Metrology in Industry 4.0 Metromeet 2016 25.2.2016 Toni Ventura-Traveset DATAPIXEL Innovalia Metrology Pag. 1
Industry 4.0
Cyberphysical systems Interoperable Virtualization Decentralized Self-configuration Self-optimization Self-diagnosis. Intelligent Manufacturing Real Time Physical Digital
Industry 4.0 Metrology 1.0 Metrology 2.0 Metrology 3.0 Metrology 4.0?
Not so new.
Holon Arthur Koestler (1905-1983) Although it is easy to identify sub-wholes or parts, wholes and parts in an absolute sense do not exist anywhere Agent Autonomous Self-regulating Cooperating Information flow Holo (whole) + on (parts) 1967
The holonic factory (Christensen 1994) Brain Body
1966- Theory of self-reproducing automata (edited by Arthur Burks) Von Neumman (1903-1957)
Von Neumman Theory of self-reproducing automata Agent (cognitive automata) Autonomous: decide for themselves Active: active control Flexible: Reactive, pro-active, social Intelligence: goal oriented, learning, adaptative, cognitive
Multi-agent System (MAS) Self-organisation Decentralization Local views Multiple cognitive entities acting in communities
Mapping physical to virtual Multiple cognitive entities acting in communities
Industry 4.0
Revolution!!!!
Revolution a sudden, extreme, or complete change in the way people live, work, etc.
Industry 4.0
Industry 4.0 WHY?
Energy
Energy Productivity
Less people Robotics China: 25% 2015-2018: 1.3 M robots
FOXBot 2013- iphone manufacturer, Foxconn, which employs 1.2 million workers, has announced that will deploy 1 million robots at their production lines in five years (70% automation). 2015- At Foxconn, the robots will only replace 30% of workers, all of them will be promoted to higher tasks Foxconn produces 10.000 robots/year
Energy Productivity Quality
owner-reported problems in the first 90 days of new-vehicle ownership the industry experiences a 3% year-over-year improvement in initial quality
Energy Productivity Quality?
Energy Productivity Quality Efficiency
Input Output Efficiency Energy Products Labour Services Material Customer satisfaction
Energy
New (old) materials
Energy Productivity Quality Efficiency Sustainability
400 ppm!!
Source : European Commission Competitiveness report 2013
Source : European Commission Competitiveness report 2013
Energy Productivity Quality Efficiency Sustainability?
General Systems Theory
Systems Theory Ludwig von Bertalanffy (1901, 1972) Organismic Theory Kurt Goldstein (1878, 1965)
Systems Theory Holism: system, macro, micro Open systems and isomorphism Meta-theory Law of exponential growth Xt = X0(1+r) t
Law of exponential growth and demography
Law of exponential growth and GDP
Law of exponential growth Moore s law Energy consumption Productivity Internet of things
Law of exponential growth limits
Law of exponential growth of knowledge Arthur Conan Doyle (1859-1930) The Great Kleinplatz Experiment Knowledge begets knowledge as money bears interest
Law of exponential growth of knowledge Knowledge is doubling every 8 years
Law of exponential growth and publications No evidence of exponential increase of traditional scientific publications (peer-reviewed journals) Publications using other channels is growing fast (internet, conferences ) (but not measured ) Traditional scientific dissemination ( and academia knowledge generation) is not able to address new knowledge creation needs
Next Generation Sequencing (base pairs)
Typical video game makes 60 to 70 percent of its money in the first four or five days LED TVs have 18 to 24 month to maximize revenues before being replaced by next generation
Law of exponential growth of manufacturing knowledge Traditional manufacturing model is not able to address new knowledge generation needs Cyberphysical systems Machines that can store and generate knowledge Continuous process transformation
Organismic theory Humans instinctively dissect the situation in an attempt to understand it better. Yet, in doing so, humans miss the essence or intrinsic nature of the organism itself. Interpretations should be taken in as a whole without giving special preference towards one part of the phenomena. When describing the phenomena, attention should not be diverted to one aspect of the phenomena that may be of interest. The holistic approach calls instead for the phenomena to be described from all angles without bias towards one part.
