PresentationPDF Available

Semantic Web in Industrial Companies - Methods, Architectures, Applications

Authors:

Abstract and Figures

Semantic Web Technologies short introduction and exploration of deployment architectures and application areas in industry, exemplified by b2b communication examples (today’s EDI domains) in Supply Chains.
No caption available
… 
No caption available
… 
No caption available
… 
No caption available
… 
No caption available
… 
Content may be subject to copyright.
Semantic Web in Industrial
Companies
Methods, Architectures, Applications
Prof. Hartwig Baumgärtel
2017/11/08
- restricted -
Copyright © Infineon Technologies AG 2017. All rights reserved. 2
About the Presenter
Prof. Dr.-Ing. Hartwig Baumgärtel
51 years, married, 2 daughters (22, 17)
Currently in research term in cooperation with Infineon AG, CSC E IN
2017-09-29 restricted
1988-1900 Mathematics (FSU Jena, TU Dresden, TU Berlin)
1990-1994 Computer Science (TU Berlin)
Majors: Artificial Intelligence and Complexity Theory
1994-1997 PhD thesis (in cooperation with Daimler Research & Technology)
Combination of Multi-Agent and Finite Domain Constraint Solving technology to solve a
production planning system at a car manufacturing plant“
Studies
1997-1999 Scientific Researcher; Daimler Research & Technology
Projects on decentralized automated production system control with Agents
2000-2004 Head of Daimler Research & Technology Supply Chain Management Team
Involved in foundation of automotive web platform Covisint
Created supply chain simulation tool SNS, applied in ~20 Daimler-internal projects
2004-2006 Consultant for process and workflow management in car development, Daimler
Professional
2006-today Professor for Logistics and SCM; University of Applied Sciences Ulm
B.Eng. and M.Eng. programs on Logistics/SCM
PhD class Cognitive Computing in cooperation with University Ulm
Professorship
Target Goals for this Presentation
Learn about the application areas
for semantic web in Industrial
companies & its potential
Get an insight into methods and
Architectures for semantic web
(like RDF, OWL, HTTP and
SPARQL)
… and preliminary answers on
how to apply it systematically
Ontology Engineering
3
2017-10-25 restricted Copyright © Infineon Technologies AG 2017. All rights reserved.
Agenda
Semantic Web Technologies
Application Areas in Industrial Companies
Semantic Web Deployment Architectures
Course of Action for SW-based B2B communication
1
2
3
4
Summary & Next Steps5
4
2017-07-11 restricted Copyright © Infineon Technologies AG 2017. All rights reserved.
Agenda
Semantic Web Technologies
Application Areas in Industrial Companies
Semantic Web Deployment Architectures
Course of Action for SW-based B2B communication
1
2
3
4
Summary & Next Steps5
5
2017-07-11 restricted Copyright © Infineon Technologies AG 2017. All rights reserved.
Semantics is a big step towards Industry
4.0 enabling…
… Big Data, Prescriptive Analytics,
Connectivity, Automation,
Business Agility, Internet of Things
A Step Towards Industry 4.0
Industry 1.0
Mechanization
Industry 2.0
Mass production
Industry 3.0
Computerization
Industry 4.0
Internet of Things
...but what is Semantics?
6
2017-06-21 restricted Copyright © Infineon Technologies AG 2017. All rights reserved.
Semantics and the World Wide Web
"The Semantic Web is an extension of the
current web in which information is given well-
defined meaning, better enabling computers and
people to work in cooperation."
“The Semantic Web” by Berners-Lee et. al., 2001, Scientific American, 29-37
Semantics Example
Used by
… that retrieves data from semantic sources
to understand and manage information.
Tim Berners-Lee
inventor of the World Wide Web
Knowledge
Information
Data
7
2017-06-21 restricted Copyright © Infineon Technologies AG 2017. All rights reserved.
Traditional Web vs. Semantic Web
The web is constructed upon machines who do the management of connecting and transferring raw data.
But how can these machines make any sense of it? All of the text and pictures on the internet are created
by humans for humans.
Thus, information is ultimately only human-readable.
Semantic web is about to provide information in ways machines can interpret.
By providing the information in a more computer-friendly way, we enable smart web analytics and can
bring automated operations forward and into reality.
8
2017-02-01 restricted Copyright © Infineon Technologies AG 2017. All rights reserved.
