Foresights and Recommendations of Digital Manufacturing ...

ConnectedFactories 2

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 873086.

D 3.3 Foresights and Recommendations of Digital

Manufacturing Platforms for a Digital Europe - First Iteration v1.0

Action acronym

ConnectedFactories 2

Action Full Title Global-leading smart manufacturing through digital platforms, cross-cutting features and skilled workforce

Grant Agreement Number 873086

Instrument CSA: Coordination and Support Action

Project coordinator VTT

Deliverable Number D3.3

Deliverable Title Foresights and Recommendations of Digital Manufacturing Platforms for a Digital Europe

Lead Beneficiary FPM

Work package 3

Work package leader INNOVALIA

Dissemination level1 Public

Type2 R

Due date according to DoA 31 May 2021

Actual submission date 23 June 2021

Main editors and contributors: FPM, INNOVALIA, EFFRA, LSEC

1 PU: Public, CO: Confidential, only for members of the consortium (including the Commission Services) 2 RE: Report, OT: Other; ORDP: Open Research Data Pilot

Ref. Ares(2021)4188793 - 28/06/2021

D 3.3 Foresights and Recommendations of Digital Manufacturing Platforms for a Digital Europe - First Iteration v.1.0

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 873086. 2 (52)

VERSION MANAGEMENT

Name Beneficiary Author(s): Sergio Gusmeroli, Silvia Razzetti FPM Contributor(s): Chris Decubber

Jesus Alonso, José de Andrés, Oscar Lázaro EFFRA, INNOVALIA

Reviewed by: Chris Decubber Ulrich Seldeslachts

EFFRA LSEC

Revision No.

Date Description Author

0.1 29/03/2021 Draft outline S.G. 0.2 10/06/2021 Final draft S.R 0.3 11/06/2021 Review C.D. 0.4 17/06/2021 Review U.S. 0.5 21/06/2021 Consolidated version S.R. 1.0 22/06/2021 Final version S.G.

Abbreviations and acronyms

TERMS, ABBREVIATIONS AND ACRONYMS

AI Artificial Intelligence AQ Answer Question AAS Asset Administration Shell B2B Business-To-Business CC Competence Center CE Circular Economy CF Connected Factories CSA Coordination and Support Actions DEP Digital Europe Programme DFA Digital Factory Alliance for Digital Manufacturing Platforms DGA Data Governance Act DIH Digital Innovation Hub DMP Digital Manufacturing Platform DT Digital Transformation EC European Commission



EDIH European Digital Innovation Hub EU European Union HEP Horizon Europe Programme HPC High Performance Computing ICT Information and Communication Technology IDSA International Data Space IDSA International Data Spaces Association IT Information Technology MPFQ Materials, Processes, Functions and Quality OEE Overall Equipment Effectiveness OT Operation Technology PPP Public Private Partnership SME Small Medium Enterprise TEF Testing and Experimentation Facility



TABLE OF CONTENT Executive Summary ........................................................................................................................................... 6 1 DMP Projects and Cases in EFFRA Innovation Portal ................................................................................ 8

1.1 CF1 Pathways ..................................................................................................................................... 9 1.1.1 The Autonomous Smart Factories pathway ................................................................................ 9

1.1.2 The Collaborative Product-Service Factories pathway .............................................................. 10

1.1.3 The Hyperconnected Factories pathway ................................................................................... 11

1.1.4 The CyberSecurity Pathway ....................................................................................................... 11

1.2 CF2 Pathways ................................................................................................................................... 12 1.2.1 The Circular Economy pathway ................................................................................................. 12

1.2.2 The Data Spaces pathway for DMP ........................................................................................... 13

1.3 Mapping process and status of DMP Projects on the Catalogue .................................................... 14 1.3.1 DMP Projects’ Cases in Smart Autonomous Factory ................................................................. 16

1.3.2 DMP Projects’ Cases in Hyperconnected Factory ...................................................................... 19

1.3.3 DMP Projects’ Cases in Product-Service Factory ....................................................................... 20

1.3.4 DMP Projects’ Cases in Circular Economy Pathway .................................................................. 24

1.3.5 DMP Projects’ Cases in Data Space Pathway ............................................................................. 26

1.4 Collection of lessons learned from cases in EFFRA Innovation Portal ............................................ 27 1.5 Ongoing works ................................................................................................................................. 28

2 Foresight and Recommendations towards Digital Europe programme .................................................. 29 2.1 Data Spaces in Manufacturing ......................................................................................................... 30

2.1.1 Data Spaces in Manufacturing; Interactive Workshop .............................................................. 34

2.2 AI TEF in Manufacturing .................................................................................................................. 36 2.2.1 AI TEF in Manufacturing; Interactive Workshop ....................................................................... 38

2.3 European Digital Innovation Hubs in Manufacturing ...................................................................... 40 2.3.1 EDIH in Manufacturing; Interactive Workshop.......................................................................... 43

3 Foresight and Recommendations towards Horizon Europe programme ............................................... 45 3.1 Human-Robot Interaction and Industry 5.0 .................................................................................... 45 3.2 Zero Defect, Zero Waste, Green and Resilient Manufacturing ....................................................... 47 3.3 Artificial Intelligence for Autonomous, Hyperconnected and Product-Service Factories ............... 49

4 Conclusions and Future Outlook ............................................................................................................. 51

TABLE OF FIGURES Figure 1 ConnectedFactories, Projects and Pathways....................................................................................... 9 Figure 2 Autonomous Smart Factories Pathway ............................................................................................. 10 Figure 3 Collaborative Product-Service pathway ............................................................................................ 10



Figure 4 Hyperconnected Factories pathway .................................................................................................. 11 Figure 5 Cybersecurity Pathway ...................................................................................................................... 12 Figure 6 Circular Economy Pathway ................................................................................................................ 13 Figure 7 Data Spaces pathway ......................................................................................................................... 13 Figure 8 Example of use case mapping ........................................................................................................... 15 Figure 9 ZDMP and QU4LITY maturity levels (Autonomous Smart Factories) ................................................ 16 Figure 10 ZDMP cases in Autonomous Smart Factories pathway ................................................................... 17 Figure 11 QU4LITY cases in Autonomous Smart Factories pathway ............................................................... 18 Figure 12 ZDMP maturity level (Hyperconnected) .......................................................................................... 19 Figure 13 ZDMP cases in Hyperconnected pathway ....................................................................................... 20 Figure 14 ZDMP cases in Collaborative Product-Service pathway .................................................................. 21 Figure 15 ZDMP maturity level (Collaborative-Product service) ..................................................................... 22 Figure 16 QU4LITY cases in Collaborative Product-Service pathway .............................................................. 23 Figure 17 QU4LITY maturity level (Collaborative-Product service) ................................................................. 24 Figure 18 EFPF cases in Circular economy pathway ........................................................................................ 24 Figure 19 EFPF maturity level (Circular economy) .......................................................................................... 25 Figure 20 QU4LITY cases in Data Spaces pathway .......................................................................................... 26 Figure 21 QU4LITY maturity level (Data Spaces) ............................................................................................. 27 Figure 22 Lessons learned filter in EFFRA Portal ............................................................................................ 28 Figure 23 ‘Manufacturing Service Buses for Data Interoperability’ example ................................................. 28 Figure 24 The workshop interactive board ..................................................................................................... 30 Figure 25 European Strategy for Data outlined by the Digital Europe Programme ........................................ 32 Figure 26 The nine key sectors for European Data Spaces ............................................................................. 33 Figure 27 Schema for Smart Factory and Value Chain Data Space ................................................................. 34 Figure 28 The "Data Space in manufacturing" DEP session ............................................................................ 34 Figure 29 AI TEF main concepts ....................................................................................................................... 37 Figure 30 TEF-centric Collaboration ................................................................................................................ 38 Figure 31 The "AI TEF in manufacturing" DEP session .................................................................................... 39 Figure 32 The four main classes of services offered by an EDIH ..................................................................... 41 Figure 33 The JRC platform for EDIH catalogue .............................................................................................. 42 Figure 34 The "European DIHs in manufacturing" DEP session ...................................................................... 43 Figure 35 The "Human-Robot interaction & Industry5.0" HEP session........................................................... 46 Figure 36 The "Zero Waste, Zero Defect, Green & Resilient manufacturing" HEP session ............................. 48 Figure 37 The "AI technologies for manufacturing" HEP session .................................................................... 50

file://Users/silviarazzetti/Dropbox%20(DIG)/0_CF2/WP3/CF2%20Deliverable%203.3_1.0_FINAL.docx#_Toc75277812



Executive Summary Digital Manufacturing Platforms (DMP) are implementing the Digital Transformation of Manufacturing Industry, enabling Products to become smarter, Processes to become more efficient, Platforms to become open and interoperable. This also impacts the Performance of the manufacturing companies, the People skills and competencies and implies the set-up of new Partnership between Digital and Manufacturing stakeholders (e.g. DIHs for Manufacturing) as well as the definition of new Performance indicators.

In 2018-2020 H2020 program, DMPs are mostly developed by 6 Innovation Actions in DT-ICT-07 two waves and are coordinated and supported by Connected Factories 2 CSA (CF2), but many other projects are being developing DMPs with different targets and foci, es for instance DT-ICT-03 DIH projects on AI, Digital Twins, Human-Robot interaction or ICT-08 projects about Cybersecurity applied to Manufacturing.

CF2 WP3 (Cases: from industrial state-of-the art to demonstrators and pilot lines) is focussing on Industrial Cases and how they are demonstrating the benefits of DMPs in realistic industrial environments. The key question addressed by D3.3 is if and to what extent such current or recent experimentations could pave the way towards the future of manufacturing in the next five years.

Connected Factories CSA (CF1) (which ran from 2017 until 2019) and its successor CF2 CSA aim at accomplishing this future perspective task by elaborating pathways, which could depict in a two dimensional matrix, how certain manufacturing dimensions (in the rows) could evolve along DMP-driven pathways in a 5-steps approach (in the columns). Six Pathways have been for the moment considered: the Smart Autonomous Factories Pathway, the Collaborative Product-Service Factories Pathway, the Hyperconnected Factories Pathway and the Cybersecurity Pathway in CF1, and in addition - for the time being - the new Circular Economy and the Data Space pathways in CF2 (considering that the Cybersecurity Pathway is further evolving in CF2) and a third one – Artificial Intelligence Pathway – under development.