Manufacturing as organismic entities From narrow manufacturing knowledge mgmt to wide holistic knowledge mgmt Open innovation The customer. The Factory.. The individual process and product
Growing Mastering the environment Maintaining coherence
Richard Ryan / Edward Deci Self-determination Organismic meta-theory
(Meta-theory: growing, mastering environment, coherence) Learning Autonomy Richard Ryan / Edward Deci Self-determination Organismic meta-theory Relational
Relational Colaborating to learn Learning to colaborate Learning Techno-social network Self-monitoring Self-configuration Self-adaptation Autonomy
Relational Colaborating to learn Learning to colaborate Learning Techno-social network Self-monitoring Self-configuration Self-adaptation Autonomy
Learning systems
Jean Piaget (1896-1980) Theory of cognitive development (Genetic epistemology) Sensorimotor stage Preoperational stage (symbolic, intuitive thought) Operational stage (logical thinking, classes) Formal operational stage (metacognition)
Jean Piaget (1896-1980) Theory of cognitive development Sensors and maps (schemata) Symbols (ontologies) Patterns and simple algorithms Analytics and complex algorithms
Is not only about productivity and quality Is not only about efficiency Is not only about sustainability Is about holistic evolution of intelligence of complex systems Is about human growth, creativity, energy, passion. Art&Science
https://youtu.be/3xgobli_fdg
Dicebamus hesterna die.
Agents in Manufacturing?
Agents in Manufacturing?
Agents in Manufacturing?
Agents in Manufacturing?
Human Machine Product
Human Machine Product Society Industry Organisation Manufacturing Unit Individual
Human Machine Product Society Industry Intelligence Organisation Manufacturing Unit Individual Relational Learning Autonomous
Dicebamus hesterna die.
Intelligence Robert Sternberg, Department of Human Development at Cornell University
Intelligence the ability to adapt to the environment and to learn from experience (Sternberg & Detterman, 1986).
Intelligence 1) the ability to achieve business goals, given a technological, sociocultural, economical context 2) by capitalizing on strengths and correcting or compensating for weaknesses; 3) in order to adapt, to shape, and select environments; and, 4) through a combination of practical, analytical and creative abilities (Adapted from Sternberg, 1997, 1998, 1999)
Intelligence 1) the ability to achieve business goals, given a technological, sociocultural, economical Goal context oriented 2) by capitalizing on strengths and correcting or compensating for weaknesses; Learning 3) in order to adapt, to shape, and select environments; and, Adaptative 4) through a combination of practical, analytical and creative abilities Cognitive abilities (Adapted from Sternberg, 1997, 1998, 1999)
Intelligent agent System 2: System 1: agent System 3,4: System 5
Intelligent agent System 2: simulation System 1: sensors agent System 3,4: cognition System 5 Techno-Social
SYSTEM 1 Sensors Body map Body schema Body image
SYSTEM 1: Smart Sensors
SYSTEM 1 Sensors Map Schema Edge Distributed sensing-measurement Semantic representation Perimetral / extended data Filtered information
Human Machine Product Society Industry Organisation Manufacturing Unit Individual Big Data Edge Computing Externalist Schema Internalist map Sensimotor: physical actuators & sensors
SYSTEM 2: Simulation
Stimulus Action
Simulation Stimulus Simulation of action Action Stimulus Simulation of action Blocked Action Visualization Simulation of action Blocked Action
SYSTEM 2: Simulation Simulation of System 1: sensing + map Simulation of the environment Simulation of actions Digital Factory
SYSTEM 2: Simulation Embedded Simulation in Machines Simulation systems embedded in the real machines Real mode Simulation mode
Repetition Diversity Wide / Open mind
Stable production
Learning by doing Learning by visualizing
Learning by simulating Creating diverse simulated conditions to estimulate factory learning
Human Machine Product Society Industry Organisation Manufacturing Unit Individual Big Data Edge Computing Externalist Schema Internalist map Sensimotor: physical actuators & sensors
SYSTEM 3 SYSTEM 4 Cognition
100 + 1
100 + 1 2 x 4
100 + 1 2 x 4 8 x 1
+ 1.10 How much does the ball cost?
