A Semantic Architecture for Industry 4.0

Transcription

1 A Semantic Architecture for Industry 4.0 Cecilia Zanni Merk cecilia.zanni unistra.fr Deputy Head of the SDC team of ICube 4P Factory e lab March 16 th, 2016

2 Agenda Introduction Basic principles of Semantic Technologies A semantic architecture for Industry 4.0 Current projects A word to finish Cecilia Zanni Merk A semantic architecture for Industry 4.0 2

3 Introduction

4 Industry 4.0 It is a collective term embracing a number of contemporary automation, data exchange and manufacturing technologies. Technologies and concepts of value chain organization which draws together Cyber Physical Systems, the Internet of Things and the Internet of Services. AKA «the fourth industrial revolution» Cecilia Zanni Merk A semantic architecture for Industry 4.0 4

5 Cyber Physical Production Systems CPPSs are online networks of social machines organized as social networks. They link IT with mechanical and electronic components that then communicate with each other via a network. They therefore create a smart network of machines, ICT systems, smart products and individuals across the entire value chain and the full product life cycle. Sensors and control elements enable machines to be linked to plants, fleets, networks and human beings. These smart networks support what are called Smart Factories Cecilia Zanni Merk A semantic architecture for Industry 4.0 5

6 Trying to put it all together Industry 4.0 can be seen as an engineering ecosystem, where processes govern themselves, smart products take corrective actions to avoid damages, individual parts are automatically replenished, and raw parts and production machines enter into a dialogue in order to optimise manufacturing processes. Cecilia Zanni Merk A semantic architecture for Industry 4.0 6

7 The Role of Big Data and Analytics Big data is high volume, high velocity, and/or high variety information assets that require new forms of processing to enable enhanced decision making, insight discovery and process optimization (Gartner, 2012) Explain inconsistent component performance and availability Implement predictive manufacturing Cecilia Zanni Merk A semantic architecture for Industry 4.0 7

8 The Role of Semantic Technologies But what are semantic technologies? Semantics is the study of meaning Central to the study of communication (a crucial factor in social organization) Central to the study of the human mind thought processes, cognition, conceptualization Cecilia Zanni Merk A semantic architecture for Industry 4.0 8

9 Some things we know Consider The small blue circle is in front of the square. The square is behind the small blue circle. We are capable of verifying that both sentences are true in this particular situation. This is because we know what the world must be like in order for these sentences to be true. Taken from Assimakopoulos s course on Semantics (University of Malta) Cecilia Zanni Merk A semantic architecture for Industry 4.0 9

10 Some things we know Now consider: She drove past the bank This sentence then can mean more than one thing (it is ambiguous). This seems to be related to our knowledge of what bank denotes. Taken from Assimakopoulos s course on Semantics (University of Malta) Cecilia Zanni Merk A semantic architecture for Industry

11 Some things we know Finally, consider: 1. John murdered the president. 2. The president is dead. We also know that sentence two follows from sentence 1 (technically: sentence 1 entails sentence 2 or sentence 1 is a consequence of sentence 2) In this particular case, it seems to be related to the meaning of murder. Taken from Assimakopoulos s course on Semantics (University of Malta) Cecilia Zanni Merk A semantic architecture for Industry

12 Agenda Introduction Basic principles of Semantic Technologies A semantic architecture for Industry 4.0 Current projects A word to finish Cecilia Zanni Merk A semantic architecture for Industry

13 Basic principles of Semantic Technologies

14 Knowledge and Semantic Engineering Knowledge and Semantic Engineering proposes concepts, methods and techniques to model, formalize or acquire knowledge with the aim of capitalization, implementation or management in the broadest sense of the terms. An approach to problem solving based on KSE generally comprises three stages: 1. Knowledge acquisition Development of 2. Knowledge modelling Intelligent 3. Knowledge implementation Semantic Systems Cecilia Zanni Merk A semantic architecture for Industry

15 Intelligent Semantic Systems Software capable of reproducing the cognitive mechanisms of an expert in a particular field. Capable of answering questions, by performing reasoning from known facts and rules. Particularly suitable for decision support. Cecilia Zanni Merk A semantic architecture for Industry

16 Intelligent Semantic Systems The classical architecture of a ISS Cecilia Zanni Merk A semantic architecture for Industry

17 An Example Carrots are plants Rabbits only eat carrots If somebody eats only plants, he is a vegetarian carrot(x) => plant(x) rabbit(x) => [eats(x, y) => carrot(y) ] vegetarian(x) => [eats(x, y) => plant(y) ] Rabbits are vegetarian Bugs Bunny is vegetarian Cecilia Zanni Merk A semantic architecture for Industry

