EMMC International Workshop 2017 April 5-7, 2017 - Vienna / … · 2018-08-10 · EMMC...

Venue

Hotel Courtyard Vienna Prater/Messe

Trabrennstrasse 4, 1020 Vienna

Austria

EEMMMMCC IInntteerrnnaattiioonnaall WWoorrkksshhoopp 22001177

AApprriill 55--77,, 22001177 -- VViieennnnaa // AAuussttrriiaa

EMMCInternationalWorkshop2017

2

Sponsors

© EMMC 2017

Responsible for content: Nadja Adamovic, Ernst‐Dieter Janotka

TU Wien Institute of Sensor and Actuator Systems

Gusshausstrasse 27‐29, 1040 Vienna, Austria

This event has received funding via the EMMC‐CSA project from the European Union‘s Horizon 2020 research and innovation programme

under Grant Agreement No 723867


3

TThhee EEuurrooppeeaann MMaatteerriiaallss MMooddeelllliinngg CCoouunncciill

iiss oorrggaanniissiinngg tthhee

EEMMMMCC

IInntteerrnnaattiioonnaall WWoorrkksshhoopp 22001177

AApprriill 55--77,, 22001177

VViieennnnaa // AAuussttrriiaa

Venue

Hotel Courtyard Vienna Prater/Messe

Trabrennstrasse 4, 1020 Vienna

Austria


4

Workshoporganisation Nadja Adamovic ([email protected])

Ernst D. Janotka (ernst‐[email protected])

TU Wien Institute of Sensor and Actuator Systems Gusshausstrasse 27‐29/ E366 1040 Vienna Austria

ScientificCommittee

Nadja Adamovic, EMMC Coordinator, TU Wien, Austria

Pietro Asinari, Politecnico di Torino, Italy

Gerhard Goldbeck, Goldbeck Consulting, United Kingdom

Adham Hashibon, IWM Fraunhofer, Germany

Kersti Hermansson, Uppsala University, Sweden

Denka Hristova‐Bogaerds, Dutch Polymer Institute, The Netherlands

Tom Verbrugge, DOW Benelux, The Netherlands

Erich Wimmer, Materials Design, France

Rudolf Koopmans, Koopmans Consulting, Switzerland

Georg J. Schmitz, Access Technology, Germany


5

Introduction WED,April5,2017 13:00‐13:30 Room1

Talk 1 Nadja Adamovic (TU Wien, AT)

The European Materials Modelling Council

Talk 2 Anne de Baas (European Commission, BE)

The NMBP programme past, present and future

PlenaryTalks WED,April5,2017 13:30‐14:30 Room1

Plenary Talk 1 Peter Littlewood (Argonne National Laboratory, US)

Materials modelling as a driver for industrial innovation ‐ breakthroughs, barriers, and bottlenecks

Plenary Talk 2 Emanuele Ghedini (University of Bologna, IT)

MODA, a common ground for MOdeling DAta generalization: introduction, use case and possible improvements

THU,April6,2017

09:00‐10:00 Room1

Plenary Talk 3 Surya R. Kalidindi (Georgia Institute of Technology, US)

A new data and informatics enabled ecosystem for materials innovation (IDEAS:MD3)

Plenary Talk 4 Welchy Leite Cavalcanti (Fraunhofer IFAM, DE)

A case study of translation practice for coating development: from modelling to the market

FRI,April7,2017

09:00‐10:00 Room1

Plenary Talk 5

Tanja Graf (Volkswagen, DE)

Challenges and opportunities for industrial materials modelling

Plenary Talk 6 Hein Koelman (DOW Benelux, NL)

The role and impact of materials modelling in accelerating innovation in an industrial context


6

Session1 WED,April5,2017 14:30‐16:15 Room1

Barriersorbreakthroughs?Towardsbettermodelsinthechemicalandsemiconductorelectronicsindustrialsectors

Chair Kersti Hermansson (Uppsala University, SE)

Impulse 1 Kurt Kremer (Max Planck Institute for Polymer Research, DE)

Challenges for Simulations of Soft Materials: Structure Process Property Relations

Impulse 2 Kurt Stokbro (QuantumWise, DK)

Towards better models in the semiconductor electronics industrial sector

Impulse 3 Ignacio Pagonabarraga Mora (CECAM, EPFL Lausanne, CH)

E‐CAM, chemical (and other) industries, and model development

Session presenter at Podium Discussion Georg Kresse (Universität Wien, AT)

Introduction

The overall focus of Session 1 deals with this issue, namely Model development and model

gaps in the context of (especially) industrial end‐users and their needs. Our underlying

objectives, are

(i) to find ways to identify the gaps between the models that are needed by industrial

stakeholders, and the models that are in fact available, and

(ii) to produce a roadmap for how to remedy these gaps.

We want to identify which are the most urgent fields where model development is needed and

what type of model developments that are needed.

Discussion points and questions

The following questions summarize the issues for this session. They primarily, but not

exclusively, refer to the industrial sectors targeted in this session (see title). If you want to

bring forward challenges in other sectors, or if your comments are general, please make the

context clear.

