Towards AI powered manufacturing services, processes, and products in an edge-to-cloud-knowlEdge continuum for humans [in-the-loop]

Informazioni relative al progetto

knowlEdge

ID dell’accordo di sovvenzione: 957331

DOI

10.3030/957331

Progetto chiuso

Data della firma CE 28 Luglio 2020

Data di avvio 1 Gennaio 2021

Data di completamento 31 Marzo 2024

Finanziato da

• INDUSTRIAL LEADERSHIP - Leadership in enabling and industrial technologies - Information and Communication Technologies (ICT)

Costo totale

€ 5 996 151,25

Contributo UE

€ 5 996 151,25

Coordinato da

TEKNOLOGIAN TUTKIMUSKESKUS VTT OY
Finland

CORDIS fornisce collegamenti ai risultati finali pubblici e alle pubblicazioni dei progetti ORIZZONTE.

I link ai risultati e alle pubblicazioni dei progetti del 7° PQ, così come i link ad alcuni tipi di risultati specifici come dataset e software, sono recuperati dinamicamente da .OpenAIRE .

Risultati finali

Documents, reports (27)

Initial Description of KnowlEdge Repository

describe the development and the functionalities of projects repository

Final evaluation KPIs

Evaluation methodology and KPIs

Edge AI Learning Pipeline Orchestration

design of the pipeline for model generation based on the data analysis, the (semi-)automatic knowlEdge extraction and the automated model generation

Report on the contribution of standardization

a report that documents contributions to standardization

Evolutionary Requirement Engineering and Innovations [Initial/Updated/Final]

this report describes the requirements analysis all stakeholder requirements (functional and non-functional)

Initial Business models and requirements

the deliverable presents a first version of the results of the stakeholders business requirements and models analyses for AI implementation

Vision, Specifications and System Architecture [Initial/Updated/Final]

a public report documenting the requirements and architecture specifications

Technical review (Interim management report)

Interim management report for the reviewers without the financial part otherwise required for management reports issued at the end of reporting periods

Final Business models and requirements

the deliverable presents the final version of the results of the stakeholders’ business requirements and models analyses for AI implementation

Market Radar and Technology Adaptations [Initial/Updated/Final]

this report describes the evolving landscape of AI technologies related with the industry and manufacturing sector

Final Description of KnowlEdge Repository

this deliverable updates D5.2

Data management plan

specifies how research data are handled during and after a research project

Bootstrapping on AI Models

this deliverable describes all the necessary procedures for execution of AI models in HPC or cloud environment

Pilot Evaluation methodology and implementation plan

Implementation and management plan

Project Manual (quality assurance and risk assessment)

helps the Consortium with quality assurance and identification of relevant quality standards and best practices and means to achieve them and supports the implementation of internal quality checks

User need specification and scenario definition [Initial/Updated/Final]

this report describes the knowlEdge scenarios based on the analysis of user needs. As part of the iterative methodology, the report is updated twice

Dissemination Roadmap & Activities

this deliverable provide the projects dissemination activities

Update on Exploitation strategy & IPR management

Updates the report on the project's Exploitation strategy & IPR management

Last report on dissemination activities
Exploitation Strategy & IPR Management

The deliverable describes how to facilitate the acceptance and utilisation by the market of the developed solutions

AI model Description

describe all the technologies and APIs to make AI models functional in the process environment

Training, Workshops, and Seminars

a report that documents all the workshops and seminars that were planned during the project, their contents and their goals

Generalized Automated Learning for Industrial Environments

this deliverable describes a multi-scale dynamic environment for autonomous learning and re-training of the AI models

Final Data Management and Data Quality modules

this deliverable will contain the final work done in task T3.2. In particular will describe the adopted approach, architecture and structure of the data quality and management modules with respect to the platform.

(Semi-)Automatic Knowledge Discovery for AI Model Generation

initial data analysis results including the description and evaluation of used methods

First report on dissemination activities
Initial Data Management and Data Quality modules

this deliverable will contain the initial work done in task T32 In particular will describe the initial approach architecture and structure of the data quality and management modules

Other (15)

Initial knowlEdge website

this deliverable provide a brief description of the initial projects website and is functionalities

Initial Site-wide Data Storage and Governance suit

provides the overview of the results produced in T3.3. This deliverable lists the data models, protocols and planned architecture of the components to be developed.