The first objective of D3.3 is to map the industrial cases collected in the EFFRA Innovation Portal and to conduct an analysis of future foresights and recommendations. The information gathered during the collection work has been used to populate the EFFRA Innovation Portal, where projects can showcase their experience. This activity has enabled the preparation of a catalogue of well-described cases, called ‘Digital Transformation Pathway Cases and European demonstration Infrastructure’, which is a dynamically evolving resource, embedded within the EFFRA Innovation Portal that brings together inspiring use cases and demonstrators, pointing also to additional information and contact points. This tool allows quick and easy access to cases containing relevant information on technologies, standards and tools, which enable high levels of maturity to be achieved across the digitalisation pathways. The work’s main focus was on acquiring information on projects belonging to the DMP Cluster. However, experiences on other innovation initiatives at national and regional level have also been collected.

The second objective of D3.3 is to consider the industrial cases in a future perspective as directed by the EC in its 2021-2027 Multiannual Financial Framework. Two programs have been considered: the Digital Europe program and the Twin Transition Dimension of Cluster 4 in the Horizon Europe program. In both cases, CF2 organised highly interactive sessions via the MURAL digital workspace for visual collaboration and subsequent 1:1 interviews to better understand the meaning of the sticky notes produced. The analysis was run in collaboration with 15 projects who participated to the interactive workshops, discussing about the six main manufacturing challenges (boosted by the two European programs) that were presented. The exercise was useful to identify the different approaches and specificities of each project with

https://portal.effra.eu/results/list/result?result_search%5Btaxon%5D=&result_search%5BtextQuery%5D=&result_search%5Bproject%5D=&result_search%5Bsort%5D=Relevancy&result_search%5BsortDirection%5D=Descending&result_search%5Blist__1045%5D%5B%5D=2092&result_search%5Blistcheckbox__1045%5D=1045




respect to the six topics, sharing results and ideas through the entire ecosystem of projects. In D3.3, the six challenges are presented, their main objectives and the European Commission expectations are described. For each of them, the inputs collected in the online board and during the interviews with partners are summarized, in order to:

• Identify projects and use cases dealing with that challenge; • Provide a full picture of different approaches that have been adopted, with the objective of

presenting concrete examples and success stories to stimulate the interest and the discussion; • Identify main foreseen challenges and recommendations for the future, comparing several opinions

and point of views, with the perspective of participating to next European calls.



1 DMP Projects and Cases in EFFRA Innovation Portal As part of ConnectedFactories 2 WP3, work has been done on collecting use cases and best practices within European innovation projects. All the information gathered has been used to populate the EFFRA Innovation Portal (EFFRA Innovation Portal3) as reported in D3.2.

Specifically, the EFFRA Innovation Portal, in which projects can showcase their experience, will support the dissemination of information on innovation projects and foster SMEs' access to success stories and best practices. A catalogue of well-described cases has been developed with all the information collected in the EFFRA Innovation Portal.

This catalogue, called ‘Digital Transformation Pathway Cases and European demonstration Infrastructure’, is integrated into the EFFRA Innovation Portal and properly announced on the ConnectedFactories website4.

This information acquisition work has focused on extending the information available on projects belonging to the DMP Cluster and associated to the same call (DT-ICT-07) as ConnectedFactories 2. However, experiences on other innovation initiatives at national and regional level have also been collected. A more extensive extract of the results of the work done can be found in deliverable 3.2.

Some additional contributions have been helpful for this report, such as the workshop that CF2 project organised on 24th March 2021 ‘Pathways to digitalisation of manufacturing and associated use cases’ (Presentations available here5 and Recordings available here6) and the associated deliverable 5.2.

The EFFRA Innovation Portal includes information on standards, business model aspects and technologies used in the projects (see also CF2 Deliverables 1.1 and 1.2), also building on the work of the CSA 'ConnectedFactories'.

In the first stage the mapping of projects and use cases or demonstrators focusses on four pathways that were developed and validated during ConnectedFactories 1: Autonomous Smart Factories Pathway, Hyperconnected Factories Pathway, Collaborative Product-Service Factories Pathway and the CyberSecurity Pathway; while in the continuation project (currently active) ConnectedFactories 2 (CF2), two more pathways have been developed on Data Spaces and Circular Economy and a third one on Artificial Intelligence is still work in progress.

As can be seen in Figure 1, during the two CF CSAs (CF1 and CF2) several innovation projects have been covered and several pathways have been developed. The pathways provide a quick and intuitive way to place the projects, demonstrators or use cases according to the context and goals of manufacturing optimisation by means of digital technologies. Part of the work done in support of this deliverable focuses on collecting information and uploading it to the EFFRA Innovation Portal.

3 https://portal.effra.eu/projects 4 https://www.connectedfactories.eu/news/digital-transformation-cases-catalogue-now-launched 5 https://cloud.effra.eu/index.php/s/ilUESNNA5leeQS5 6 https://www.youtube.com/playlist?list=PLpaccoqg8rxbKwVkTO5fKXRL5eDlQauVt


https://portal.effra.eu/projects

https://www.connectedfactories.eu/news/digital-transformation-cases-catalogue-now-launched

https://cloud.effra.eu/index.php/s/ilUESNNA5leeQS5

https://www.youtube.com/playlist?list=PLpaccoqg8rxbKwVkTO5fKXRL5eDlQauVt



Figure 1 ConnectedFactories, Projects and Pathways

The following sub-sections briefly explain the pathways of CF1 and CF2, provide some pointers to where further information is available, shows examples of mapping and summarise the results obtained.

1.1 CF1 Pathways

This section presents a summary of the work done in ConnectedFactories 1. Below on this section can be found pictures of the developed pathways, a brief description of each of them and some pointers to places where more detailed information can be obtained.

1.1.1 The Autonomous Smart Factories pathway This pathway focuses on digitalisation within the Factory environment (From machinery, shop floor up to entire production site) and is based on the so-called ‘Automation Pyramid’.

The Autonomous Smart Factories pathway (Figure 2) is developed with a scope towards optimised and sustainable manufacturing including advanced human-in-the-loop workspaces. Further information could be found on the ConnectedFactories portal7.

7 https://www.connectedfactories.eu/autonomous-smart-factories-pathway

https://www.connectedfactories.eu/autonomous-smart-factories-pathway



Figure 2 Autonomous Smart Factories Pathway

1.1.2 The Collaborative Product-Service Factories pathway The Collaborative Product-service pathway (Figure 3) is developed around the concept of servitisation. The manufacturing company is the primary perspective of the pathway aiming at reflecting the progressive adoption of service-oriented business model without dropping their product-oriented business model. Unlike large companies, SMEs have more difficulties to understand the product-service concept, particularly in the production high-tech components. For more information consult CF website8.

Figure 3 Collaborative Product-Service pathway

8 https://www.connectedfactories.eu/collaborative-product-service-factories-pathway

https://www.connectedfactories.eu/collaborative-product-service-factories-pathway



1.1.3 The Hyperconnected Factories pathway The Hyperconnected Factories pathway (Figure 4) was developed for networked enterprises in complex, dynamic supply chains and value networks.

At its levels, multi-purpose digital tools are progressively implemented and complemented by dedicated tools, towards inter-factory integration through the IoT connection until last level where a dynamic IT connection can be established with new business partners or suppliers.

A complete description can be found on the Hyperconnected Factories Pathway section9 on the ConnectedFactories website.

Figure 4 Hyperconnected Factories pathway

1.1.4 The CyberSecurity Pathway The CyberSecurity Pathway (Figure 5) was added to the ConnectedFactories project as it was seen as a major inhibitor and enabler to the DMP developments.

During the first stage, as presented on the ConnectedFactories platform10, it indicated a maturity model alongside the Pathway evolutions from the three other Pathways, where the evolution of the Manufacturing company in its digitalization process is being supported by a series of basic requirements needed for the Digital Manufacturing Platforms to contain.

9 https://www.connectedfactories.eu/hyperconnected-factories-pathway 10 https://www.effra.eu/news/challenges-explored-cybersecurity-workshop

https://www.connectedfactories.eu/hyperconnected-factories-pathway

https://www.effra.eu/news/challenges-explored-cybersecurity-workshop



Figure 5 Cybersecurity Pathway

These were set-up on the basis of existing CyberSecurity standards, processes and control frameworks being widely used in the industry. They were both guidance to DMP platform developers as well as manufacturing companies, containing instructions for solutions and controls.

During the CF2 project, and alongside the two additional Pathways, the CyberSecurity Pathway evolves for DMP considering intensive use of data or AI capabilities. With this additional CyberSecurity measures will need to be put into action supporting new solutions and ongoing developments.

1.2 CF2 Pathways

Building further on the works done in ConnectedFactories in the previous section, the following is an overview of the ongoing work done in CF2. Below can be found pictures of the two developing pathways, with a brief description of each of them and some pointers to deliverables where more detailed information can be obtained.

1.2.1 The Circular Economy pathway The Circular Economy pathway rises to highlight the importance of the Circular Economy (CE) paradigm for the sustainable development of the manufacturing sector. It is a maturity assessment model which allows companies to estimate their position into the CE paradigm, considering the levels and the dimensions depicted in Figure 6. This is one of the CF2 developed pathways. For further information consult deliverable 2.1.

https://www.connectedfactories.eu/sites/default/files/d2.1_pathways_from_dt-ict-07_-_first_iteration.pdf

https://www.connectedfactories.eu/sites/default/files/d2.1_pathways_from_dt-ict-07_-_first_iteration.pdf



Figure 6 Circular Economy Pathway

1.2.2 The Data Spaces pathway for DMP The Data Spaces pathway (Figure 7) developed by CF2 is made by crossing six Data Technology enablers (Data Management, Data Protection, Data Processing, Data Analytics, Data Visualisation and Data Sharing) versus a five-levels maturity model (No Data Control, Data Silos, Data Bridges, Data Interoperability and Data Valorisation). Information of the development is available on the deliverable 2.2.