• In your opinion, which are the largest barriers hindering the wider use of modelling

in industry? ‒ Hopefully, some of your answers will relate to the lack of adequate

computational models and the need to develop improved ones, as this is the main

topic of Session 1.

• Thus: which applications/materials/phenomena are the most interesting and urgent

for todays (and tomorrow's) model developments to target, i.e. in what directions

should the model development efforts go to meet the needs of industry?


7

• In your experience, are industrial stakeholders mostly interested in modelling results

expressed as absolute values or as trends, i.e. would they best benefit from

quantitative or descriptive answers?

• Do you foresee applications where discrete models (i.e. electronic, atomistic,

mesoscopic) will become more crucial in the (near) future? (Why is this so?)

• Who should do the needed model development (academic researchers?, industrial

people?). If academic researchers are expected to engage in targeted model

development for industrial needs, how shall it best be organized and financed? Or

differently put: How can we turn the different perspectives and interests of industry

and academia into a win‐win situation?

• Do you discern any important differences between the two sectors in this session in

terms of how they ought to approach the issue of model developments and model

gaps?


Vocabularyandtaxonomyforimprovedcommunityintegration,communicationandinteroperability

Chair Gerhard Goldbeck (Goldbeck Consulting, UK)

Impulse 1 Anne de Baas (European Commission, BE)

Making sense of materials modelling

Impulse 2 Geoff Gross (Osthus, US)

Establishing a common semantics across sub‐disciplines: Lessons from Allotrope and other projects

Session presenter at Podium Discussion Emanuele Ghedini (University of Bologna, IT)

Introduction

Due to the complexity of materials and the wide range of applications, the materials modelling

field consists of a number of communities. With the focus on a specific application domain

(along with specific models and domain specific software codes) each community has evolved

a domain related terminology. However, applications to industrial problems in advanced

materials and nanotechnology require a strong interdisciplinary approach between these fields

and communities. There is hence a need to establish a common terminology (definition of

concepts and vocabulary) in materials modelling which will lead to greatly simplified and much

more efficient communication. A standardised terminology could improve future exchanges

among experts in the entire area of materials modelling and also facilitate the exchange with

industrial end‐users and experimentalists and lowering the barrier to utilising materials


8

modelling. The common language is hoped to foster dialogue and mutual understanding

between industrial end‐users, software developers, scientists and theoreticians.


What is the experience and lessons‐learnt from other communities in establishing a

shared vocabulary, classification and metadata?

How can the benefit to industry be maximised? I.e. how can common terminology and

MODA be used to lower the barriers to benefiting from materials modelling in

industry?

What are the expectations from stakeholders (Modellers, Translators, Software

Owners, Manufacturing industry)?

What improvements are needed/recommended for the terminology and MODA?

How can the common terminology and metadata support publications about materials

modelling? A metadata schema would help to communicate, disseminate, store,

retrieve and mine data about materials modelling. Should/could CEN/CWA become a

guidelines for use of terminology in publications? Should/could MODA be introduced

as Supplementary Information in publications?

Should there be a MODA repository?

How can classification and MODA be built upon to drive forward interoperability?


Translationinpractice

Chair Pietro Asinari (Politecnico di Torino, IT)

Impulse 1 Chris Eberl (Fraunhofer IWM, DE)

Connecting cutting edge fundamental research to applications – current and future ways for translation

Session presenter at Podium Discussion Pietro Asinari (Politecnico di Torino, IT)

Introduction

There is a need to advance an intermediate translation and transfer process of innovation from

the academy to the manufacturers. To this end, the development of an intermediate

framework able to improve the materials modeling uptake for the European industry must be

stimulated. Therefore, the so called translation network, is thought and expected to act as a

bridge between the academic and the industrial sector. Some of the characteristics of this

emerging framework ‐ such as, the existence of a common workflow for the translation and

the contest from which it should arise (e.g. academy, centers for applied research and/or

industrial R&D departments) ‐ are still unclear and will be discussed in this session.


9


Discuss the possible scenario for translation workflow as general as possible, divided in

steps, starting with that proposed by the impulse speaker.

o Who is the translation activity mainly planned for? (large companies, SMEs,

etc)

o Who is the translator? Should/could he/she be independent?

o Is it possible to define “universal” steps in the translation workflow?

o Which are the most common steps of the translation activity?

o In your experience, which are the best practices for making translation

workflow successful in practice?

o Which are the most relevant features of the small medium enterprises (SME)

with regards to the translation process?

List of ideas about how to overcome barriers

o In your experience, which are the bottlenecks that may prevent translation

workflow in being successful?

o Which tools are missing for overcoming such barriers?

o Which supporting actions, shared documents and dissemination activities are

missing for overcoming such barriers?

o In your experience, a European recognition of the role of translation would be

useful in overcoming such barriers?

List of ideas about how to promote translation further

o Which official tools (talks, seminars, program‐funding, projects) are most

promising for promoting translation further?

o What should be the synergy between European and national actions for

promoting translation further?

o Could we provide a list of SME to master students for collaborations and,

eventually, for employment in the field of translation?

o What do you expect from the EMMC coordination and supporting action (CSA)

for promoting translation further?