Human-AI Collaboration and Domain Knowledge Fusion

Describe the collaboration of humans with an automated AI pipeline.

Final Decision Support Framework

this deliverable updates D7.2.

Initial explainable mechanisms DFS

a prototype deliverable that documents the design and implementation of the integrated 2-Axis DSF, that encapsulates decision strategies and rules, KPIs, as well as a number of parameters that affect product and process quality.

Initial user-centric dashboards as enable explainable AI

a prototype deliverable that documents all the enhanced visualizations that support the decision making mechanisms of the 2-Axis DSF

Initial Provisioning and Deployment Management Tool

planned Concept, architecture and communication protocols that shall be developed as part of T6.3

Final site-wide data collection and integration toolkit

provides the overview of the results produced in T3.1. This deliverable updates on D3.1 with the details of the deployments and results.

Final knowlEdge Marketplace Platform

this deliverable updates D5.4

Full knowledge website

this deliverable provide a brief description of the full project website and is functionalities

Final Provisioning and Deployment Management Tool

this deliverable updates D6.4

Initial site-wide data collection and integration toolkit

provides the overview of the results produced in T31 This deliverable lists the data models protocols and planned architecture of the components to be developed

Final Site-wide Data Storage and Governance suit

provides the overview of the results produced in T3.3. This deliverable updates on D3.5 with the details of the deployments and results.

Final user-centric dashboards as enable explainable AI

this deliverable updates D7.4

Initial KnowlEdge Marketplace Platform

a public prototype deliverable that documents the architecture, the features and the capabilities of the knowlEdge marketplace

Demonstrators, pilots, prototypes (6)

"Pilot#2: K-T, Plastic parts for Car Industry"
"Pilot#3: Bonfiglioli, Gear machines for Industrial applications – Final"

"Final assessment of the pilot#3: Bonfiglioli, Gear machines for Industrial applications"

"Pilot3#: Bonfiiglioli, Gear machines for Industrial applications"
"Pilot#1: Parmalat, Milk Industry"
"Pilot#1: Parmalat, Milk Industry – Final"

"Final assessmentof the pilot#1: Parmalat, Milk Industry"

"Pilot#2: Kautex Textron, Plastic parts for Car Industry – Final"

"Final assessmen tof the pilot#2: Kautex Textron, Plastic parts for Car Industry"

Open Research Data Pilot (1)

Evaluation results and KPI assessment

AI maturity tool results and KPI performance indicators (assessment framework). Describes the overall impact of the application of AI models on the pilot sites. These results are provided as open research data for others to access and re-use and may for example be used to validate results in scientific publications.

Pubblicazioni

Conference proceedings (6)

Towards Industry 5.0 – A Trustworthy AI Framework for Digital Manufacturing with Humans in Control

Autori: Usman Wajid, Alexandros Nizamis, Victor Anaya
Pubblicato in: Proceedings of Interoperability for Enterprise Systems and Applications Workshops co-located with 11th International Conference on Interoperability for Enterprise Systems and Applications (I-ESA 2022), Numero 2022, 2022, Pagina/e 6, ISSN 1613-0073
Editore: CEUR

LOGIC: Probabilistic Machine Learning for Time Series Classification

Autori: Fabian Berns, Jan David Hüwel, Christian Beecks
Pubblicato in: IEEE International Conference on Data Mining, Numero 2021, 2021, Pagina/e 1000-1005, ISBN 978-1-6654-2398-4
Editore: IEEE
DOI: 10.1109/icdm51629.2021.00113

Automated Kernel Search for Gaussian Processes on Data Streams

Autori: Jan David Hüwel; Fabian Berns; Christian Beecks
Pubblicato in: IEEE International Conference on Big Data, Numero 2021, 2021, Pagina/e 3584-3588, ISBN 978-1-6654-3902-2
Editore: IEEE
DOI: 10.1109/bigdata52589.2021.9671767

knowlEdge Project –Concept, Methodology and Innovations for Artificial Intelligence in Industry 4.0