Figure 7 Data Spaces pathway

Linearity (1°) Industrial CE Piloting (2°)

Systemic Material Management (3°)

Circular Economy Thinking (4°) Circularity (5°)

1 2 3 4 5

Products Any toxic substances should not be used to create products

Material used should be the minimum amount required to respect product functionalities and design (resource sufficiency)

Systematical identification of possibilities to reuse, refurbish and remanufacture

Ecodesign of products (must be easy to be disassembled, repaired, remanufactured and its components should be recyclable)

Changing business model towards product-service-systems and X as service approaches at ecosystem level

Process Quality monitoring to avoidunnecessary scraps

Production processes must require limited amount of energy

Transportations modes (reverse logistics), internal recycling of materials

Building industrial synergies/closed loop models

Circular systems and process at value network/ecosystemlevel

Platform Information technologies to gather processes data

Company systems integration (e.g. ERP, MRP, PDM, PLM…)

Disassembly and Remanufacturing enabling technologies introduced on the shop-floor

Digital platform integration enabling the interaction among value chain actors

Collaborative business processes and workflows are used over the product life cycle

People Ad-hoc engagement of individuals, not comprehensive engagement

Engagement and awareness raising, systemic empowering through champions

Cultural transformation and qualified people (skills)

Circular suppliers selection and value network level indicators

Sustainable government requirements and European Green Deal?

Partnership Contractual obligations based onregulations

Code of conduct guidelines for circularity/sustainability of materials. There areoccasionalcollaboration around circularity/sustainability

Circularity objectives discussed at value chain level. Stakeholder engagement systematically considered.

Capabilities to dynamic collaboration new partners ,research and innovation partnerships launched

Digital ecosystems enabling circularity, partnering with customers making use of knowledge

Performance Reporting and measuring onlythe legal responsibilities related to recycling and otherenvironmental obligations

Targets and KPIs set up for circularity pilots

Operative KPIs for resourcesufficiency/ decrease of waste etc.

LCA analysis over the productslifecycle, KPIs for foot print.

Circularityindicators(basedon the forthcoming standard)

Dimensions / Levels

Level INo Data Control

Level IIData Silos

Level IIIData Bridges

Level IV Data Interoperability

Level VData Valorization

Data Management

Data Protection

Data Processing

Data Analytics

Data Visualization

Data Sharing

Digital Transformation - Industry 4.0

Data are generated, processed

and visualized by CPPS and I4.0 systems

Enterprise Applications (ERP, SCM, PLM, CRM)

collect, store and visualize

Data

Data Spaces Interoperability

Complex applications require data

from different sources

Data Engineering &

Security Privacy

Data Sovereignty and GDPR

Data Sharing Spaces

AI-driven applications;

Digital Assistants;

VR/AR

Data-driven Business Models

Flexible Data Marketplaces

Data Economy and Industrial Data Platforms

https://www.connectedfactories.eu/sites/default/files/d2.2_pathways_cross-fertilisation.pdf



1.3 Mapping process and status of DMP Projects on the Catalogue

The mapping of cases is supported by the EFFRA Innovation Portal, so the mapping can be applied to any project and the associated cases (use cases, demonstrators, pilots) that are included on the EFFRA Innovation Portal and consists of positioning according to the different levels of the pathways. This process is based on selecting within the pathway the maturity level, as well as providing information about key enablers and cross-cutting aspects, such as standards, technological enablers as well as business model aspects. More information on the mapping process can be found in deliverable 3.1.

As aforementioned the focus has been on DMP projects. During this task, projects have been engaged to map their cases in the EFFRA Innovation Portal, to show the mapping process in case they had doubts and to collect the generated results.

At the time of submission of this deliverable, only the first wave of DT-ICT-07 projects have mapped their pilots, as it is an early stage for the second wave of projects (DT-ICT-07 2019), which is not yet sufficiently developed to map its pilots. More detailed results on the mapping of DT-ICT-07 pilots can be found in deliverable 3.2

Figure 8 below shows an example of mapping according to the Autonomous Smart Factories pathway of one of the pilots of the QU4LITY project. In addition, most of the cases collected during this process have been selected as part of the Catalogue, of well described cases, ‘Digital Transformation Pathway Cases and European demonstration Infrastructure’.

This process brings knowledge on diverse technologies, that enable pilots to reach the different levels of the digitalisation pathways, together with inspiring use cases and demonstrators, pointing also to additional information and contact points.

https://www.connectedfactories.eu/sites/default/files/d3.1_initial_scouting_collection_and_fine-tuned_mapping_methodology.pdf





Figure 8 Example of use case mapping

https://portal.effra.eu/result/show/3922



Apart from the DMP projects, many other cases have been mapped and registered in the ‘Digital Transformation Pathway Cases and European demonstration Infrastructure’ catalogue. Deliverable 3.2 provides a more detailed summary of the results obtained with this work. In addition, the actual use cases mapped according to the ConnectedFactories pathways can be consulted here, which could also be restricted to cases that were retained for the catalogue here.

At the time of submission of this deliverable, the following subsections 1.3.1 to 1.3.5 report on the status of the mapped DMP project cases and present structured descriptions of particular cases, which reinforce the explanation of the relationship between the DT-ICT-07 projects and the itineraries. This information is also presented in Annex I of D3.2. This still is work in progress and as said before only the DT-ICT-2018 projects were developed enough to map their pilots.

This mapping process has been done by the experts working on each pilot and have been engaged to collaborate with this task through different internal and public workshops, such as the aforementioned on 24th March 2021 ‘Pathways to digitalisation of manufacturing and associated use cases’.

The following sections present the results of the mapping of DT-ICT-07 projects. As stated above, only the first wave of projects (ZDMP, QU4LITY and EFPF) were sufficiently developed to map their pilots. Each Project has selected the pathways that match their pilots to carry out the mapping process. This is still a work in progress and more information and use case mapping is expected to be collected in the coming months.

1.3.1 DMP Projects’ Cases in Smart Autonomous Factory The Smart Autonomous pathway is one of the three pathways developed in CF1 and it is also the most selected pathway for mapping use cases. Namely, two (out of three) EU Horizon 2020 DT-ICT-07-2018-2019 DMP projects have chosen this pathway to map their use cases, positioning mainly between the fourth and the fifth level: ZDMP and QU4LITY (Figure 9).

Figure 9 ZDMP and QU4LITY maturity levels (Autonomous Smart Factories)

https://portal.effra.eu/results/list/result?result_search%5Btaxon%5D=&result_search%5BtextQuery%5D=&result_search%5Btype%5D%5B%5D=Demonstrators&result_search%5Bcalltopic%5D=&result_search%5Bcall%5D=&result_search%5Bprogramme%5D=&result_search%5Bproject%5D=&result_search%5Bsort%5D=Relevancy&result_search%5BsortDirection%5D=Descending&result_search%5Blist__909%5D%5B%5D=all&result_search%5Blistcheckbox__909%5D=909

https://portal.effra.eu/results/list/result?result_search%5Btaxon%5D=&result_search%5BtextQuery%5D=&result_search%5Btype%5D%5B%5D=Demonstrators&result_search%5Bproject%5D=&result_search%5Bsort%5D=Relevancy&result_search%5BsortDirection%5D=Descending&result_search%5Blist__1045%5D%5B%5D=all&result_search%5Blistcheckbox__1045%5D=1045&result_search%5Blist__909%5D%5B%5D=all&result_search%5Blistcheckbox__909%5D=909&result_search%5Bpublishable%5D=1



The ZDMP Innovation Action11 has uploaded ten cases for the Smart Autonomous Factory pathway (Figure 10).

Figure 10 ZDMP cases in Autonomous Smart Factories pathway

During the process of mapping these ten use cases, the maturity levels to which they belong were selected according to the technologies, methods and results obtained. For example, the pilot ‘Engine block manufacturing: Defects reduction by the optimization of the machining process’12 has achieved third (Connected IT and OT) and fourth (Off-line optimisation) maturity levels. In this case the machines of ETXE industrial partner are installed on the FORD plant as a part of cylinder block production line. The raw process data are collected about the equipment state, as well as about the production process itself (e.g. type of cylinder blocks being produced at the moment). The operation of machining equipment installed in production line can be optimised and improved through analysis of these data to detect anomalies/abnormalities and to develop a mitigation plan to return to the efficient production state. In some cases, the optimisation process can be done by the platform. However, some cases might require involvement of engineers from equipment manufacturer to analyse data and provide recommendations on optimisation. The effectiveness of recommendations can be further assessed by the platform.

See information on the other ZDMP use cases mapped on the Autonomous Smart Factories here.

The QU4LITY Innovation Action has uploaded eight cases for the Smart Autonomous Factory pathway (Figure 11).

11 https://www.zdmp.eu/ 12 https://portal.effra.eu/result/show/4500

https://portal.effra.eu/results/list/result?result_search%5Btaxon%5D=&result_search%5BtextQuery%5D=&result_search%5Btype%5D%5B%5D=Demonstrators&result_search%5Bproject%5D=1866&result_search%5Bsort%5D=Relevancy&result_search%5BsortDirection%5D=Descending&result_search%5Blistcheckbox__1045%5D=1045&result_search%5Blist__909%5D%5B%5D=911&result_search%5Blistcheckbox__909%5D=909

https://www.zdmp.eu/




Figure 11 QU4LITY cases in Autonomous Smart Factories pathway

As part of the mapping process of pilots, the technologies used have determined the maturity level positioning. For example, the pilot ‘PRIMA Additive Manufacturing Pilot Adaptive Control Technology’13 has achieve the fifth (Realtime optimisation) maturity level. The scope of this use case is to enhance process monitoring and control for producing metal components and make Additive process more productive and robust. This aim is reached with the new modular monitoring and control system approach that can be used with many different sensors and adaptable process models. Real-time process and machine signals are analysed by machine-learning algorithms to find structures and pattern related to the required key quality indicators (critical defects per track, distortion, keeping of dimensions). The system is also connected to a higher-level factory data interface which allows to exchange process information and reassign the production strategy based on additional factory conditions. This allows to: have equipment condition reporting, reduce reject rate by application of data-driven process model that has been derived by AI algorithms, increase OEE by recommending process adjustments to the operator or directly change the parameters in real time; so, to reduce also the operator costs.

See information on the other QU4LITY use cases mapped on the Autonomous Smart Factories here.