18 An Example Roger Rabbit eats hay Is he a rabbit? Is he vegetarian? We can t say Our knowledge base is incomplete!! Cecilia Zanni Merk A semantic architecture for Industry

19 The capitalization of experience An expert can tell us Roger Rabbit is a vegetarian rabbit OR If a rabbit eats hay, then it necessary eats carrots New knowledge without explanation Evolution of the rule base OR Both Use of a new knowledge structure to capitalize both, the new knowledge and the new rules SOEKS Cecilia Zanni Merk A semantic architecture for Industry

20 The capitalization of experience Set of Experience Knowledge Structure (SOEKS) Experience based knowledge representation To store uncertain and incomplete data To make qualitative and quantitative extractions of knowledge from available information name: Roger Rabbit Four components eatenfood: hay typeofanimal: rabbit variables, typeofeater: vegetarian functions, constraints, #eatenfood<=2 rules rabbit(x) eats(x,y) hay(y) => eats (x,z) carrot(z) If a rabbit eats hay, then it necessarily eats carrots Cecilia Zanni Merk A semantic architecture for Industry

21 An example Extra terrestrial rabbits eat meat Clearly, our initial knowledge base cannot be used to classify the Raving Rabbids Cecilia Zanni Merk A semantic architecture for Industry

22 We need more Both rules regarding the feeding of rabbits can be included, but they make the KB inconsistent. We need mechanisms to lead us to launch de appropriate rules according to the context We need meta knowledge Cecilia Zanni Merk A semantic architecture for Industry

23 Meta Knowledge Meta knowledge is knowledge about the domain knowledge, the rules or the experience Context Any information that characterizes a situation related to the interaction between human beings, applications and the surrounding environment (Dey et al, 2001) Four types: identity, location, status, time Culture, Protocols Meta knowledge is closely related to experience knowledge Cecilia Zanni Merk A semantic architecture for Industry

24 The KREM model Four components, with the goals of managing the complexity of developing ISSs, trying to solve the problems associated with knowledge elicitation, where knowledge can be tacit or not incorporating the capitalization of experience to improve decision making Cecilia Zanni Merk A semantic architecture for Industry

25 The KREM model Knowledge to operate, Domain ontologies to be developed. Rules to allow different types of reasoning, Monotone, spatial, temporal, fuzzy, or other depending on the application. Experience, Allowing the capitalization and reuse of previous knowledge. Meta Knowledge, Knowledge about the other three components Gives the context of the problem to be solved. Cecilia Zanni Merk A semantic architecture for Industry

26 The KREM model The way the domain knowledge is formalized will shape the way the rules are expressed. Experience will come complete the available knowledge and rules. Finally, meta knowledge will directly interact with the rules and the experience to indicate which rules (coming from experience or from the initial rule set) need to be launched according to the context of the problem. Cecilia Zanni Merk A semantic architecture for Industry

27 Why should we use semantic technologies in an industrial context? To share common understanding of the structure of information among people or software agents To enable reuse of domain knowledge To make domain assumptions explicit To separate domain knowledge from the operational knowledge To capitalize experience knowledge To be able to make decisions in different contexts Cecilia Zanni Merk A semantic architecture for Industry

28 Agenda Introduction Basic principles of Semantic Technologies A semantic architecture for Industry 4.0 Current projects A word to finish Cecilia Zanni Merk A semantic architecture for Industry

29 One possible architecture Taken from Toro s «A perspective on Knowledge Based and Intelligent systems implementation in Industrie 4.0» Cecilia Zanni Merk A semantic architecture for Industry

30 One possible architecture A physical layer where collections of signals from sensors and actuators are consumed and produced, such information travels through the data bus to a standardization layer that is able to perform data exchange and alignment in required time frames. The communication layer divides and controls packages of information and should assure security and anonymity where necessary. The CPS layer can support localized decisions on its own but also based on contextual knowledge. Contextual knowledge allows considering the impact of a given element in the machine (local information or data gathered) on the whole production line and vice versa The application layer includes the front end applications that will exploit the semantic enriched information with different human to machine interaction modules. Cecilia Zanni Merk A semantic architecture for Industry

31 Contributions at different levels Taken from Toro s «A perspective on Knowledge Based and Intelligent systems implementation in Industrie 4.0» «BigData» comingfrom sensors This information needs to be converted into knowledge for further use and exploitation Cecilia Zanni Merk A semantic architecture for Industry

