Graph Databases, RDF Triplets, Ontologies, SWRL and SPARQL Queries are modern technologies for knowledge-based artificial intelligence. In this presentation, wou will see how Industry 4.0 standards such as AutomationML (AML), OPC-UA and eCl@ss create the conditions for unparalleled engineering power in manufacturing in general, and in assembly lines in particular. AI is not only data-hungry deep learning. Use what we already know before you try to guess what's new. Let's make the fourth industrial revolution elegant and energy-efficient! Weiss GmbH provides unique automation components and their Digital Twin partial models by the I4.0 standards.
Gurgaon ✡️9711147426✨Call In girls Gurgaon Sector 51 escort service
Towards batch one size with industrial semantics email
1. BATCH ONE SIZE WITH INDUSTRIAL
SEMANTICS
The advantages of analytical solutions in manufacturing automation
and how to create the environment to make them work for you
4. THE WORLD OF WEISS COMPONENTS
UNPARALLELED DIVERSITY
5. WE KNOW WHAT YOU ARE TALKING ABOUT
25.11.2018 Our customers PAGE 5
Automotive Medical /
pharmaceutical
Food Machinery and plant
engineering
Consumer / electronics
6. A SELECTION OF OUR CUSTOMERS
25.11.2018 Our customers PAGE 6
7. 1 Types and qualities of knowledge
2 Knowledge engineering using AutomationML
3 Semantic interoperability – RAMI4.0
4 Expert systems and the software stack of truth
5 Autonomous system- and process engineering for unique products
25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 7
8. TYPES AND QUALITIES OF KNOWLEDGE
25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 8
» Immanuel Kant
* Critique of the pure Reason
A Priori
•Knowledge is necessary
and universal
•Rules of Logic
•Axioms of Mathematics
A Posteriori
•Based on Experience
•Knowledge is Intuition
•All Sciences, including
•“Laws” of physics
9. TYPES AND QUALITIES OF KNOWLEDGE
25.11.2018 Towards Batch One Size with Industrial Semantics
» Albert Einstein
“How can it be that mathematics, being after all a product of
human thought which is independent of experience, is so
admirably appropriate to the objects of reality?”
PAGE 9
10. TYPES AND QUALITIES OF KNOWLEDGE
25.11.2018 Towards Batch One Size with Industrial Semantics
Deep?
PAGE 10
Deep!
11. TYPES AND QUALITIES OF KNOWLEDGE
25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 11
Deep?
Praxis / Machine Learning
» Can only Interpolate
» Domain Specific
» Subject to Bias / Prejudice
» Computationally expensive
Deep!
Theory / Analytical Models
» Can Extrapolate
» Broadly applicable
» Unbiased
» Computationally cheap
12. TYPES AND QUALITIES OF KNOWLEDGE
A Priori
Knowledge
Exact
Sciences
Analytical
Models
25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 12
Leverage Engineering
Knowledge
Let the computer
reason upon it
Use data-driven as a
last resource
13. 1 Types and qualities of knowledge
2 Knowledge engineering using AutomationML
3 Semantic interoperability – RAMI4.0
4 Expert systems and the software stack of truth
5 Autonomous system- and process engineering for unique products
25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 13
14. 25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 14
KNOWLEDGE ENGINEERING USING
AUTOMATIONML
Source:
AutomationML in a Nutshell
Nicole Schmidt, Arndt Lüder
State: November 2015
» Open Association since
2009
» XML-Based data format
» Industry 4.0 Standard
IEC 62714
IEC 62424
15. 25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 15
KNOWLEDGE ENGINEERING USING
AUTOMATIONML
Source:
AutomationML in a Nutshell
Nicole Schmidt, Arndt Lüder
State: November 2015
16. 25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 16
KNOWLEDGE ENGINEERING USING
AUTOMATIONML
Source:
AutomationML in a Nutshell
Nicole Schmidt, Arndt Lüder
State: November 2015
AutomationML
Taxonomy