Autori: Alvarez-Napagao, Sergio; Ashmore, Boki; Barroso, Marta; Barrué, Cristian; Beecks, Christian; Berns, Fabian; Bosi, Ilaria; Chala, Sisay Adugna; Ciulli, Nicola; Garcia-Gasulla, Marta; Grass, Alexander; Ioannidis, Dimosthenis; Jakubiak, Natalia; Köpke, Karl; Lämsä, Ville; Megias, Pedro; Nizamis, Alexandros; Pastrone, Claudio; Rossini, Rosaria; Sànchez-Marrè, Miquel; Ziliotti, Luca
Pubblicato in: IEEE 19th International Conference on Industrial Informatics, Numero 2021, 2021, Pagina/e 7, ISBN 978-1-7281-4395-8
Editore: IEEE
DOI: 10.1109/indin45523.2021.9557410

Evaluating the Lottery Ticket Hypothesis to Sparsify Neural Networks for Time Series Classification

Autori: Georg Stefan Schlake; Jan David Hüwel; Fabian Berns; Christian Beecks
Pubblicato in: IEEE International Conference on Data Engineering Workshops, Numero 2022, 2022, Pagina/e 70-73, ISBN 978-1-6654-8104-5
Editore: IEEE
DOI: 10.1109/icdew55742.2022.00015

Local Gaussian Process Model Inference Classification for Time Series Data

Autori: Fabian Berns, Joschka Hannes Strueber, Christian Beecks
Pubblicato in: 33rd International Conference on Scientific and Statistical Database Management, Numero 2021, 2021, Pagina/e 209-213, ISBN 978-1-4503-8413-1
Editore: ACM
DOI: 10.1145/3468791.3468839

Book chapters (7)

Anomaly Detection in Manufacturing

Autori: Jona Scholz Maike Holtkemper Alexander Graß Christian Beecks
Pubblicato in: Artificial Intelligence in Manufacturing, Numero 2024, 2024, Pagina/e 351-360, ISBN 978-3-031-46452-2
Editore: Springer Nature Switzerland
DOI: 10.1007/978-3-031-46452-2_20

Boosting AutoML and XAI in Manufacturing: AI Model Generation Framework

Autori: Marta Barroso Daniel Hinjos Pablo A. Martin Marta Gonzalez-Mallo Victor Gimenez-Abalos Sergio Alvarez-Napagao
Pubblicato in: Artificial Intelligence in Manufacturing, Numero 2024, 2024, Pagina/e 333-350, ISBN 978-3-031-46452-2
Editore: Springer Nature Switzerland
DOI: 10.1007/978-3-031-46452-2_19

Human-AI Interaction for Semantic Knowledge Enrichment of AI Model Output

Autori: Sisay Adugna Chala Alexander Graß
Pubblicato in: Artificial Intelligence in Manufacturing, Numero 2024, 2024, Pagina/e 43-54, ISBN 978-3-031-46452-2
Editore: Springer Nature Switzerland
DOI: 10.1007/978-3-031-46452-2_3

Designing Human and Artificial Intelligence Interactions in Industry X

Autori: Stefan Walter
Pubblicato in: Service Design for Emerging Technologies Product Development, Numero 21 July 2023, 2023, Pagina/e 207-232, ISBN 978-3-031-29306-1
Editore: Springer, Cham
DOI: 10.1007/978-3-031-29306-1_12

Designing a Marketplace to Exchange AI Models for Industry 5.0

Autori: Alexandros Nizamis Georg Schlake Georgios Siachamis Vasileios Dimitriadis Christos Patsonakis Christian Beecks Dimosthenis Ioannidis Konstantinos Votis Dimitrios Tzovaras
Pubblicato in: Artificial Intelligence in Manufacturing, Numero 2024, 2024, Pagina/e 27-41, ISBN 978-3-031-46452-2
Editore: Springer Nature Switzerland
DOI: 10.1007/978-3-031-46452-2_2

