  • Services to manage domain knowledge

Cross-cutting services

Knowledge Management

Naming

Quality

Traceability

...

 

OSLC & Linked Data

 

Systems Engineering

 

&

A "car" as a system...

Taxonomy

LET'S START WITH THE VOCABULARY 

Bridging the gap between natural language and concepts using a standardized form

SKOS in a nutshell

  • S IMPLE

  • K NOWLEDGE

  • O RGANIZATION

  • S YSTEM

Use of SKOS

  • Controlled vocabularies
  • Taxonomies
  • Thesauri
  • Folksonomies
  • ...

Objectives

Define concepts and concept schemes

Operations

Store

Retrieve

Annotate

Link and merge

Concepts

SKOS Definitions

  • Classes and categories
    • skos:Concept and skos:ConceptScheme
  • Properties
    • Lexical labels
    • Notation
    • Documentation
    • Mapping
    • Semantic

STEPS

  • Model Concepts in SKOS
  • URI Design
  • Create concepts
  • Add authoring properties
  • Add labeling properties
  • Add notation properties
  • Add   documentation properties
  • Add semantic properties
  • Add mapping properties
  • Add concepts to a scheme
  • Describe the concept scheme

Similar to a Linked Data lifecycle...

A SKOS-based Car...

km-demo2:http://http://thereusecompany/km/demo2/.
dbpedia-res:http://dbpedia.org/resource/
<http://thereusecompany/km/demo/1381307095/car> a km-demo:System;
	dcterms:author  jose-foaf:me;			
	dcterms:created “2013-10-10"^^xsd:date ;    	
	dcterms:modified "2013-10-10"^^xsd:date ;.
	rdfs:label “Car”@en;
	skos:prefLabel “Car”@en;
	skos:prefLabel “Voiture”@fr;
	skos:altLabel “Motor”@en;
	skos:altLabel “Bus”@en;
	skos:altLabel “Wagon”@fr;
	skos:hiddenLabel “Auto-bus”@en;
	skos:notation "c1"^^xsd:string ;
	dcterms:subject "1"^^xsd:string ;  	
	km-demo:level "1";
	km-demo:related-to km-demo:Automative;
	skos:changeNote "An example of change note."@EN;
  	skos:editorialNote "An example of editorial note"@EN;
	skos:historyNote "An example of history note"@EN;
	skos:scopeNote "An example of scope note"@EN;.
	skos:narrower km-demo:car/subsystem/braking.
	skos:closeMatch dbpedia-res:Automobile;
	skos:exactMatch km-demo2:otherCar;
	skos:inScheme <http://thereusecompany/km/demo/1381307095/ds>.

SKOS Benefits

  • A concept-oriented model
  • RDF-based 
    • a common and shared data model
  • Crucial contribution to interoperability
  • A way of porting KOS to RDF
    • Align to existing standards (ISO 25964)
  • Lightweight reasoning
  • Simple but powerful
  • Extensible
  • Widely-used in existing semantic web vocabularies
    • Boosting the Linked Data initiative

Outcomes & On-going work

  • OSLC Vocabulary Management
    • A first prototype to tackle the aforementioned challenges through an OSLC interface for existing services (RAT)
      • A SKOS-based approach
      • Use of the Eclipse Lyo OSLC API and .NET OSLC API
  • OSLC Quality Management*
    • OSLC Adapter for RQA
    • Quantitative indexes in RDF 
      • To model (reuse existing quality metrics)  and
      • ... compute
      • ... quantitative indexes
  • Services for knowledge management

Overview

Vocabulary Management

Naming Service

Quality Management

As RDF...

Interoperability


OSLC & Linked Data


Systems Engineering

Conclusions

  • Re-use information and data to create a real collaborative environment
  • OSLC to ensure compatibility among tools
    • Linked Data in a controlled environment...
  • Requirements are  the key-entities  for the development of critical systems
    • Requirement as a basic unit of specification
    • Link requirements to models 
    • Quality is a MUST
    • Verification and Validation
    • Model-driven Requirements Engineering ( MDRE )

What's next...

  • Research
    • Identify more services (for knowledge  management) to be offered via OSLC
    • Index and retrieve any kind of artifact
    • Traceability as a  semantic-based retrieval process
  • Continuos engineering via OSLC
  • Collaborative complex systems development via OSLC
  • Other fields:
    • Multiplex network analysis
    • Expertise retrieval
    • ...

E.g. Index and retrieve anything

(math formulas, electrical networks, tables, etc.)

OSLC

KNOWLEDGE MANAGEMENT

The next big thing...?

For your attention!

Q & A

Credits

  • Speaker: Dr. Jose María Alvarez-Rodríguez
  • Some members of the KR roster...
    • Prof. Dr. Juan Llorens
    • Prof. Dr. Juan Miguel Gómez, 
    • Dr. Anabel Fraga
    • Dr. Valentín Moreno
    • Ms. Manuela Alejandres
    • Mr. Jose Fuentes (The ReuseCompany Inc.)

SESE-Zurich 2014

By Jose María Alvarez