13 https://portal.effra.eu/result/show/3922





1.3.2 DMP Projects’ Cases in Hyperconnected Factory Like the Smart Autonomous pathway, the Hyperconnected pathway was developed in CF1 and it is still active and used to position different projects and cases. On this pathway, one of the EC Horizon2020 DT-ICT-07 DMP projects (ZDMP) has positioned some of its pilots.

The ZDMP Innovation Action has uploaded nine cases for the Hyperconnected Factory pathway (Figure 13), which are mainly involved in the development of dedicated IT connection to some supply chain partners, which matches with the fourth maturity level (Figure 12).

Figure 12 ZDMP maturity level (Hyperconnected)



Figure 13 ZDMP cases in Hyperconnected pathway

One of the most representative cases is the ‘Construction supply chain: quality control at construction site and quality traceability’14 pilot, which involves 3 industrial partners among the construction supply chain. The supply delays sometimes can significantly affect the working schedule causing the productivity loss. To avoid delays, it is important that all the parties have an early access to information about potential delays and about the quality of supplies to act quickly and, if needed, to reschedule activities. The material suppliers provide the product quality data, as well as data about material types and amount to be delivered. Thus, it is possible for construction to react in agile way and reschedule in the case of unexpected accidents.

See information on the other ZDMP use cases mapped on the Hyperconnected Factories here.

1.3.3 DMP Projects’ Cases in Product-Service Factory The Collaborative Product-Service pathway is the last of the CF1’s pathways and it is also one of the most used in this mapping process.

The ZDMP Innovation Action has uploaded six cases for the Product Service Factory pathway (Figure 14).






Figure 14 ZDMP cases in Collaborative Product-Service pathway

As part of the mapping process of pilots, the technologies used have determined the maturity level positioning. For example, the pilot ‘Moulds manufacturing: Process alert system for machine tool failure prevention and Smart process parameter tuning’15 has achieved the fifth (Realtime optimisation) maturity level (Figure 15). The goal of this case is to collect the equipment and machining process data to detect the abnormalities in the production process and inform the operator, while adjusting the parameters of the production process to achieve the optimal quality results. The quality assessment of the products is performed by the FORM industrial partner that utilizes equipment delivered by FIDIA. Degradation in the equipment performance can cause significant quality drop. In order to avoid this, ZDMP platform aggregates the process and equipment data and performs analysis to identify the defects occurring, as well as to detect possible equipment degradation. The ZDMP platform acquires the process and equipment data from FORM and provides a set of innovative services, namely: (i) parameter adjustment, (ii) operator alerting in case of defect, (iii) correlation among various parameters related to equipment and parameters selected. In the case, when equipment manufacturers need to be involved in the problem solving, data can be securely shared through ZDMP platform.





Figure 15 ZDMP maturity level (Collaborative-Product service)

See information on the other ZDMP use cases mapped on the Collaborative Product-Service pathway here.

The QU4LITY Innovation Action has uploaded two cases for the Product Service Factory pathway (Figure 16).




Figure 16 QU4LITY cases in Collaborative Product-Service pathway

The most representative case from QU4LITY on the collaborative Product-Service pathway is the pilot called ‘GHI Real-time cognitive hot stamping furnace 4.0’16. This case has achieved the fourth maturity level (Figure 17) of the pathway (Product-Service Innovation). It is building a ZDM scenario based on the development of a smart and connected hot stamping process with the ability to correlate the furnace operation parameters with the quality control of the stamped parts, extending in this way the product lifecycle control loop, making the operator more involved in the process. This is achieved thanks to the newly developed platform, which integrates the information provided through the product quality control process. In this use case, GHI and Innovalia share data in a trusted manner through an IDS connector17, so that GHI is able to find correlation between the quality of the parts and the furnace operation parameters. By the time this sharing data solution has been devise just for the use case, but also a business model in here can emerge as GHI is getting more restrictions with customers that do not want to give the governance of their process data.

16 https://portal.effra.eu/result/show/3873 17 https://industrial-data-space.github.io/trusted-connector-documentation/


https://industrial-data-space.github.io/trusted-connector-documentation/



Figure 17 QU4LITY maturity level (Collaborative-Product service)

See information on the other QU4LITY use cases mapped on the Collaborative Product-Service pathway here.

1.3.4 DMP Projects’ Cases in Circular Economy Pathway The Circular Economy pathway is one of the newest pathways in CF2. The EFPF Innovation Action18 has uploaded two cases for the Circular Economy pathway (Figure 18), according to their publicly available information.

Figure 18 EFPF cases in Circular economy pathway

18 https://www.efpf.org/


https://www.efpf.org/



At the time of submission of this document, the Circular Economy pathway is still in its validation process. For the moment only the EFPF Innovation Action has used this pathway to map their cases. The two pilots mapped in this pathway have reached the second maturity level (Industrial Circular Economy Piloting), see Figure 19. One of these two cases is called ‘Wastes Tracking Decentralized Application for Circular Economy’19 and involves multiple European companies from the manufacturing ecosystem, where the lack of capability to track and trace assets and transactions throughout the entire supply chain is one of the critical challenges that the companies face towards the circular economy. This lack of traceability hinders companies from quickly adapting, planning, and managing their assets effectively and optimised. A solution is needed to help companies track and trace their assets at every supply chain stage and increase supply chain visibility without significantly increasing operational costs.

Figure 19 EFPF maturity level (Circular economy)

In particular, for EFPF project supports a circular scenario, from wastes to energy. The wastes producer sells its wastes to a waste management company which pre-processes and sells them again to a bioenergy company that produces energy. A part of this energy flows back to the company that had produced the wastes as an outcome of its production processes. A track and trace mechanism for the exchanged wastes over this circular closed loop is needed.

EFPF provides a Blockchain-based mobile app available to the stakeholders of the transportation (i.e. drivers etc.) to enable a secure and trusted handshake between them. Furthermore, a web application based on blockchain enables adding permissions to drivers, vehicles etc. that participate in a transportation process, monitor the various stages of the transported wastes and issues a digitally signed document related to wastes transportation and handling. Both mobile and web apps share a common blockchain backend in logging


Linearity (1°) Industrial CE Piloting (2°)

Systemic Material Management (3°)

Circular Economy Thinking (4°) Circularity (5°)

1 2 3 4 5

Products Any toxic substances should not be used to create products

Material used should be the minimum amount required to respect product functionalities and design (resource sufficiency)

Systematical identification of possibilities to reuse, refurbish and remanufacture

Ecodesign of products (must be easy to be disassembled, repaired, remanufactured and its components should be recyclable)

Changing business model towards product-service-systems and X as service approaches at ecosystem level

Process Quality monitoring to avoidunnecessary scraps

Production processes must require limited amount of energy

Transportations modes (reverse logistics), internal recycling of materials

Building industrial synergies/closed loop models

Circular systems and process at value network/ecosystemlevel

Platform Information technologies to gather processes data

Company systems integration (e.g. ERP, MRP, PDM, PLM…)

Disassembly and Remanufacturing enabling technologies introduced on the shop-floor

Digital platform integration enabling the interaction among value chain actors

Collaborative business processes and workflows are used over the product life cycle

People Ad-hoc engagement of individuals, not comprehensive engagement

Engagement and awareness raising, systemic empowering through champions

Cultural transformation and qualified people (skills)

Circular suppliers selection and value network level indicators

Sustainable government requirements and European Green Deal?

Partnership Contractual obligations based onregulations

Code of conduct guidelines for circularity/sustainability of materials. There areoccasionalcollaboration around circularity/sustainability

Circularity objectives discussed at value chain level. Stakeholder engagement systematically considered.

Capabilities to dynamic collaboration new partners ,research and innovation partnerships launched

Digital ecosystems enabling circularity, partnering with customers making use of knowledge

Performance Reporting and measuring onlythe legal responsibilities related to recycling and otherenvironmental obligations

Targets and KPIs set up for circularity pilots

Operative KPIs for resourcesufficiency/ decrease of waste etc.

LCA analysis over the productslifecycle, KPIs for foot print.

Circularityindicators(basedon the forthcoming standard)




transportation information in the same blockchain instance. Smart Contracts are used as a back-end part of the two DApps (web and mobile). Identity Smart Contracts, Supply Chain Smart Contracts, Logging Smart Contracts and Notification Smart Contracts are being used. The combination of the two apps (mobile and web) provides a complete solution of wastes’ track and trace in a circular closed-loop and promotes full visibility of the processes.

More information on the other EFPF cases mapped on the Circular Economy pathway is available here.

1.3.5 DMP Projects’ Cases in Data Space Pathway The QU4LITY Innovation Action has uploaded one case for the Data Space pathway (Figure 20).

Figure 20 QU4LITY cases in Data Spaces pathway

The pilot called ‘WHR Dryer Factory Holistic Quality Platform’20 is mapped according to the Data Spaces pathway. It is wanted to reach a holistic approach to ZDM considering the full product lifecycle: from Product Design to Customer Service, cycling back to Product Design; by digitalising the factory. This pilot leverages the outcomes of a previous research project (NMBP FP7 GRACE) and integrate the QU4LITY digital enablers and platforms (through APIs) and the AQ control loops.






It has reached the maximum maturity level among the Data Spaces pathway (Data Valorisation), see Figure 21, where the main innovation is represented by the introduction in production of MPFQ model fused with AQ control loops: Functional Integration and Correlation between Material, Quality, Process and Appliance Functions. This innovative way to control quality and model data inherent to quality is the fundamental approach that will lead to the vision of holistic Quality system. In addition, it will be deployed AQ reference implementations to address unresolved problems in the vertical integration of data management (from data gathering to visualization and decision-making), enabling a holistic vision to be achieved.

Figure 21 QU4LITY maturity level (Data Spaces)

At the time of submission of this deliverable only one case from QU4LITY was uploaded, but it is able to check future updates on QU4LITY cases mapped on the Data Spaces pathway here.

1.4 Collection of lessons learned from cases in EFFRA Innovation Portal

As already said, this Catalogue not only contains information on DT-ICT-07 projects. Other initiatives at national/regional level were involved on this collection of information work, which have a lot of valuable information regarding lessons learned and best practices.

The EFFRA Innovation Portal support a really interesting section (see Figure 22) where projects can describe their know-how and share publicly not only lessons learned during their experiences, but also their significant innovation achievements or the significance of the results for SMEs.