Eva has two friends: Adam and
Kate.
Eva likes a lot of things on the
Internet, for instance YouTube
and Last.fm.
Eva saw Peter‘s photo and liked
it!
Traditional Web SEMANTIC WEB
Links
information
on
document
level
For people
based on
XML which
is
machine-
readable
Links
information
on data level
For machines
& people
based on RDF
which is
understand-
able for
machines
Resource Description Framework (RDF)
The Resource Description Framework (RDF) is a fundamental concept for expressing semantic data.
RDF is a method for modeling resources as well as their intrinsic properties and relationships using series of
logical assertions.
RDF was defined by the W3C and requires some specifications:
Semantic Specification:
Each resource or property must be uniquely identified (URI, IRI)
Relational Specification:
Resources should be defined according to intrinsic properties or relationships to other resources
Ontological Specification
These relationships must be presented as logical assertions that are consistent with the universe
This That
…is somehow related to… Resource
Property
9
2017-02-01 restricted Copyright © Infineon Technologies AG 2017. All rights reserved.
10
2017-10-09 restricted Copyright © Infineon Technologies AG 2017. All rights reserved.
This That
…is somehow related to… Resource
Property
RDF expresses data in triples with subject, predicate, and object. Predicates are called properties.
An object property relates things with other things, a data property connects a thing with a data literal.
The subject of a property is called domain, the object is called range.
Each resource and each property has a unique identifier (URI or IRI). Class and property URI’s refer to
the address (server + path) where the ontology which defines them is stored.
Subject Predicate Object
Class Object Property
DOMAIN RANGE
Class
Class Data literalData Property
Data Property
Resource Description Framework (RDF)
Copyright © Infineon Technologies AG 2017. All rights reserved. 11
Ontology and Individuals
An ontology is intended to organize data about a certain domain.
Therefore, it needs concepts like classes, class hierarchy, properties, property hierarchy,
property characteristics, domain and range of properties and basic data types.
The basic data is intended to be modelled as individuals in an ontology language.
Individuals are elements of classes.
Individuals may have properties and hence relations with other individuals.
Individuals may have properties which can be expressed by data types (literals).
The data of individuals is stored in so-called RDF-graphs.
Individuals have also URI’s. They refer to the address (server + path) of the graph store in
which they are initially defined.
2017-09-29 restricted
John is a Person. The age has the value of 42.
A person has an age.
Person Age
…hasAge… Resource
Property
Copyright © Infineon Technologies AG 2017. All rights reserved. 12
RDF (Resource Description Framework)
Data model for things (“resources”) and relations
between them
Provides a simple semantics for this data model
These data models can be represented in an XML
syntax
2017-09-29 restricted
Subject
Artefacts (things)
Object
Artefacts (things) or
Literals (string, integer, etc.)
Properties of subjects
also called
Attributes or Slots
RDFS (RDF Schema)
Extension of RDF
Adds more vocabulary for describing properties and
classes of RDF resources
With a semantics for generalization-hierarchies of
such properties and classes
OWL (Web Ontology Language)
Extension of RDF and RDFS by many useful concepts
Adds more vocabulary for describing properties: e.g. disjointness,
cardinality, equality, symmetry,
Adds more vocabulary for describing classes, e.g. enumerated classes
This That
…is somehow related to… Resource
Property
Ontology Languages
Copyright © Infineon Technologies AG 2017. All rights reserved. 13
Ontology and Individuals
2017-09-29 restricted
Parent isA Class
Mother isA Class
Mother subClassOf Parent
Child isA Class
isMotherOf isA Property
isMotherOf domain Mother
isMotherOf range Child
Based on “Introduction to Ontology Concepts and Terminology” by Steven J. Miller (2013)
isA
Child AdamJ
Taylor
Mother isA MariaI
Taylor
isMotherOf
subClassOf
Parent
Mother
Child
isMotherOf
Ontology Proper (here: family relationship ontology)
Individual Statements (for a specific family)
MariaITaylor isA Mother
AdamJTaylor isA Child
MariaITaylor isMotherOf AdamJTaylor
defines structure
For more information please look to SC Colloquium presentations by L. Mericle (2017-02-01) and A. Söderström et.al. (2017-06-21) at SC Colloquium iWiki
Agenda
Semantic Web Technologies
Application Areas in Industrial Companies
Semantic Web Deployment Architectures
Course of Action for SW-based B2B communication
1
2
3
4
Summary & Next Steps5
14
2017-07-11 restricted Copyright © Infineon Technologies AG 2017. All rights reserved.