17 x 14
17 x 14 5211 x 23
100 + 1 17 x 14 Fast system Slow system
100 + 1 17 x 14 Fast system Fast, automatic, frequent, emotional, stereotypic, subconscious Slow system Slow, effortful, infrequent, logical, calculating, conscious 99% 1% Daniel Kahneman: Nobel prize 2002
100 + 1 17 x 14 Fast system Based on patterns Intuitive intelligence Learning Slow system Based on analysis Analytical intelligence 99% 1%
Situation Pattern? Algorithm repository Pattern based intelligence Analytical Intelligence New Algorithm
SYSTEM 3: Fast thinking: Pattern based Intelligence SYSTEM 4: Slow thinking: Analytical Intelligence
SYSTEM 3: Fast thinking: Pattern based Intelligence Pattern based algorithms Automation PLC systems Logic systems / Fuzzy logic Smart Factory Neural networks Control systems
SYSTEM 4: Slow thinking: Analytical intelligence Data Mining systems Statistical analysis systems Agent-based planning systems Deep-learning Learning Factory
How to grow the algorithm repository?
Diversity of Situations Algorithm? Algorithm repository System 3: Pattern based intelligence System 4: Analytical Intelligence New Algorithm
7 month Novelty seeking The relationship of novelty preferences during infancy to later intelligence and later recognition memory, Joseph F. Fagan (1984)
7 month 5 years Novelty seeking ++ Intelligence The relationship of novelty preferences during infancy to later intelligence and later recognition memory, Joseph F. Fagan (1984)
Laboratories for creative exploratory learning Creating diverse conditions to estimulate factory learning
Learning sources Reality Simulation Exploration Labs Cognitive system Pattern based intelligence Analytics
Human Machine Product Society Industry Organisation Manufacturing Unit Individual Big Data Analytic intelligence Pattern based intelligence Edge Computing Externalist Schema Internalist map Sensimotor: physical actuators & sensors
Exploratory mode of intelligence Creative intelligence System 1 Sensing + mapping System 3 Intuition System 2 Simulation System 4 Analytical
Human Machine Product Society Industry Organisation Manufacturing Unit Individual Big Data Analytic intelligence Pattern based intelligence Edge Computing Externalist Schema Internalist map Sensimotor: physical actuators & sensors
System 5 Social cognition: knowing about relationships To know and To be known Inter-agent ties Cooperative agents Techno-social intelligence
To know and to be known: internet of things To know To be known
Human Machine Product Society Industry Organisation Manufacturing Unit Individual Techno-social intelligence Big Data Analytic intelligence Pattern based intelligence Edge Computing Externalist Schema IoT Internalist map Sensimotor: physical actuators & sensors
Standards QIF Quality information framework
Human Machine Product Society Industry Organisation Manufacturing Unit Individual Techno-social intelligence Big Data Analytic intelligence Pattern based intelligence Edge Computing Externalist Schema IoT Internalist map Sensimotor: physical actuators & sensors
Metrology 4.0 THE CHALLENGES
Human Machine Product Society Industry Organisation Manufacturing Unit Individual Techno-social intelligence Big Data Analytic intelligence Pattern based intelligence Edge Computing Externalist Schema IoT Internalist map Sensimotor: physical actuators & sensors 1
1st challenge for metrology 4.0 To allow future growth of manufacturing, metrology has to exponentially grow in terms of information generation and knowledge sharing
High density metrology Physical Digital
Metrology knowledge sharing
Human Machine Product Society Industry Organisation 2 Manufacturing Unit Individual Techno-social intelligence Big Data Analytic intelligence Pattern based intelligence Edge Computing Externalist Schema IoT Internalist map Sensimotor: physical actuators & sensors 1
2nd challenge for metrology Metrology has to expand their narrow view inside the laboratory towards a wide holistic metrology
Metrology everywhere
Human Machine Product Society Industry Organisation 2 Manufacturing Unit Individual Techno-social intelligence Big Data Analytic intelligence Pattern based intelligence Edge Computing Externalist Schema IoT Internalist map Sensimotor: physical actuators & sensors 1 3
3rd challenge for metrology Metrology has to develop more powerful simulation tools to facilitate integration in the digital factory
Metrology simulation Simulation of sensors Simulation of interaction with the parts Simulation of sources of error (noise) Simulation of influence of humans..