Figure 22 Lessons learned filter in EFFRA Portal

Several national/regional cases have reported their lessons learned, such as those collected by IMR or UNOTT colleagues.

For example, the case ‘Manufacturing Service Buses for Data Interoperability’21 developed by FA3D uses a manufacturing systems bus concept to enable a manufacturing system to be easily reconfigurable from a logical control perspective, and to integrate multi-vendor equipment and software into a single source of data which is used throughout. As part of the use case, they have collected and shared several lessons learned about their data format and structure through the use of B2MML and the OPC UA intermediate protocol used to communicate all components of the system directly with the service bus.

This collection of lessons learned and innovation achievements is part of an ongoing work that has just started during this first CF2 period and will be fully accomplished by the following CF2 activities. The next steps in relation to this work will be directly related to the DT-ICT-07 projects. There is a clear intention to enter and enlarge all the information gathered in the 19 May workshop (in which the DT-ICT-07 projects participated) in the EFFRA Innovation Portal, in order to give full visibility to the projects' results.

1.5 Ongoing works

This collection of cases and lessons learned, as well as the mapping process, is still work in progress. At the moment, all EC Horizon 2020 DT-ICT-07 2018/2019 are engaged in these tasks. Future work will complete the catalogue of these projects and add more national/regional cases. It will allow validation of newer pathways and improve the information collected on these topics.


Figure 23 ‘Manufacturing Service Buses for Data Interoperability’ example

https://portal.effra.eu/results/list/result?result_search%5Btaxon%5D=&result_search%5BtextQuery%5D=imr&result_search%5Bproject%5D=&result_search%5Bsort%5D=Relevancy&result_search%5BsortDirection%5D=Descending&result_search%5Blist__1028%5D%5B%5D=all&result_search%5Blistcheckbox__1028%5D=1028

https://portal.effra.eu/results/list/result?result_search%5Btaxon%5D=&result_search%5BtextQuery%5D=nottingham&result_search%5Bproject%5D=&result_search%5Bsort%5D=Relevancy&result_search%5BsortDirection%5D=Descending&result_search%5Blist__1028%5D%5B%5D=all&result_search%5Blistcheckbox__1028%5D=1028




2 Foresight and Recommendations towards Digital Europe programme With the objective of collecting recommendations and lessons learnt to be addressed in the coming Digital Europe Programme (DEP) and Horizon Europe Programme (HEP), CF2 project organized an interactive workshop “Shape the future of Digital Manufacturing Platforms in Horizon and Digital Europe” on 19th of May, 2021. Six topics related to manufacturing innovation were identified (three for each programme) and, focusing one by one, attendees were invited to analyse them from a present and future perspective.

• Firstly, participants were asked to provide evidence of how the topic can be addressed, by telling their experiences, discussing challenges and barriers that were faced in specific projects, the solutions adopted for the implementation.

• To complement it, the discussion was projected toward future activities, by considering un-solved challenges to be addressed shortly, recommendations for next projects and applications, to provide food for thought in preparation to next European calls.

Attendees were the representatives of the ecosystem of digital manufacturing platforms associated to ConnectedFactories2. More than 15 projects participated: AI REGIO as organizer of the workshop, the 7 DT-ICT-07 cluster members (CF2, DIGIPRIME, EPFP, KYLOS 4.0, QU4LITY, SHOP4CF, ZDMP) and other projects specialized in smart and digital manufacturing (MARKET4.0, MAS4AI, ESMERA, TRINITY, MANUHUB, CSAI, SHERLOCK, DIH2 , FIAB).

The choice of organizing a session where participants could interact one to each other was not driven by the attempt of creating an event to build consortia but to stimulate reflections and ideas, in preparation to HEP and DEP calls proposals. The main goal was to stimulate a wider discussion, where the others’ ideas and suggestions could be of interest and of examples for everyone. Actually, the workshop had a high level of participation and interaction: a total of 50 attendees (from the digital manufacturing ecosystem associated to ConnectedFactories) left almost one hundred comments, by telling past experiences and recommendations for the future.

Due to covid-19 pandemic restriction, the workshop was organized online, with the support of MURAL online board, where participant could easily drag and drop sticky notes and write comments (Figure 24)22.

22https://app.mural.co/t/polimi3712/m/polimi3712/1620211267161/284521480c93a05c74ffbb31a10bea1d4385550d?sender=silviarazzetti4266

https://app.mural.co/t/polimi3712/m/polimi3712/1620211267161/284521480c93a05c74ffbb31a10bea1d4385550d?sender=silviarazzetti4266

https://app.mural.co/t/polimi3712/m/polimi3712/1620211267161/284521480c93a05c74ffbb31a10bea1d4385550d?sender=silviarazzetti4266



Figure 24 The workshop interactive board

In the context of the Digital Europe Programme23 (at that time under development) the three topics that were selected were:

• Data Space in manufacturing, to discuss their application both at single Smart Factories level and at Network level

• AI Testing and Experimentation Facility (TEF) in manufacturing, to outline required services and activities that can help manufacturing to benefit from a TEF during their digital journey

• European DIHs (EDIH) in manufacturing, to identify differences and similarities among manufacturing and other DIHs, to identify next steps toward European DIHs programme.

In next paragraph, main results from the workshop are presented in detail.

2.1 Data Spaces in Manufacturing

A Data Space is not simply a collection of data and information, but could better be defined as a techno-business artefact formed by:

• High Value Datasets, structured according to understandable data models, that encourage and facilitate the use by third parties, providing, for instance, set of metadata and being organized by ontologies. “High Value Dataset” doesn’t mean simply “large amount of data”, but it refers to the fact that information contained inside can be easily accessed and exploited and that they are structured in order to get all the information to generate business.

• A trustworthy Industrial Data Platform, that is, a set of architecture, components, tools and repository to use and exchange data, that allows to access information, to control data usage and

23 https://digital-strategy.ec.europa.eu/en/activities/digital-programme

https://digital-strategy.ec.europa.eu/en/activities/digital-programme



presenting an easy integrability. Different ways and solutions to implement it are available in the market, from proprietary to Open-Source ones.

• Business and Governance Rules, defined for each level of the data value chain. The goal is to overcome the traditional “piece of paper” that typically store the agreement among parties, but to leverage on a more mature technology in order to create an automated system able to manage business and governance rules.

In last years, the European Commission (EC) is boosting the creation and exploitation of Data Spaces, as an effective way toward European Digitalization, with consequent increase of productivity and competitiveness in the international scenario. In 2018, driven by the awareness that in next years, the need of data will considerably grow, the EC published a specific guideline to enforce sharing of private sector data, opening the discussion about the always increasing need to share information among different stakeholders in a reliable way, avoiding risks of loss or misuse.

Besides Open Data (free data, with very few restrictions) and Data Marketplace (where data are sold and monetized), the third way to share data is the exchange in a closed platform, where quality and reliability are guaranteed. To achieve the latter option, the EC is annually publishing working documents to pave the path toward the creation of Common Data Spaces, stressing the importance of having pan-European infrastructure to access data. Four main pillars were identified, for implementing a successful strategy:

• Governance framework, to regulate Data Spaces and to define governance rules. Recently the EC published the Data Act, containing all legal aspects related to data access and use;

• Enablers, that is, the Data Platforms, in which EC is investing billions of euros; • Competences, that is, digital skills required to manage Data Spaces. The digitization, the data

exploitation and the adoption of innovative AI-based technology solutions require new competencies and new professional figures among the workforces. Companies (SMEs in particular, with a reduced number of employees) often lack adequate skills and need resource to re-train/acquire new workers;

• Rollout of common European Data Spaces, in several crucial economic sectors and with the participation of European and regional stakeholders.



Figure 25 European Strategy for Data outlined by the Digital Europe Programme

A common European data space is nothing more than a Data Space shared among European stakeholders, with common purposes and interests, and of course it presents the same key elements described above.

• High Value Dataset, in the sense of FAIR (Findable, Assessible, Interoperable and Reusable). Data will be made available of a voluntary basis and can be reused against remuneration or for free, depending on the data holder’s decision.

• An infrastructure to use and exchange data, managed by a secure IT environment for processing of data

• Appropriate Governance Models, that is, a set of rules of legislative, administrative and contractual nature that determine the rights of access for processing data

The European Commission objective is ambitious since a successful Data Space exists only after a trust relationship among different actors has been established. The final goal is the creation of nine different Data Spaces, dealing with nine different sectors: Manufacturing, Green, Mobility, Health, Finance, Energy, Agriculture, Public Administration and Skills.



Figure 26 The nine key sectors for European Data Spaces

To create a strong network, scalable and interoperable, that attracts stakeholders and meets standard requirements, the DEP call (2nd call) recommends following steps:

1. To collaborate with other sectoral projects, in order to develop a Data Space by connecting different infrastructures deployed

2. To collaborate with Data Space Support Centres to identify common standard and architectures 3. To federate deployed Data infrastructure with other European facilities 4. To make use of existing services offered by the Common Services Platform

The call foresees two main application scenarios:

• a Value Chain Data Space interoperating non-hierarchical networks of suppliers sharing their capability and capacity data and implementing Manufacturing as a Service industrial cases. The supply chain management is continuously monitored, and status data are exchanged across the value chain

• a Smart Factories Data Space interoperating distributed industrial assets aimed at the development of advanced solutions like predictive/prescriptive maintenance, by continuously monitoring and exchanging industrial data



Figure 27 Schema for Smart Factory and Value Chain Data Space

2.1.1 Data Spaces in Manufacturing; Interactive Workshop The 50 participants were asked to identify use cases that address(ed) main features related to the implementation of data spaces (considering standard, data models, interoperability, trust, people engagement), to tell their experience and to identify issues and un-solved challenges.

During the session in the limited time people had to reply, 28 sticky notes were collected.

Figure 28 The "Data Space in manufacturing" DEP session



In AI REGIO project24, all activities related to implementation of a Data Space are addressed, starting from the identification of common ontologies and data models, with a specific focus on RAMI Asset Administration Shell (AAS) solution. The project started in October 2020 so the implementation expected to be achieved in some of the 17+16 experiments is still in a preliminary stage, where requirements and specifications are gathered. Currently, the main activity is focused of definition of tools, methodology and development of software component.