Hierarchy
RolesInterfaces
External
References
17. 25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 18
KNOWLEDGE ENGINEERING USING
AUTOMATIONML
Geometry:
Collada
Logic:
PLCOpen
XML
Semantic
Reference:
eCl@ss
18. 1 Types and qualities of knowledge
2 Knowledge engineering using AutomationML
3 Semantic interoperability – RAMI4.0
4 Expert systems and the software stack of truth
5 Autonomous system- and process engineering for unique products
25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 20
19. SEMANTIC INTEROPERABILITY – RAMI4.0
HTTP://I40.SEMANTIC-INTEROPERABILITY.ORG/
25.11.2018
Towards Batch One Size with Industrial Semantics PAGE 22
20. SEMANTIC INTEROPERABILITY – RAMI4.0
HTTP://I40.SEMANTIC-INTEROPERABILITY.ORG/
25.11.2018
Towards Batch One Size with Industrial Semantics PAGE 23
Semantic References
•Objects and Properties referred globally
•Unified Classification
•Standardized Parameter Sets
•Language-Neutral
Queries:
•What is the average Power Factor of ALL my
electric motors weighted by nominal power?
•Which are ALL my ball screws past warranty?
•Which machine tool is most likely to cause a
downtime?
21. 1 Types and qualities of knowledge
2 Knowledge engineering using AutomationML
3 Semantic interoperability – RAMI4.0
4 Expert systems and the software stack of truth
5 Autonomous system- and process engineering for unique products
25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 24
22. EXPERT SYSTEMS / SOFTWARE STACK OF TRUTH
AI
Symbolic
Logic Based
Knowledge
Based
Subsymbolic
Autonomous
Systems
Distributed AI
Statistical
Probabilistic
ML
25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 25
Facts
Rules
Queries
Prolog, 1972
Declarative Rules Language
Modus Ponens, backward chaining
Engine
Datalog…RuleML…Drools…
23. EXPERT SYSTEMS / SOFTWARE STACK OF TRUTH
SEMANTIC WEB
» Proposed by Sir Tim Berners-Lee
- Inventor of the WWW
25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 26
“I have a dream for the Web [in which
computers] become capable of analyzing
all the data on the Web – the content,
links, and transactions between people
and computers. A "Semantic Web",
which makes this possible, has yet to
emerge, but when it does, the day-to-day
mechanisms of trade, bureaucracy and
our daily lives will be handled by
machines talking to machines. The
"intelligent agents" people have touted
for ages will finally materialize.[6]”
24. EXPERT SYSTEMS / SOFTWARE STACK OF TRUTH
25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 28
»The Semantic WEB failed.
»But we will save it. And
I4.0 will be born.
Engineering Ontologies
Ecossystem Data Exchange
Querying complex problems
25. EXPERT SYSTEMS / SOFTWARE STACK OF TRUTH
25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 29
Source: AutomationML Analyzer form TU-Wien
http://data.ifs.tuwien.ac.at/aml/analyzer/
What is the weight and power
consumption of the whole system?
SELECT (SUM(xsd:integer(?deviceWeight)) AS ?systemWeight)
(SUM(xsd:integer(?devicePowerConsumption)) AS ?
systemPowerConsumption)
WHERE {
aml:myConveyor aml:hasPart* ?device
?device a aml:InternalElement
?device aml:hasAttribute ?attribute
?attribute aml:hasAttributeName "Weight„
?attribute aml:hasValue ?deviceWeight
?device aml:hasAttribute ?attribute
?attribute aml:hasName "PowerConsumption„
?attribute aml:hasValue ?devicePowerConsumption . }
26. 1 Types and qualities of knowledge
2 Knowledge engineering using AutomationML
3 Semantic interoperability – RAMI4.0
4 Expert systems and the software stack of truth
5 Autonomous system- and process engineering for unique products
25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 30
27. AUTONOMOUS SYSTEM- AND PROCESS
ENGINEERING FOR UNIQUE PRODUCTS
25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 31
» How can we produce <this> with the current
resources ?