A Manufacturing Digital Twin Framework

Autori: Victor Anaya Enrico Alberti Gabriele Scivoletto
Pubblicato in: Artificial Intelligence in Manufacturing, Numero 2024, 2024, Pagina/e 181-193, ISBN 978-3-031-46452-2
Editore: Springer Nature Switzerland
DOI: 10.1007/978-3-031-46452-2_10

Advancing Networked Production Through Decentralised Technical Intelligence

Autori: Stefan Walter Markku Mikkola
Pubblicato in: Artificial Intelligence in Manufacturing, Numero 2024, 2024, Pagina/e 281-300, ISBN 978-3-031-46452-2
Editore: Springer Nature Switzerland
DOI: 10.1007/978-3-031-46452-2_16

Peer reviewed articles (7)

Production Scheduling Optimization enabled by Digital Cognitive Platform

Autori: Konstantinos Georgiadis, Alexandros Nizamis, Thanasis Vafeiadis, Dimosthenis Ioannidis, Dimitrios Tzovaras
Pubblicato in: Procedia Computer Science, Numero 204, 2022, Pagina/e 424-431, ISSN 1877-0509
Editore: Elsevier
DOI: 10.1016/j.procs.2022.08.052

Automated Model Inference for Gaussian Processes: An Overview of State-of-the-Art Methods and Algorithms

Autori: Fabian Berns, Jan Hüwel, Christian Beecks
Pubblicato in: SN Computer Science, Numero 3 (4), 2022, Pagina/e 300, ISSN 2661-8907
Editore: Springer Nature
DOI: 10.1007/s42979-022-01186-x

AI Lifecycle Zero-Touch Orchestration within the Edge-to-Cloud Continuum for Industry 5.0

Autori: Alberti, Enrico; Alvarez-Napagao, Sergio; Anaya, Victor; Barroso, Marta; Barrué, Cristian; Beecks, Christian; Bergamasco, Letizia; Chala, Sisay Adugna; Gimenez-Abalos, Victor; Graß, Alexander; Hinjos, Daniel; Holtkemper, Maike; Jakubiak, Natalia; Nizamis, Alexandros; Pristeri, Edoardo; Sànchez-Marrè, Miquel; Schlake, Georg; Scholz, Jona; Scivoletto, Gabriele; Walter, Stefan
Pubblicato in: Systems, Numero 12, 48, 2024, Pagina/e 32, ISSN 2079-8954
Editore: MDPI
DOI: 10.3390/systems12020048

Explaining the Behaviour of Reinforcement Learning Agents in a Multi-Agent Cooperative Environment Using Policy Graphs

Autori: Domenech i Vila, M.; Gnatyshak, D.; Tormos, A.; Gimenez-Abalos, V.; Alvarez-Napagao, S.
Pubblicato in: Electronics, Numero 13(3), 573, 2024, Pagina/e 19, ISSN 2079-9292
Editore: MDPI
DOI: 10.3390/electronics13030573

Impacts of AI driven manufacturing processes on supply chains: the contributions of the knowlEdge project

Autori: Stefan Walter
Pubblicato in: Transportation Research Procedia, Numero 72, 2023, Pagina/e 3443-3449, ISSN 2352-1465
Editore: Elsevier
DOI: 10.1016/j.trpro.2023.11.773

AI impacts on supply chain performance : a manufacturing use case study

Autori: Stefan Walter
Pubblicato in: Discover Artificial Intelligence, Numero 3:28, 2023, ISSN 2731-0809
Editore: Springer Nature
DOI: 10.1007/s44163-023-00061-9

Testing Reinforcement Learning Explainability Methods in a Multi-Agent Cooperative Environment

Autori: Domènech Vila, Marc; Gnatyshak, Dmitry; Tormos Llorente, Adrián; Álvarez Napagao, Sergio
Pubblicato in: Frontiers in Artificial Intelligence and Applications, Numero Volume 356: Artificial Intelligence Research and Development, 2022, Pagina/e 355-364, ISSN 0922-6389
Editore: IOS Press
DOI: 10.3233/faia220358