Application Areas in Industrial Companies
Marketing and
Sales Application
Internal
Application
15
2017-06-21 restricted Copyright © Infineon Technologies AG 2017. All rights reserved.
Product catalogues
E.g. semantic product catalogues in Open
Linked Data
Current topic in Productive 4.0 (EU funded)
and SmartStage (BMWi funded)
2009
2011
196,000 + Unique Vehicles
<Cars
Vans
Busses
Trucks
Knowledge management, e.g. for large
IT projects with many stakeholders
Data warehousing, e.g. for spare parts
in a car repair information service
Vocabulary 1
Vocabulary 2
Vocabulary 3
Vocabulary 4
Searchable
Knowledge Graph
Semantics for Dummies (Taylor 2015)
The development of Open Linked Data (Glimm 2016)
Application Areas in Industrial Companies
Procurement and SCM
Application
Business Transaction
Application
16
2017-06-21 restricted Copyright © Infineon Technologies AG 2017. All rights reserved.
Unifying supplier catalogues
Knowledge management on potential
suppliers and their offers
Information exchange for tender
processes / RFQs
Order and Call-off processing
Agenda
Semantic Web Technologies
Application Areas in Industrial Companies
Semantic Web Deployment Architectures
Course of Action for SW-based B2B communication
1
2
3
4
Summary & Next Steps5
17
2017-07-11 restricted Copyright © Infineon Technologies AG 2017. All rights reserved.
Copyright © Infineon Technologies AG 2017. All rights reserved. 18
Ontology Development and Deployment
2017-09-29 restricted
Ontology
Development
Ontology
Deployment
Tools:
Ontology Editor
Reasoner
Visualizer
Engineering:
Develop own ontologies
Import existing ontologies
Merge several ontologies
Create test data (individuals)
Test and validate ontology
Data generation and input
Mass data conversion to graphs,
e.g. RDB2RDF
Deployment architecture (where
ontology and data are provided):
Graph stores
Web interfaces
SPARQL protocols
Web crawlers
Analytics and learning
algorithms
Security and authentication
Semantic Web Deployment Architectures
Linked Open Data (LOD)
Today’s Internet-(TCP/IP)-based networks Today’s and expected Internet-(TCP/IP) + semantics based networks
Linked Open Data – “the semantic web”
DBpedia, Europeana, GeoNames, FOAF, …
Open access network
How to publish at Linked Open Data:
http://www.euclid-project.eu/
World-wide web
(“the web”)
Logical architecture:
Ontologies and data from different sources
are connected (interlinked) by object properties and
queried by browsers, mashup services or search engines
Software components architecture:
RDF graph stores (Triple stores), SPARQL services,
Adapters, Mapping tools, ontology stores working
together in the existing TCP/IP and HTTP based Internet
Source: W3C
19
Copyright © Infineon Technologies AG 2017. All rights reserved.2017-09-29 restricted
Semantic Web Deployment Architectures
Company-internal intranet and internet server
Today’s Internet-(TCP/IP)-based networks
Company internal semantic servers or networks
Cross-department knowledge management (e.g. Marketing, Sales,
Product development, Production, Procurement, SCM / Logistics, IT)
Cross-department workflow and data flow management
Intranets
Today’s and expected Internet-(TCP/IP) + semantics based networks
Internal Domain
Public Domain
Internal Open Data External Open Data
Logic
(php)
Logic
(php)
Logic
(php)
same as
same as
20
Semantic Web Deployment Architectures
Today’s Internet-(TCP/IP)-based networks Today’s and expected Internet-(TCP/IP) + semantics based networks
(Open) linked data – “the semantic web”
DBpedia, Europeana, GeoNames, FOAF,
Open access network
World-wide web
(“the web”)
Cross-industry community semantic networks
Company directory
Product catalogues
Secured network, restricted access
Examples: Industrial Data Space (IDS), Productive 4.0
Horizontal
e-Markets /
e-biz Platforms
Industry sector-specific semantic networks
Company directory
Product catalogues, product classification schemes, numbering systems
Trade initiation (e.g. tenders, RFQs, quotes)
Collaboration initiation (e.g. R&D, production, service cooperation)
Vertical
e-Markets /
e-biz Platforms
Supply chain-specific semantic networks
Network for one Company and it’s direct customers and suppliers
Business transactions (e.g. contract negotiation, call-off, order, advises)
Cross-company workflow and data flow management
Extranets,
Supplier Portals,
EDI Connections
Company internal semantic servers or networks
Cross-department knowledge management (e.g. Marketing, Sales,
Product development, Production, Procurement, SCM / Logistics, IT)
Cross-department workflow and data flow management
Intranets
21
Semantic Web Deployment Architectures
Cross-industry sample: Industrial Data Space
System layer of architecture, components
Roles and Interactions in the IDS
“The Industrial Data Space is a virtual data space
leveraging existing standards and technologies, as
well as accepted governance models,
to facilitate the secure exchange and easy linkage of
data in a trusted business ecosystem.