» What Throughput can be reached if we add
<this> product/process/resource change?
» Which alternatives are there in the market for
<this> linear axis, exceeding its performance
but not its weight?
» Please tune the control loop of <this> 450mm
wafer manipulator eliminating the first 6
eigenfrequencies!
28. AUTONOMOUS SYSTEM- AND PROCESS
ENGINEERING FOR UNIQUE PRODUCTS
Vendors
Market
Standards
Tech Partners
Manufacturers
Society Batch 1 Size
Self-
Configuring
Production.
Inference
Engines
Expert
Rulebases
Companion
Ontologies
Industrial
Semantics
Product
Models
Analytical
Formulas
0
Experiments
25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 32
https://www.symestic.de/de/industrie-4-0.html
29. AUTONOMOUS SYSTEM- AND PROCESS
ENGINEERING FOR UNIQUE PRODUCTS
25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 33
Neo4J, ArangoDB, Jena+Hbase, CumulusRDF+ScyllaDB
Solver1
SPARQL Queries
RDF MathML
Solver2 Eval…
Reasoning Engine
Problem Solution
Vendor AML +
eCl@ss
Your AML +
eCl@ss
AMLO
Ingest
Ingest
MathML
Protégé
Your Rules
30. BONUS TAKEAWAY
Behold!
• Hire some Mathematicians to
code 400 years of science in
your domain
• Demand AML-Product Models
from ALL your vendors
• Master AML, RDF, OWL and
SPARQL yourself
• Build a powerful, scalable
inference stack
Please!
•Use Simulations, Numeric
Methods and Data-Driven
Learning only as your last
resource
•Mankind has to
spare those
Megawatts to mine
cryptocurrencies
25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 34
31. BIBLIOGRAPHY
» CUDRÉ-MAUROUX, Philippe, et al. NoSQL databases for RDF: an empirical evaluation. In:
International Semantic Web Conference. Springer, Berlin, Heidelberg, 2013. S. 310-325.
» KOVALENKO, Olga, et al. AutomationML Ontology: Modeling Cyber-Physical Systems for
Industry 4.0.
» WAGNER, Constantin, et al. The role of the Industry 4.0 asset administration shell and the
digital twin during the life cycle of a plant. In: Emerging Technologies and Factory Automation
(ETFA), 2017 22nd IEEE International Conference on. IEEE, 2017. S. 1-8.
» FRANCALANZA, Emmanuel; BORG, Jonathan; CONSTANTINESCU, Carmen. A knowledge-based
tool for designing cyber physical production systems. Computers in Industry, 2017, 84. Jg., S.
39-58.
» AutomationML in a Nutshell. Nicole Schmidt, Arndt Lüder State: November 2015
» http://i40.semantic-interoperability.org/
25.11.2018 Towards Batch One Size with Industrial Semantics PAGE 35
WEISS has enjoyed dynamic growth in the last few years. We have faced the global challenges head on and established an international network of companies that generated sales revenue of €100 million with around 450 employees in 2017.
We have employees in Europe, America and Asia.
Our teams are interdisciplinary and include engineers, technicians, designers, programmers and consultants - all of whom have many years of experience. These different knowledge backgrounds form the basis for our intelligent mechatronic modules.
A very wide range of rotary indexing tables is available with different drive types and sizes. No other company has such a broad product portfolio. The large range of handling units supplements this: Pick&Place, axes, rotating units - all available in a wide range of versions. Control systems, software. Also superstructures and substructures, frames, plates…
Every sector is different and has its own special characteristics. We have therefore structured our organisation by industry segment.
That's why we know your industry and your processes. This kind of customer proximity is extremely important to us.
It allows us to better understand requirements and link them more effectively with our solution potential.
Our consultants bring their combined practical experience from numerous projects to every meeting. They know what is feasible and what is not.