In EFPF project25, it has been implemented a data space (already in use by multiple partners) leveraging on IDS connector, with the objective of sharing IoT data (sensors outputs, for instance) among supply chain partners. The goal was not only defining standards, but also a reliable way of sharing data. According to their experience, to become a common standard shared and spread in Europe for managing Data Sovereignty, IDS should overcome some barriers:

• Lack of awareness that sharing data is possible if the right standardized and trustworthy solutions are adopted. Most of the use cases that they come across leverage on proprietary systems and often this represents a bottle neck to implement shared data spaces;

• Limited availability of examples and success stories from IDSA side, to show the results that can derive from adoption of their technologies;

• Difficulties in mapping available ontologies and data models into the data space to be implemented.

In Market4.0 project26, several use cases are focused on the implementation of data spaces concepts and framework, leveraging on IDS connector and Smart Connected Supply Network, for data exchange along the supply chain. Experiences are collected both from project partners activities and Open Calls participants. Main obstacle that typically an SME faces while adopting IDS connector is lack of resources and expertise to do it: technology providers in the project must often find a balance between IDS connectors complexity and traditional approaches.

As of now, Shop4CF27 activities didn’t aim at implementing a data space and defining data models to share data among different actors, since currently, it is not clear yet to what extend data will be shared, or used only for data visualization. Actually, the first barrier to overcome is to make people confident about existing solutions for data sharing. Here, use cases were more focused on the definition of protocols to data exchange between sensors and backend platform, by adopting main standards solutions (all different components, indeed, are based on FIWARE broker28). While discussing about protocols, participants raised their interest about ontologies and expressed their wish to increase knowledge and competences about ontologies definition.

24 https://www.airegio-project.eu/ 25 https://www.efpf.org/ 26 http://market40.eu/ 27 https://shop4cf.eu/ 28 https://fiware-orion.readthedocs.io/en/master/

https://www.airegio-project.eu/

https://www.efpf.org/

http://market40.eu/

https://shop4cf.eu/

https://fiware-orion.readthedocs.io/en/master/



ZDMP29 is working on a number of use cases to implement data spaces, but the one dealing with business interoperability, in particular, deserves to be highlighted. The goal is to develop a “mapping solution” to merge data coming from several sources and that are structured according to different models. The adoption of common ontologies (not necessarily data models) can facilitate the purpose.

To summarize, all the main aspects concerning the end-to-end implementation of a data space have been addresses and discussed:

• Standard, ontologies and interoperability to build the foundation of a data space, designed to be used by several stakeholders;

• Trustworthy data sharing and privacy management, to create a scenario where actors can benefit from it without encountering risks;

• Partner’s engagement to create a strong network of actors interested in sharing data. For instance, it came out that, talking about IDS connectors, it is difficult to take onboard new partners and newcomers, due to the apparent complexity of the solution.

2.2 AI TEF in Manufacturing

A Testing and Experimentation Facilities (TEF) is a technology infrastructure (possibly accessible also remotely) where it is possible to test mature solutions in an environment as close as possible as the real world, before deploying it in production. We talk about AI TEF in case the facility offers services related to Artificial Intelligence, providing physical and software resources to be exploited by customers who want to fully enhance their data potential.

Beside the technological provision and validation, where AI TEFs put at disposal technical facilities, skills and expertise, a Testing and Experimentation Facility is typically in charge also of supporting the regulatory framework definition, that plays a fundamental role in the creation of a trustworthy AI solution, and socio-economic aspects.

In January 2020, the European Commission organised a series of workshops to further refine the concept of Testing and Experimentation Facilities for Artificial Intelligence (AI TEFs) with the help of experts and national delegations. In the upcoming Digital Europe Programme 2021-2022, the European Commission will launch calls to create these TEFs. The objective is to come up with a network of “World-class TEF”

• Supporting the testing of mature technology, that is, playing an active role in last stages of the solution implementation;

• Limited in number: the plan is to cover all the main sectors (Manufacturing, Health, Agri-Food, Smart Cities and Communities) but with a small number of high qualified facilities;

• Available for all European stakeholders, in the context of AI-based solutions validation; • Providing both technical, socio-economic and legal support; • Able to disseminate results throughout Europe (also thanks to DIH network channel)

29 https://it.zdmp.eu/

https://it.zdmp.eu/



Figure 29 AI TEF main concepts

Reasons that are driven the European Commission to invest in such facilities are mainly related to the increasing need of effective data exploitations, to make Europe competitive in the international scenario. To achieve the goal, several steps have been identified:

• To improve hardware and software performance and efficiency, also making them energetically sustainable, to meet also Green Deal and Twin Transition requirements;

• To increase the amount of data available, also thanks to the implementation of Data Spaces; • To adopt standards and certification; • To deploy explainable AI solutions.

AI TEFs have been conceived exactly to support SMEs toward this goal, by facilitating and supporting the adoption of AI digital solutions.

As already mentioned, there is a strong relationship between TEFs and DIHs, since the latter can support the former to disseminate results, thanks to its widespread presence over regional countries and leveraging on the trust relationship established with customers. Moreover, to create a distribution channel for AI to reach also local companies and users, the European Commission strategy leverages on three main pillars:

• Testing and Experimentation Facilities, as a physical infrastructure to test and validate Data Space for AI solutions, receiving feedback and support by experts. Due to change of mindset, deriving also from covid-19 pandemics, next step is the provision of remote access to the infrastructure, for instance supported by Digital Twin, monitoring platforms and ICT as a service.

• AI-on-demand Platform, conceived as a reference central toolbox that can be easily accessed, offering high quality services, such as AI algorithms and solutions, data to train models, computing infrastructures. An example is represented by AI4EU, a EU project, aiming at deploying a platform for an on-demand AI ecosystem for Europe.

• AI DIHs, that is, Digital Innovation Hubs with a specific attitude toward Artificial Intelligence. They are less technological-centric with respect to a TEF, but typically offer a wider Portfolio of Services. In AI REGIO, for instance, it was adopted the D BEST taxonomy to describe services (it means that



they are clustered in five classes, Data, Business, Ecosystem, Skills and Technology) and, to develop new offers more AI oriented, DIH-DIH collaboration models have been studied, modelled and implemented by the AI DIH Network initiative. Indeed, the strength of a DIH derives also by its capability of creating collaboration and ecosystem, to be of support for instance to AI TEFs.

Figure 30 TEF-centric Collaboration

2.2.1 AI TEF in Manufacturing; Interactive Workshop The objective of the panel is to discuss about how a (European) AI TEF can bring benefit to the manufacturing ecosystem, with the objective of identifying the required skills and services that a Testing and Experimentation Facility should provide. This topic can be easily related with the other panel, since for instance, a TEF is expected to have competences to support in the implementation of Data Spaces, AI and Collaborative Robots. But here the focus is on the TEF itself and participants were asked to tell how their projects supported TEFs and to think about “nice to have” services and features that an Experimentation Facility should put in place in short term. What came out is that the Manufacturing AI TEFs of the future should be equipped by services to be remotely accessed and should be able to replicate real conditions as in production lines.

Overall, 16 sticky notes were collected.



Figure 31 The "AI TEF in manufacturing" DEP session

AI REGIO, whose core is working with DIHs, is trying to create synergies, exploiting what there is in common between DIHs and TEFs. In a dedicate work-package, AI REGIO is collecting requirements from AI TEFs, with the objective of enhancing the physical assets, but also of reorganizing the services provided according to the common framework used for DIH (D BEST Service Portfolio), with a specific focus on “Remote services” (DR BEST Service Portfolio). The main challenges to be addressed regard interoperability, to:

• establish a relationship between i-Space and TEF. The latter are often not equipped by adequate computing infrastructures and the gap can be filled by exploiting i-Space assets (specialized in data elaboration and High-Performance Computing) but, of course, it requires interoperability between the two systems;

• spread of computational load among embedded, edge and cloud infrastructures to optimize the computing process.

The RAMI AAS30 represents a scalable, customisable and effective solution to guarantee interoperability among TEFs.

The need of introducing in Experimentation Facilities services that can be assessed remotely is a driver also for ZDMP project, who is actively working to address it, especially after the covid-19 pandemics.

To promote robotic technologies for SMEs, both ESMERA31 and TRINITY32 projects are engaging a number of Competence Centres (CCs) that can provide an easily accessible environment for developing, evaluating, testing and demonstrating new solutions. In particular, LMS CC provides access to equipped infrastructure open call winners to test their solutions.

30 https://ec.europa.eu/futurium/en/system/files/ged/a2-schweichhart-reference_architectural_model_industrie_4.0_rami_4.0.pdf 31 http://www.esmera-project.eu/welcome/ 32 https://trinityrobotics.eu/

https://ec.europa.eu/futurium/en/system/files/ged/a2-schweichhart-reference_architectural_model_industrie_4.0_rami_4.0.pdf

https://ec.europa.eu/futurium/en/system/files/ged/a2-schweichhart-reference_architectural_model_industrie_4.0_rami_4.0.pdf

http://www.esmera-project.eu/welcome/

https://trinityrobotics.eu/



The main objective of QU4LITY project is to produce digital enablers and machine enhancement to reduce the number of defects in manufacturing and it counts on several AI TEF partners, where it is possible to test innovative solutions before presenting them to pilots. Again, the difficulties faced to physically access the facilities during covid-19 pandemics make urgent the development of remote services.

2.3 European Digital Innovation Hubs in Manufacturing

Digital technologies such as High-Performance Computing, Internet of Things, Big Data, blockchain, robotics and artificial intelligence allow businesses to produce higher value products and to improve production processes. However, European companies are not fully exploiting the potential that a digital approach may offer. This slow uptake of digital technologies has an impact on the competitiveness of European Union, that risks remaining a step behind with respect to other world powers and to affect its global role, unable to compete in the economic grow.

Therefore, the European Commission is boosting the creation of European Digital Innovation Hubs (EDIHs)33, as one-stop shops to help companies to dynamically respond to the digital challenges and with a coordination role towards other innovation hubs.

A European Digital Innovation Hub - similarly to a DIH - is a single organisation or a coordinated group of organisations with complementary expertise, with a not-for-profit objective that support companies – especially SMEs and mid-caps – and/or the public sector in their digital transformation.