It thereby provides a basis for smart service
scenarios and innovative business processes, while
at the same time ensuring data sovereignty for the
participating data owners.” (Source: IDS)
Started as german public founded project
Driven by several Fraunhofer Institutes
Meanwhile IDS Association founded
Currently ~80 members, among them ZVEI
IDS Reference Architecture Model (06/2017)
Security and identification
22
Semantic Web Deployment Architectures
Supply Chain specific architectures
Subscription
Published
Data view
Published
Data view
Chosen
Company
Data in
Semantic
format
Chosen
Company Data
in Semantic
format
Company
Data
Company
Data
Subscription
Supplier Customer
(Https)
RDF Graph Store
SPARQL end point
Web Interface
SPARQL query & update app
e.g. Oracle DB 12c EE, Stardog,
OpenLink Virtuoso,
Apache Jena Fuseki
Enterprise Connectors, RIAs,
Data Managers, …
SPARQL query, update
SPARQL Graph Store HTTP protocol
HTTPS, Tunneling (TLS), Authentication (PKIs)
23
Cf. A. Söderström et.al., 2017
Semantic Web Deployment Architectures
Supply Chain specific, using many ontologies
www.w3.org
Organizations
ontology
www.apics-
scc.org
SCOR
ontology
www.vda.de
Glossary
VDA5002
www.vda.de
EDI
messages
VDA49xx
www.vda.de
Transport
Label
VDA4902
www.bme.de
EDI messages
bmeCAT
www.gs1.org
GS1 transport
label
www.gs1.org
EANCOM EDI
messages
www.bme.de
EDI messages
openTRANS
resources.gs1
us.org
Rosetta-
Net
www.icc.org
Incoterms as
ontology
www.unece.org
UN/EDIFACT
Supplier Customer
S2.1
S2.2
S2.3
S2.4
D2.1
D2.2
D2.x
D2.11
Order
RDF
graph
Refers to
S2.5
ASN as
RDF
graph
Retrieves
Retrieves
D2.13
D2.12
Ships
Transport
label
Refers to
Refers to
24
www.gs1.org
EAN-128 data
elements
www.bme.de
eClass
Network
with public,
cross-industry,
and specific
ontologies
Ontology and Data Levels
Ontologies and user data can occur on all architecture levels.
Ontologies may be classified in a corresponding manner in
Global ontologies (should be available in open linked data)
Cross-industry ontologies
Industry branch-specific ontologies
Supply chain specific ontologies
Company-specific ontologies
Data will be typically more restricted than ontologies which structure it.
Data should refer to ontologies from higher or same architecture level
The higher (less restricted) the level of an ontology is, the more potential users
it can have
25
Agenda
Semantic Web Technologies
Application Areas in Industrial Companies
Semantic Web Deployment Architectures
Course of Action for SW-based B2B communication
1
2
3
4
Summary & Next Steps5
26
2017-07-11 restricted Copyright © Infineon Technologies AG 2017. All rights reserved.
1. Determine the domain and scope of the ontology
2. Establish requirements
3. Check existing ontologies and related material (e.g. standards)
4. Develop an Entity-Relationship-Diagram
1. Define classes and class hierarchy
2. Define properties
5. Use ontology editor – (e.g. by Protégé)
1. Add information from existing ontologies (import)
2. Implement classes and class hierarchy from ERD
3. Define object- and data properties and their characteristics
4. Create sample individuals, test, verify and validate
5. Create sample SPARQL queries
Ontology Development for B2B Transactions
Knowledge-engineering methodology
27
2017-07-11 restricted Copyright © Infineon Technologies AG 2017. All rights reserved.