32 Agenda Introduction Basic principles of Semantic Technologies A semantic architecture for Industry 4.0 Current projects A word to finish Cecilia Zanni Merk A semantic architecture for Industry

33 Current projects

34 The SemSeN project Semantic Sensor Networks The semantic architectures proposed so far do not include a formal model for the communication tasks in the context of Industry 4.0 Goal : Setting up of a formal model based on ontologies for making decisions with performance guarantee for the Internet of Things, where information can be received with losses or after a variable delay. Cecilia Zanni Merk A semantic architecture for Industry

35 The SemSeN project Proposed approach Study of the existing foundational ontologies on sensor networks (Compton et al, 2012) Cecilia Zanni Merk A semantic architecture for Industry

36 The SemSeN project Proposed approach Study of the existing foundational ontologies on sensor networks (Compton et al, 2012) Get a model of the communication layers for Industry 4.0, based on IEEE TSCH / 6tisch standard, to guarantee hard performance and deterministic access Develop an ontology from the previous model and align with the Compton ontology Cecilia Zanni Merk A semantic architecture for Industry

37 The SemSeN project Proposed approach Study of the existing foundational ontologies on sensor networks (Compton et al, 2012) Get a model of the communication layers for Industry 4.0, based on IEEE TSCH / 6tisch standard, to guarantee hard performance and deterministic access Cecilia Zanni Merk A semantic architecture for Industry

38 The SemSeN project Proposed approach Study of the existing foundational ontologies on sensor networks (Compton et al, 2012) Get a model of the communication layers for Industry 4.0, based on IEEE TSCH / 6tisch standard, to guarantee hard performance and deterministic access Develop an ontology from the previous model and align with the Compton ontology Cecilia Zanni Merk A semantic architecture for Industry

39 The SemSeN project Proposed approach Study of the existing foundational ontologies on sensor networks (Compton et al, 2012) Get a model of the communication layers for Industry 4.0, based on IEEE TSCH / 6tisch standard, to guarantee hard performance and deterministic access Develop an ontology from the previous model and align with the Compton ontology Cecilia Zanni Merk A semantic architecture for Industry

40 The SemSeN project ADVERTISING If you have students that would like to come to Strasbourg for an internship of 6 or 8 months from september 2016 on, please give them my address Thank you! Cecilia Zanni Merk A semantic architecture for Industry

41 The HALFbACk project Cross Border Highly AvaiLable Smart FActories in the Cloud French German project to be submitted to the Offensive Sciences 2016 call for projects in april 2016 Goal: Assure highly available manufacturing processes, by forecasting failures of machines, tools, product quality loss, resource flow problems, etc. and by scheduling maintenance, component replacing, process re planning, and even take over the production by another factory, in an optimized and intelligent way. Cecilia Zanni Merk A semantic architecture for Industry

42 The HALFbACk project Cross Border Highly AvaiLable Smart FActories in the Cloud French German project to be submitted to the Offensive Sciences 2016 call for projects in april 2016 Goal: Assure highly available manufacturing processes, by forecasting failures of machines, tools, product quality loss, resource flow problems, etc. and by scheduling maintenance, component replacing, process re planning, and even take over the production by another factory, in an optimized and intelligent way. Cecilia Zanni Merk A semantic architecture for Industry

43 The HALFbACk project Specific scientific challenges in semantic technologies The use of both inductive and deductive techniques to make sense of data and to gain new insights. The joint use of semantics and of data mining algorithms to improve the quality of the interpretation of huge amounts of data and foster data retrieval, reuse, and integration. Top down engineered ontologies and logical inferencing play a key role in providing the vocabularies for querying data, Machine learning techniques will enable the linkage of these data. Cecilia Zanni Merk A semantic architecture for Industry

44 The HALFbACk project Specific scientific challenges in semantic technologies The HALFbACk system will first learn higher level features, expressed as logical axioms from data (coming from sensors, measurements of the characteristics of the final products, machine statistics, and so) and then use these higher level features for non trivial deductive inferences (the optimized prediction of maintenance or the need to use the broker to shift the production). Cecilia Zanni Merk A semantic architecture for Industry

45 Agenda Introduction Basic principles of Semantic Technologies A semantic architecture for Industry 4.0 Current projects A word to finish Cecilia Zanni Merk A semantic architecture for Industry

46 A word to finish

47 A Word to Finish Industry 4.0 is not something that will suddenly arrive overnight. But in the coming years there will be a gradual change to more networked components and machinery, which are connected to the enterprise, which have the intelligence to decide what data is important to share and what isn t. Cecilia Zanni Merk A semantic architecture for Industry

48 Questions or Comments? Thank you!!!