It is conceived as an access point to latest digital capacities including high performance computing (HPC), artificial intelligence, cybersecurity, as well as other existing innovative technologies such as Key Enabling Technologies, available also in fablabs or citylabs. Hubs are embedded in a local economy and have as an objective to strengthen it by supporting the digital transformation of the local industry and public sector. Considering, for instance, the manufacturing sector, the hub is expecting to support companies in the adoption of Industry 4.0 and circular economy methods.

An EDIH Service Portfolio typically includes services such as:

• Test before Invest, to help enterprises to do the right choice of the solution to be implemented both doing consultancy activities (awareness raising, digital maturity assessment, demonstration activities) and putting at disposal infrastructure and competencies to perform tests in an environment that doesn’t present same costs and risks of the production one (adaptation and customisation of various technologies, testing and experimentation with digital software and hardware technologies). Of course, a special focus will be put on the key technologies promoted in Digital Europe Programme: HPC, AI, and Cybersecurity;

• Skills and training, including advertising, hosting or providing of training, boot-camps, traineeships, as well as supporting the implementation of the short-term advanced digital skills training courses and job placements developed as part of the DEP Advanced Digital Skills pillar;

• Support to find investments, including access to financial institutions and investors, supporting the use of InvestEU and other relevant financing mechanisms, in close co-operation with the foreseen InvestEU Advisory Hub4 and the Enterprise Europe Network (EEN);

• Innovation ecosystem and networking, establishing relationship between companies and other enterprises belonging to the same value chain, between companies and technology providers,

33 https://digital-strategy.ec.europa.eu/en/activities/edihs

https://digital-strategy.ec.europa.eu/en/activities/edihs



between public administrations and GovTech companies to promote co-creation, supporting international matchmaking.

Figure 32 The four main classes of services offered by an EDIH

The Commission proposed to invest between 0.5 and 1M€ per year in each EDIH. Together with the contribution of the related Member State, this would add up to a significant investment between 1 and 2 M€ per year per EDIH. With the current budget proposal for Digital Europe, the goal is to support between 100 – 200 hubs in the EU, with at least one hub per Member State.

In order to identify the bunch of DIHs designated to become EDIHs, the European Commission has started a specific selection process. The first step consists of a call for expression of interest, where each Member State delivers a list of designated hubs, in order to apply a preliminary filter on existing DIH that would like to compete to become an EDIH. The eligible proposals will be evaluated by the EC and all those above a predefined threshold will be ranked. Finally, the selected proposals will get a grant from the European Commission and the initial payment (pre-financing) will be disbursed. As of now, 323 candidates from 25 countries (24 Member States plus Norway) were selected, taking into account their different technological and sectoral specialisation, aiming at creating a heterogeneous network of hubs able to cover several innovation aspects. DIHs selected to become EDIH must provide expertise in at least one of the following subjects: Artificial Intelligence, High Performance Computer and Cybersecurity.

Table 1 Specialization of the 323 EDIHs candidates

AI HPC Cybersecurity

211

95

137

The DIHs profiles are available in the platform34 developed by the JRC, where Digital Innovation Hubs are catalogued and it is possible to browse candidates, filtering by country, technologies and services provided, sectors….

34 https://s3platform.jrc.ec.europa.eu/digital-innovation-hubs-tool

https://s3platform.jrc.ec.europa.eu/digital-innovation-hubs-tool



Figure 33 The JRC platform for EDIH catalogue

Since EDIH are financed with public funds, the European Commission is creating a monitoring system based on a set of KPIs in order to effectively measure the performance of each hub, with the following indicators:

• Number of businesses and public sector entities which took advantages of the EDIH’s services. Where relevant, this will include a description of the main reason of the contact.

• Number of additional investments and total investment amount successfully triggered • Number of collaborations foreseen with other EDIHs and stakeholders outside the region at EU level,

and description of infrastructures jointly shared / joint investments with other EDIH.

A set of additional impact indicators will be collected and analysed with the support of the Digital Transformation Accelerator35, such as the impact on digital maturity that the EDIH had on supported organisations, taking into account aspects as: the motivation of workers to relate to Artificial Intelligence, the environmental impact and the sustainability of the solution implemented, the evaluation of the business model around the product.

The vast majority of DIHs currently present in the JRC Catalogue expressed their interest in supporting the full adoption of Digital Technologies in Manufacturing. The DIH4INDUSTRY36 portal aims at becoming a single

35 https://www.eitdigital.eu/accelerator/ 36 https://dih4industry.eu/welcome/

https://www.eitdigital.eu/accelerator/

https://dih4industry.eu/welcome/



point of access for DIHs and Policy Makers to know the coverage and mapping of DIH services in the Manufacturing Industry domain.

The role of DIHs as mediators between SMEs (willing of a digital transformation) and technology providers has recently been as fundamental and this is why the European Commission is investing lot of money and effort to support them and to increase their visibility.

2.3.1 EDIH in Manufacturing; Interactive Workshop Participants to the workshop in CF2 were asked to reflect about the commonalities and specificities of Manufacturing-oriented DIHs with respect to other DIHs and how will they be developed and materialised in the EDIH program, taking into account also the role of the DT Accelerator to promote and coordinate an EDIH Vertical on manufacturing.


Figure 34 The "European DIHs in manufacturing" DEP session

AI REGIO project is part of I4MS Phase 4 cluster, that focuses on business definition for service suppliers (as Innovation Actions, Digital Innovation Hubs and Competence Centers) and so it has a strong propensity toward DIHs. AI REGIO, namely, includes a cluster of 13 DIHs and, to support the exploitation of their activities, it has developed a standard methodology (METHODIH), to describe, according to a common taxonomy, services provided and customer journeys and pipelines. Currently it is working to include more services in the standard Portfolio provided to DIHs (as Legal services for instance) and to enhance cross-regional and cross-domain collaboration among Hubs.



In QU4LITY the goal is the creation of virtualized DIHs, providing Zero Defect manufacturing solutions and services, building them up on top of its “market platform”, where different services are presented and offered.

ZDMP is supporting DIHs enhancement by implementing a platform based on a set of components that contribute to SMEs digitalisations, to be deployed in the Hubs. The next step is to equip DIHs of Zero-Defect solutions to be offered as services.

About the need of boosting the establishment of DIHs able to support the manufacturing industry, besides AI REGIO project, also TRINITY and DIH237 are working on it. The latter is setting up Hubs to provide services as IoT and robotics, but also able to drive manufacturing SMEs to access the GAIA-X38 and IDSA communities.

At least in the manufacturing domain, it is fundamental for DIHs to play the role of technology scouting, in order to drive customers towards the most suitable technological solution for their purposes.

37 http://dih-squared.eu/content/we-accelerate-factories-through-robotics 38 https://www.gaia-x.eu/

http://dih-squared.eu/content/we-accelerate-factories-through-robotics

https://www.gaia-x.eu/



3 Foresight and Recommendations towards Horizon Europe programme As already mentioned in Section 2, the workshop “Shape the future of Digital Manufacturing Platforms in Horizon and Digital Europe” on 19th of May, 2021 focused also on Horizon Europe Programme. Three topics were selected, to be analysed in details:

• Human-Robot Interaction and Industry 5.0, to discuss the implementation of a Human-Machine coexistence (and collaboration) that could bring advantages not only in terms of digital but also Green Transition;

• Zero Defect, Zero Waste, Green and Resilient Manufacturing, to discuss about defects preventions and waste reduction (obtained with advanced technologies) that can drive toward a Green Manufacturing;

• Artificial Intelligence for Autonomous, Hyperconnected and Product-Service Factories, to identify how Artificial Intelligence can boost sustainability, resiliency and exploitation of product-service lifecycle in manufacturing.

The objective of the workshop was twofold: on the one side participants were involved to share their personal experience with other projects’ representatives in order to provide a picture of how that specific challenge was addressed; on the other, they were asked to share recommendation about opportunities and expectations for the future.

In next paragraph, main results from the Horizon Europe session are presented in detail.

3.1 Human-Robot Interaction and Industry 5.0

The INDUSTRY 5.0 challenge implies the harmonisation of human-centric manufacturing with the green-digital twin transition. This is even more true in these times of recovery and resilience, where manufacturing industry needs an even deeper transformation of its technologies, organisation and workforce. In particular, such a topic implies

• To support fast response to repurposing changes in production requirements with a new role of humans;

• To demonstrate seamless social collaboration in teams of human workers; • To create a network of open-access pilots to experiment new technologies and to enable data and

knowledge sharing through the European industrial ecosystems.

Participants were asked to reflect about projects (if any) where they were involved to realize at least an embryonic approach toward I5.0, to highlight difficulties, lessons learnt and best experiences, to describe which goals were achieved and to identify next steps.

The discussion dealt only partially with competences required to develop a system where human and machine collaborate together, but it was widely focused on its management/maintenance and on the role and needs of the worker in that scenario. Beside the technology itself that is required to build an “intelligent” robot, there are other side (but key) aspects to be taken into account while talking about Industry5.0. For instance,

• the worker’s ability to collaborate with a machine and its propensity to accept robot limits, while evaluating benefits;



• the need of reallocating (and training) workforce towards new tasks; • privacy of (human) data gathered by machine, to avoid misuse and improper handling.


Figure 35 The "Human-Robot interaction & Industry5.0" HEP session

AI REGIO together with KITT4SME39 are the two Innovation Actions in I4MS (phase 4) addressing AI innovations, to be exploited also in the scenario of human-machine interaction. In particular, some AI REGIO’s pilots are moving towards the concept of Industry5.0, by developing collaborative intelligence solutions, that is, an evolution of the simple “interaction” where the machine learns from the human and both actively support each other. Considering what happened during covid-19 pandemic and of the increasing needs of resilient attitudes, collaborative intelligence seems to be a great challenge to be boosted in next years.

TRINITY project is developing a set of tools and methods to reach the goal of Agile Production in Manufacturing, specifically focused on the adoption of advanced robotics. Cybersecurity is identified as one of the main issues to be addressed, not only in Robotics but also dealing with AI, Data Spaces, DIH competences. Actually, Cybersecurity is itself a challenge that European Commission wishes to direct in next years and, indeed, is inside Pillar II – Cluster 3 of Horizon Europe Programme.