Copyright © Infineon Technologies AG 2017. All rights reserved. 28
Example:
Ontology Development for B2B Transactions
Ontologies domain and scope:
Business transactions in B2B environment
Process support and communication acts for operational business transactions in a SC
Requirements:
Ontology has to map reality in B2B business transactions
Ontology has to cover relevant concepts, processes, standards
Ontology has to be developed with focus on later deployment use
Relevant use cases should exist
Existing knowlegde and material:
Concepts
Processes
IT standards (B2B communication: EDI standards)
Existing Ontologies from the domain
2017-09-29 restricted
Copyright © Infineon Technologies AG 2017. All rights reserved. 29
Existing Knowledge and Material (1):
B2B Company Relations and Roles*
2017-09-29 restricted
Supplier Customer
Principal
Freight Forwarder
Shipper
(Ship-from party)
Consignor
Carrier Consignee
(Ship-to party)
Forwarding
contract
Contract of
carriage
Forwarding instructionWay bill
Delivery Contract
From Plant
Free Plant
documents
produces and sells products to
is Customer of
is Customer of
hands over goods to
Closed
between
closed
between
is usually**
is (in case of EXW, …)
is site of (e.g. Plant, Wareh.)
is Site of (e.g. Plant, DC)
Closed between
with Incoterm
with freight terms
has on board
*: partially based on DSLV and VDA concept definitions **: except FF conducts physical transport himself (own-name transaction)
is (in case of DDP, …)
hands over goods to
refines for specific
consignment
Source: Prof. H. Baumgärtel, University of Applied Sciences Ulm)
Copyright © Infineon Technologies AG 2017. All rights reserved. 30
Existing Knowledge and Material (1):
Concepts for deliveries and transports
2017-09-29 restricted
order and delivery related
Load
Consignment
Delivery
SKU
Article
transport related
1:1
n : m
Source: Prof. H. Baumgärtel, University of Applied Sciences Ulm)
Means of transport
Transport unit
Load unit
Handling unit
Article
Copyright © Infineon Technologies AG 2017. All rights reserved. 31
Existing Knowledge and Material (2):
Logistics Glossary of VDA and DSLV
The glossary contains definitions of logistics terms as well as several graphs explaining their
relation (current version: V2 published in 2016)
Authoring associations:
VDA: Verband der deutschen Automobilindustrie
(German Association of Automotive Industry)
DSLV: Deutscher Speditions- und Logistik-Verband
(German Association of Freight Forwarders and Logistics Service Providers)
Accessible at https://www.vda.de/
Information type: pdf document, 30 pages, 2 columns: German / English
136 concept definitions, 10 explaining graphics:
Transport chain
Transport activities
Packaging and transport units
Date and time
Transport and delivery documents
Graphical explanations
2017-09-29 restricted
Copyright © Infineon Technologies AG 2017. All rights reserved. 32
Existing Knowledge and Material (2):
Logistics Glossary of VDA and DSLV
2017-09-29 restricted
Copyright © Infineon Technologies AG 2017. All rights reserved. 33
Existing knowledge and material (2)
B2B fulfillment strategies and processes
2017-09-29 restricted
Copyright © Infineon Technologies AG 2017. All rights reserved. 34
Existing knowledge and material (2)
B2B fulfillment mapped with SCOR*processes
2017-09-29 restricted
D1.2: Receive,
Enter and
Validate Order
D1.5:
Build
Loads
D1.7: Select
Carriers, Rate
Shipments
D1.8: Receive
Product from
Source or Make
D1.9:
Pick
Product
D1.10:
Pack
Product
D1.12:
Ship
Product
D1.13: Re-
ceive Product
by Customer
S1.1:
Schedule
Deliveries
S1.2:
Receive
Product
P3.3:
Balance
Make R&R
M1.1: Schedule
Production
Activities
M1.3:
Produce
and Test
M1.5:
Stage
Product
M1.6: Release
Product to
Deliver
M1.2:
Issue
Material
D2.2: Receive,
Configure,
Enter and
Validate Order
D2.3: Reserve
Invent. & Det.