Human Machine Product Society Industry Organisation 2 Manufacturing Unit Individual Techno-social intelligence Big Data 4 Analytic intelligence Pattern based intelligence Edge Computing Externalist Schema IoT Internalist map Sensimotor: physical actuators & sensors 1 3
4th challenge for metrology Metrology has to provide analytical tools for algorithm generation and deep-learning
The algorithm marketplace
Human Machine Product Society Industry Organisation 2 Manufacturing Unit Individual Techno-social intelligence Big Data Analytic intelligence 4 5 Pattern based intelligence Edge Computing Externalist Schema IoT Internalist map Sensimotor: physical actuators & sensors 1 3
5th challenge for metrology Metrology has to provide tools for creative exploratory learning labs (disruptive metrology)
Disruptive metrology
Human Machine Product Society Industry Organisation 2 Manufacturing Unit Individual Techno-social intelligence Big Data Analytic intelligence 4 5 Pattern based intelligence Edge Computing Externalist Schema IoT 6 Internalist map Sensimotor: physical actuators & sensors 1 3
6th challenge for metrology Metrology has to develop standards and practical examples regarding IoT standards
Metrology and standards
Human Machine Product Society Industry Organisation 2 Manufacturing Unit Individual 7 Big Data Techno-social intelligence Analytic intelligence 4 5 Pattern based intelligence Edge Computing Externalist Schema IoT 6 Internalist map Sensimotor: physical actuators & sensors 1 3
7th challenge for metrology Metrology has to provide/adapt solutions for big data storage and analytics
Data replication Auto-associative patterns Hierarchical levels Invariant patterns Federated DataBases Ontological networks BIG DATA need to be organised and contextualized for fast retrieval and analysis
Human Machine Product Society Industry Organisation 2 Manufacturing Unit Individual 7 Big Data Techno-social intelligence Analytic intelligence Pattern based intelligence 8 4 5 Edge Computing Externalist Schema IoT 6 Internalist map Sensimotor: physical actuators & sensors 1 3
8th challenge for metrology Metrology has to develop new methods and tools for advanced human-machine interaction in order to deal with complexity
Metrology for humans
Human Machine Product Society Industry Organisation 2 Manufacturing Unit Individual 7 Big Data Techno-social intelligence Analytic intelligence Pattern based intelligence 8 4 5 Edge Computing Externalist Schema 9 IoT 6 Internalist map Sensimotor: physical actuators & sensors 1 3
9th challenge for metrology Metrology has to facilite more multi-disciplinary knowledge generation, sharing and colaboration (multi-disciplinary innovation)
Multi-disciplinary metrology
Human Machine Product Society Industry Organisation 2 Manufacturing Unit Individual 7 Big Data Techno-social intelligence Analytic intelligence Pattern based intelligence 8 4 5 Edge Computing Externalist Schema 9 10 IoT 6 Internalist map Sensimotor: physical actuators & sensors 1 3
10th challenge for metrology Metrology has to facilite more multi-disciplinary education, gain visibility and increase attractiveness to young talents
Metrology as creative space of value creation and business opportunities
Metrology challenges for Industry 4.0 1- Metrology growth, high density metrology 2- Metrology everywhere 3- Simulation 4- Analytical tools 5- Disruptive metrology, innovation labs 6- Standards 7- Big data storage 8- Human machine interaction 9- Multi-disciplinary innovation 10- Multi-disciplinary education
Industry 4.0 is driven by the global evolution of intelligence and complex systems Exponential growth of knowledge generation Reference model of intelligent manufacturing The challenges of metrology 4.0
Thank You!!