SHERLOCK project40 deals with Human-Robot collaboration, exploring solutions to collaboratively assembly heavy parts. The project is working to implement solutions able to guarantee privacy of human data, but lack of regulatory framework is an obstacle to achieve it. Of course, having at disposal specific guide lines outlined

39 https://kitt4sme.eu/ 40 https://www.sherlock-project.eu/home

https://kitt4sme.eu/

https://www.sherlock-project.eu/home



by a central regulator could speed up and facilitate the task, since it is not easy to identify a priori all kind of intrusion and data attacks that could be carried out.

In Shop4CF, a specific work-package is in charge of defining an evaluation framework of worker acceptance in a scenario of human-machine collaboration, with the goal to find the right balance where both the workforce and the company can benefit from robots adoption. To reach a full exploitation of collaboration, it’s important to teach workers how the can take advantage from machines and how their job may improve, guaranteeing privacy of the data caught by the robot to track, in every moment, the exact location and activity of the worker.

3.2 Zero Defect, Zero Waste, Green and Resilient Manufacturing

This challenge aims at applying the latest manufacturing and digital technologies for the benefit of a Green Transition of the Manufacturing Industry, especially for what regards Waste production and management. In particular, advanced manufacturing technologies are required to

• demonstrate a significant increase of sustainable production through improved control systems and non-destructive inspection methods;

• develop methodologies and tools to prevent the generation of defects at component level and its propagation to the system level;

• create new diagnostic methods for in-situ monitoring of industrial production; • ensure efficient use of materials, repair strategies, and reduced production cost and time.

One key point addressed to participants is to reflect about the economic impact that the transition toward a green solution can have for an enterprise (mainly, if small or medium). The European Commission is pushing toward the 2050 climate neutrality, but of course it represents a cost for companies that should be balanced with alternative revenues. In order to achieve this goal, manufacturing industry should completely change its mentality in next years: participants were asked to reflect about the effort required to do it. Moreover, the Twin Transition programme is boosting also the adoption of new digital solutions and it is fundamental to understand how to combine both the green and the digital aspects, to avoid mutual exclusion.




Figure 36 The "Zero Waste, Zero Defect, Green & Resilient manufacturing" HEP session

In DIGIPRIME41, the circular aspect of the automotive sector is enhanced, by the adoption of AI for optimization of compounding parameters in machines. The goal is to develop circular economy for the recycle of plastic, metals components, mechatronics, batteries, … Due to the lack of data in the circular value chain, it is not straightforward to combine green transition with the digital one (as the Twin Transition European goal asks), in particular in the plastic domain where recycle activities are performed in different “siloed” facilities. Actually, stakeholders are reluctant to share own data, mainly because they don’t feel the need of knowing what happened in previous stages depending probably on limited awareness of benefits coming from data exploitation.

Similarly to DIGIPRIME, MAS4AI project42 is developing innovative applications concerning AI and zero defect solutions for optimization of compounding parameters in machines.

QU4LITY and ZDMP are the two cluster projects involved the most in the topic of zero defect product and process in manufacturing by digitization. ZDMP is developing solutions to reduce the number of defects in five different sectors (machine tool, automotive, electronic, construction and agri-food) and the objective for the future is to be able to combine it with green solutions and circular economy. QU4LITY is working on the definition of a Reference Architecture and digital enablers to achieve Zero Defect Manufacturing. Indeed, reduction of defects (and consequently also the decrease of wastes) can be reached only transforming the entire process line, by adding digital devices, sensors, cloud/edge computers, but avoiding the increase of pollution and carbon emissions.

41 https://www.digiprime.eu/ 42 https://mas4ai.eu/

https://www.digiprime.eu/

https://mas4ai.eu/



On one side, it is required the support of government and external funds and incentives to concretely start the transformation process; on the other, Research and Innovation projects should train people basing on success stories, translating experiences and adapting them to the customer scenario. The role of the projects cluster shouldn’t be only to manage marketplaces where to sell assets and solutions, but also to “sell” experiences and success stories, showing KPIs and measurable results. Green options will become appealing and convenient for companies only after a complete change of mindset, for instance when, by adopting low-impact solutions or implementing circular economy, there will be advantages also in term of visibility and reputation.

3.3 Artificial Intelligence for Autonomous, Hyperconnected and Product-Service Factories

This topic focuses on applying and adopting Artificial Intelligence to level-4 autonomous factories, to highly dynamic and heterogeneous value networks and to the entire lifecycle of products and services from design to remanufacturing and including all the aspects primarily relevant for industrial production.

In particular, proposals are required to

• establish European industry as leader in sustainable manufacturing through the application of AI technologies, by improving the environmental, economic and social sustainability of industrial production smart factories;

• improve the agility of European industry and its resiliency to external and internal influences in the whole value chain, supporting new as a Service business models;

• integrate state-of-the-art AI technologies with advanced manufacturing and re-manufacturing technologies and systems, exploiting their potential across the entire product and service lifecycle, including the interaction with users and consumers.

Participants were asked to present their projects and experiences measuring the impact that the adoption of AI in manufacturing had (or has) in terms of sustainability and environmental footprint. Moreover, the topic was selected to open a wide discussion dealing with all different steps required to implement AI solutions, from data acquisition (standards and interoperability) to user interface (explainability and trust), from implementation of AI models to measurements of their impact.

Since Artificial Intelligence for Autonomous manufacturing will be the subject for the third pathway of Connected Factories 2 (to be developed in June, 2021) and so, there will be lot of chance to deal with it in details, the topic was left at the end of the workshop. However, it raised the interest of participants, as AI is a very present component in the majority of projects involved.




Figure 37 The "AI technologies for manufacturing" HEP session

DIGIPRIME platform will contain several AI solutions: for instance, it is developing an intelligent tool to support decision making about material recycle and, as already mentioned in the second panel, also an optimization system for compounding parameters.

As the name suggests, the core of MAS4AI project is Artificial Intelligence: the goal of the solution is to develop an AI-based multi-agent system to address the interaction among different agents, gathered from six use cases. Main implementations are a tooling selection agent, a matching parameters optimization agent and a scheduling optimization agent, each based on different technologies and models according to requirements. The architecture behind has just been defined leveraging on AAS concept.

AI REGIO‘s main purpose is to create regional collaboration between DIHs for an AI-driven transformation, hence, also here Artificial Intelligence represents (together with DIHs) the core of the project. Currently, as a preliminary task before starting to concretely implement AI solutions in the experiments, two assessment activities are run: firstly, the definition and validation of an AI maturity model to measure the capability of SMEs to accept AI-driven technologies; secondly, the definition of an interoperability framework with AI4EU, which is developing a platform to create an ecosystem of AI stakeholders, resources and innovations.

Artificial Intelligence is not the main challenge addressed in ZDMP, but defect detection during the production process is often AI-based, for instance in case of predictive analysis. The final goal would be to reach full automation in quality check process, through extensive usage of AI.



4 Conclusions and Future Outlook D3.3 has a twofold objective: on the one side to identify foresights and Recommendations for DMP of the future through the analysis of CF1 and CF2 pathways; on the other side, DMP industrial cases have been projected into the future of next EC Digital Europe and Horizon Europe programs

It is worth to highlight that all the projects belonging to the first wave of the DT-ICT-07/2018 have been engaged in this information collecting activity and have mapped their use cases according the ConnectedFactories pathways. As for the second wave of projects DT-ICT-07/2019, it is valuable to note that each of them is now aware of the Catalogue and the dissemination opportunities it offers. As it is a continuous work, different fine-tuning actions will be carried out, since the Catalogue and its information are continuously growing and updated along the development of the projects. The work done by all different involved tasks of ConnectedFactories 2 will support the dissemination of information on innovation projects and encourage SMEs to access success stories and best practices.

Regarding the projection of DMP industrial cases toward Digital Europe Programme, all the three selected topics (Data Space, AI TEFs and DIHs for manufacturing) raised the interest of the attendees, but of course with a different level of participation. For instance, the first one, that deals with a subject that presents large applicability and it is common in several projects (since it is the foundation of a number of other implementations), worked as ice-breaker and allowed to engage several partners in the discussion. On the other side, not all projects involve AI TEFs and DIHs, so, last two topics were mainly driven by AI REGIO, QU4LITY, DIH2, TRINITY, ESMERA, whose main business is focused on such supporting facilities.

Regarding Horizon Europe Programme, more time was dedicated to discuss about the first two selected topics (Human machine collaboration and Zero Defect and Zero Waste manufacturing), while the third one (AI for Autonomous manufacturing) will be deeply analysed in following months, with the development of the third CF2 pathway (“AI for Autonomous manufacturing pathway”, part of WP2 activities). Also in this session, projects whose main core business is exactly one of the three main challenges presented in the panels drove the discussion. For instance, SHERLOCK (focused on “robotic application”), QU4LITY and ZDMP (focused on “Zero Defect manufacturing”) and AI REGIO and MAS4AI (focused on “artificial intelligence”) provided concrete examples of implementation. It was a good starting point to make all participants aware of the issues faced during the implementation and to share reliable recommendations and suggestions for the future. However, the discussion involved also partners who partially deal with such topic but wanted to tell experiences and are interested in collecting more useful information.

It’s worth to mention that there are some barriers that have been identified, which are commonly spread among all the six topics analysed and affect several projects and use cases. First, lack of resources and expertise in companies who want to approach the digital transformation process (mainly in small and medium enterprises, whose workforce is typically reduced in number). Second, what is required is to improve the quality and availability of data, that or are not always systematically collected or are collected without taking into account the business that can derive from it. The aim is to pursue a complete change of mindset in lot of companies, scared of adopting new solutions (Open Source in particular) and to share data among third parties.



WP3 activities will now proceed with this analysis by more and more involving the DT-ICT-07 second wave projects (digiPRIME, KYKLOS4.043 and Shop4CF) in the EFFRA Innovation Portal and in extracting knowledge about foresights and recommendations along the new pathways that WP2 will analyse such as the AI for Manufacturing pathway. Similarly, with the advancement of DEP and HEP programs, we will also map the new cases with the new topics and priorities of the 2021-22 work programme through new interactive workshops and personalised interviews.

At M34, the D3.5 will be finally issued with the final considerations about Foresights and Recommendations.

43 https://kyklos40project.eu/

https://kyklos40project.eu/

Foresights and Recommendations of Digital Manufacturing ...

Documents

Transcript of Foresights and Recommendations of Digital Manufacturing ...