Delivery Date
D2.4:
Consolidate
Orders
D2.5:
Build
Loads
D2.7: Select
Carriers, Rate
Shipments
D2.8: Rec.
Product
from Make
D2.9:
Pick
Product
D2.10:
Pack
Product
D2.12:
Ship
Product
D2.13: Re-
ceive Product
by Customer
S2.1:
Schedule
Deliveries
S2.2:
Receive
Product
P3.3:
Balance
Make R&R
M1.1: Schedule
Production
Activities
M1.3:
Produce
and Test
M2.5:
Stage
Product
M2.6: Rel.
Product to
Deliver
M1.2:
Issue
Material
M2.3:
Produce
and Test
P3.4: Establish
and Communi-
cate Plans
P3.4: Establish
and Communi-
cate Plans
D1.6:
Route
Shipm.
D1.4:
Consolidate
Orders
D1.3: Reserve
Invent. & Det.
Delivery Date
D2.6:
Route
Shipm.
Supply Chain Operations Reference Model, www.supply-chain.org
Copyright © Infineon Technologies AG 2017. All rights reserved. 35
Existing Knowledge and Material (3):
EDI Data, Message and Document Standards
2017-09-29 restricted
EAN-128 and EANCOM by GS1: world-wide standard in retail and production of
FMCG* and CPG**
*: Fast Moving Consumer Goods ** Consumer Pacakged Goods
List of data elements and their IDs Standardized transport label using the data elements
Existing Knowledge and Material (3):
EDI Data, Message and Document Standards
*: EANCOM is a specific subset of UN/EDIFACT
EAN-128 and EANCOM*by GS1: world-wide standard in retail and production of
FMCG and CPG
List of EDI messages (using the data elements) Application scenarios (based on roles and transports)**
**: Source: GS1-Germany;
explanation document for
message type DESADV
36
Copyright © Infineon Technologies AG 2017. All rights reserved.2017-09-29 restricted
Existing Knowledge and Material (3):
EDI Data, Message and Document Standards
EDI Message DESADV
(„DESpatch ADVise“, Lieferavis)
EDI messages for freight
forwarders: IFTMIN, IFCSUM
Transport label on load units /
handling units GS1 transport
label
Electronic and paper-based
information exchange with
unique data elements
Immediate check of delivered
goods against advised goods
possible
Source: GS1 Germany, after Prof. H. Baumgärtel, University of Applied Sciences Ulm)
EAN-128 and EANCOM by GS1: world-wide standard in retail and production of
FMCG and CPG
Serial Shipping Container Code
(SSCC, NVE) of Load units
Load Units +
pick up date
Load Units +
delivery date
EDI, pre-delivery information Physical flow of goods
37
Copyright © Infineon Technologies AG 2017. All rights reserved.2017-09-29 restricted
Copyright © Infineon Technologies AG 2017. All rights reserved. 38
Existing Knowledge and Material (4):
Supply Network Organizations and Humans
2017-09-29 restricted
Source: M. Krupp: “Kooperatives Verhalten auf der
sozialen Ebene einer Supply Chain“, EUL-Verlag, 2005
Can be mapped with
organizations
ontology by the World
Wide Web Consortium
(W3C)
Source: W3C (http://www.w3.org/TR/vocab-org/ )
Copyright © Infineon Technologies AG 2017. All rights reserved. 39
Identified Gap between existing Knowledge
and further Project Needs
Most of the concepts for a B2B business transaction and communication
ontology exist:
W3C organization ontology
VDA + DSLV VDA 5002 logistics roles, documents, contracts
SCOR processes
B2B communication and transport label standards (e.g. EDIFACT, EANCOM, VDA, etc.)
…and they all fit together!
But what is needed? Only an addition for the base of business
transactions:
Delivery Master Contracts (DMC) (IFX: Volume Purchase Agreements)
Delivery Schedules (linkage to concepts from EDI and logistics contracts are possible)
Delivery: linkage to Delivery Schedule
Consignment: linkage to Delivery
2017-09-29 restricted
Copyright © Infineon Technologies AG 2017. All rights reserved. 40
Ontology of Business Transactions
(Entity Relationship Diagram)
2017-09-29 restricted
Copyright © Infineon Technologies AG 2017. All rights reserved. 41
Ontology on Business Transactions
Linkage to existing knowledge / ontologies
2017-09-29 restricted
Business
transactions
data elements
(cf. EAN-128)
Product
models and
classifications
Payment
conditions
(invoice vs.
credit voucher;
frequency, …)
Single
Orders
Information
exchange acts
(e.g. EDI messages)
Delivery documents
(papers)
Transport documents
(papers)
Transport conduction
(routes, tours,
tour stops, …)
Tracking and tracing
- of transport means
- of transport units
- of load units
fleet telematics; RFID
Means of transport
(train, truck, airplane, …)
Transport units
(trailer, semi-trailer,
swap bodies, container)
Product packaging
(logistics point of
view) (vs. Packaging
of semiconductors)
- Inner packaging
- Outer packaging
- Package unit
hierarchy, …
Labels
- of product packages
- of handling units
- of load units
(cf. GS1 transport label,
VDA4902 label,
ZVEI MAT-Label, …)
Packaging items
(receptacles, pallets,
boxes, load carrier, …) Transport companies,
roles, and contracts
Business
processes
(extern: SCOR,
intern)
Demand forecasts
Acknowledgments
Copyright © Infineon Technologies AG 2017. All rights reserved. 42
Test and validate ontology: Sample
Supply Chain for electric motors for E-Bikes
2017-09-29 restricted
Incrediblion
PLC
FlyingDutch
man Inc.
Company
Business
Unit
Electric
Drives Div.
Bicycle
Division
Site
Plant
Munich
plant
Rotterdam
plant
Product
Class
Product
Type
Electric
Bicycle
Motor
EBM_278369
Electric Bike
FD_MTB_
GenerationX
EBM_278369
isSupplierOf
isCustomerOf
isA
isA isSupplierOf
isCustomerOf
unitOf unitOf
hasUnit hasUnit
siteOf
hasSite siteOfhasSite
supplies
isSuppliedBy
isA
isA
produces producedAt produces producedAt
isA
isA
contains
hasPart partOf
sameAs
isA
isA
contains
Organizations (W3C)
Business transactions
Product modelling
Ontologies (proper) Test data (sample supply chain)
Copyright © Infineon Technologies AG 2017. All rights reserved. 43
Test and validate ontology: test graph sample
individual: Delivery Master Contract
2017-09-29 restricted
Reasoner proofs:
The ontology (i.e. all used
ontologies and their linkages)
as well as the test data is
consistent and new data
(triples) could be derived by
inference!
Copyright © Infineon Technologies AG 2017. All rights reserved. 44
Test and validate ontology: test graph sample
visualization: Delivery Master Contract
2017-09-29 restricted
Test data mapped to supply chain specific
deployment architecture
www.w3.org
Organizations
ontology
www.vda.de
Glossary
VDA5002
www.vda.de
EDI
messages
VDA49xx
www.vda.de
Transport
Label
VDA4902
www.me.com
Business
Transactions
Ontology
www.gs1.org
EANCOM EDI
messages
www.icc.org
Incoterms as
ontology
www.unece.org
UN/EDIFACT
Supplier Customer
S2.1
S2.2
S2.3
S2.4
D2.1
D2.2
D2.x
D2.11
Order
RDF
graph
Refers to
S2.5
ASN as
RDF
graph
Publishes
Retrieves
Retrieves
D2.13
D2.12
Ships
Transport
label
Refers to
Refers to
45
www.gs1.org
EAN-128 data
elements
Network
with public,
cross-industry,
and specific
ontologies
www.gs1.org
GS1 transport
label
Further Research and Next Steps on
semantic web based B2B communication
While the general deployment architecture is clear, there are still several detailed technical
questions:
Data communication / retrieval by subscribing partner:
SPARQL query vs. RDF graph transmission?
How to merge? How to avoid inconsistencies?
Could a block chain be a solution?
How to avoid technical overhead of point-to-point messaging systems?
Data security for the concept of choice:
HTTPS, Tunneling, Encryption, Public Key Infrastructures, … ?
Generation of world wide unique identifiers for:
Delivery Master Contracts
Identification of companies with D.U.N.S. numbers?
Combine customers, suppliers and article number?
Delivery Schedules
Load Units? SSCC from EAN-128 that requires a contract with GS1?
46
Copyright © Infineon Technologies AG 2017. All rights reserved. 47
Value proposition for
semantic web based B2B communication
1. New customers without EDI or supported EDI standards
Build SW based communication instead of proprietary web portals
Save additional manual work of filling boring web templates
Reduce transmission errors through elimination of media breaks
Saving potential depends on personal cost at IFAG and number of CLMs working with
such portals
Maybe an activity based costing analysis is necessary to gather the relevant quantitative
data
2. Convince existing customers without EDI but with supplier portals etc. from SW based
communication
Save current manual work of CLMs working with Web Portals
3. Conduct effort-gain comparison for new customers even with EDI capabilities between
establishing a new EDI connection vs. establishing a SW based communication
Cost approximations for establishing new EDI connections:
100.000 to 150.000€ (industry average in Europe).
Saving potential depends on real cost at IFAG for establishing a new EDI
connections
Based on an already supported / used EDI standard
Based on a new EDI standard
2017-09-29 restricted
Agenda
Semantic Web Technologies
Application Areas in Industrial Companies
Semantic Web Deployment Architectures
Course of Action for SW-based B2B communication
1
2
3
4
Summary & Next Steps5
48
2017-07-11 restricted Copyright © Infineon Technologies AG 2017. All rights reserved.
Copyright © Infineon Technologies AG 2017. All rights reserved. 49
Summary & Next Steps
Semantic web is reality today through Linked Open Data & Application
potential for industrial companies are obvious
Crucial base concepts for success exist like
RDF (simple, unique data scheme)
Ontologies and languages to express them (RDFS, OWL)
As well as general deployment methods like
SPARQL queries and updates
Reasoning algorithms, ontology mapping, data conversions
As many applications cannot run in Linked Open Data due to privacy
reasons specific deployment architectures will evolve with user data stored
and exchanged on specific architecture levels and it’s ontologies located in
higher levels
Semantic web based ontology development is beneficial per se for:
Understanding complex supply chains and the complex interactions
within the own company
Linking companies and sub projects of huge funded projects
As preparation for the applications to come
2017-09-29 restricted
Copyright © Infineon Technologies AG 2017. All rights reserved. 50
Summary & Next Steps (2)
Even if the focus of the ontology engineering example in this
presentation was b2b communication, next steps may also occur
for
Publishing information on infineon AG at Linked Open Data
Company information
Products and Services
Building up semantic based intranet services for
Knowledge management (glossaries, process documentations, …)
Semantic based learning
Data warehousing
Cross-department data models
2017-09-29 restricted
same as
same as
PLM
Target Goals for this Presentation
Learn about the application areas
for semantic web in Industrial
companies & its potential
Get an insight into methods and
Architectures for semantic web
(like RDF, OWL, HTTP and
SPARQL)
… and preliminary answers on
how to apply it systematically
Ontology Engineering
51
2017-10-25 restricted Copyright © Infineon Technologies AG 2017. All rights reserved.
Thank you!
Do you have questions?
... The following Table 4 summarizes some key application areas of semantics in industrial companies. parts in a car repair information service [85] 3. Procurement and Supply Chain Management Appli-cation  Unifying supplier catalo-gues  Knowledge management on potential suppliers and their offers [86] 4. Business Transaction Application  Information exchange for tender processes / RFQs  Order and Call-off processing [87] 5. BIG DATA ANALYTICS AND INDUSTRY 5.0 CONCEPTUAL FRAMEWORK To meet the future manufacturing complexity of increasing customization through an optimized manufacturing (robotized) process, Industry 5.0 recognizes that man and machine must be interconnected. Industry 5.0 is likely to affect the economy, ecology, and the social world. ...
Article
The digitalization of modern manufacturing systems has resulted to increasing data generation, also known as Big Data. Although there are several technologies and techniques under the term Data Analytics for gathering such data, their interpretation to information, and ultimately to knowledge remains in its infancy. Consequently, albeit engineers, currently can monitor the factory level, optimization is cut off of the data acquisition, and is based on data related methodologies. The focus should be pivoted on designing and developing suitable frameworks for integrating Big Data to process optimization based on the context of information gathered from the shopfloor. This paper aims is to investigate the opportunities and the gaps as well as the challenges arising in the current industrial landscape, towards the efficient utilization of Big Data, for process optimization based on the integration of semantics. To that end, a literature review is performed, and a data-based framework is presented.
ResearchGate has not been able to resolve any references for this publication.