Barriersandbreakthroughs:Bridgingthecommunicationandexpectationgapsbetweenmodeldevelopersandindustrialstakeholders


Impulse 1 Erich Wimmer (Materials Design, FR)

Origins of the gaps between academic model developers and industrial stakeholders and ways to build bridges


10

Impulse 2 David Cebon (Granta Design, UK)

What do companies need from material modelling? And how to find out!

Session presenter at Podium Discussion Tom Verbrugge (DOW Benelux, NL)

Introduction

Sessions 1, 4, and 7 all deal with Model development and model gaps in the context of

industrial end‐users and their needs. However, while Sessions 1 and 7 concern identification of

the model gaps and how to remedy them, Session 4 concerns how to extract this information

from the industrial stakeholders themselves. It focusses on finding the best procedures how to

reach out to the industrial community and create efficient and sustainable ways to exchange

information about model needs. The perspective in Session 4 is different, with focus on the

communication model rather than on the physics model.

Thus, in brief, the goal of this session is to identify the best way for the EMMC to learn about

the industrial stakeholder's own view of what improved models are needed to enhance

modelling in industry, or at least in selected key sectors. Software owners and translators will

most likely be considered important actors in this process by the workshop participants.


In your opinion: what could be the most useful approach to promoting the use of

materials modelling in industry (SMEs, large companies, companies with some

modelling connection, newcomers, ...?). Good examples? Making more reliable and

versatile models? Other?

Do you have some suggestions on how to efficiently involve the software owners

(without compromising their commercial interests) so that the industrial users'

_frustration with bad models and needs for better models_ reach the (academic?)

model developers? And vice versa, so that information about the skills and capabilities

of model developers reach the industrial stakeholders?

Same question as the previous, but now concerning the so‐called translators.

How can we turn the different perspectives and interests of industry and academia

into a win‐win situation? For example – how can one share the burden of developing

more reliable, realistic and powerful models? How do we jointly tackle the need for

models to become more reliable and the fact that it is a very time‐consuming process

to generate good models?


11


OntologiesandMetadataschemaandtheirimplementations

Chair Adham Hashibon (Fraunhofer IWM, DE)

Impulse 1 Colin Batchelor (Royal Society of Chemistry, UK)

How to do things with ontologies?

Impulse 2 Jesper Friis (SINTEF, NO)

Proposal for basic metadata schema to support interoperability

Impulse 3 Claudio Cacciari (CINECA / EUDAT, IT)

EUDAT: pan‐European Data Infrastructure to support metadata management and foster interoperability

Session presenter at Podium Discussion Colin Batchelor (Royal Society of Chemistry, UK)

Introduction

All stakeholders of materials modelling face barriers regarding access to, and use of,

information about materials modelling, utilisation of the wide range of modelling tools and

methods, and last but not least interoperability of models and codes.

Metadata can be defined as “data describing the context, content and structure of records and

their management through time”. They provide information that allows for categorization,

classification and structuring of data. A well‐established example is the Crystallographic

Information File (CIF) which provides metadata for atomistic structures and properties.

A metadata schema can be defined as “a logical plan showing the relationships between

metadata elements, normally through establishing rules for the use and management of

metadata specifically as regards the semantics, the syntax and the optionality (obligation level)

of values.”

An ontology is a formal naming and definition of the types, properties, and interrelationships

of the entities that really or fundamentally exist for a particular domain. Ontologies aim to

define which entities, provided with their associated semantics, are necessary for knowledge

representation in a given context. Ontologies and related information technology provide an

opportunity to share a common understanding of the structure of information within a specific

domain, the possibility to reuse domain knowledge, to make domain assumptions explicit and

to analyse domain knowledge.


What are the experiences and lessons learnt from other communities in establishing,

using and maintaining a metadata schema and ontology?

What co‐operation can be installed to come to a common approach to metadata

schema and ontologies for materials modelling?


12

How can the benefit to industry be maximised?

What are the expectations from stakeholders (Modellers, Translators, Software

Owners, or Manufacturing industry)?

How can metadata schema and ontologies be introduced and used in repositories?

What is the best path towards an “Open Simulation Platform”, a reference design that

supports wide interoperability?


GettingSMEsmoreinvolvedinmaterialsmodelling

Chair Chris Eberl (Fraunhofer IWM)

Co‐Chair Denka Hristova‐Bogaerds (Dutch Polymer Institute, NL)

Impulse 1 Marcus Kremers (Airborne International, NL),

Modelling from the perspective of a SME

Impulse 2 Paul Racec (Deutsches Zentrum für Luft‐ und Raumfahrt, DE)

Experience of SMEs doing material modelling within EUREKA/Eurostars: Alternative funding instruments beyond H2020

Impulse 3 Manuel Laspalas (ITTAINNOVA, SP)

Translation for SME: challenges and opportunities

Session presenter at Podium Discussion Manuel Laspalas (ITTAINNOVA, SP)

Introduction SMEs often have insufficient or even lack of expertise in modelling that prevents them from

integrating modelling and simulation software into their industrial development workflows.

Furthermore, many SMEs are not fully aware of the opportunities which modelling can bring to

their business. There is a need to illustrate to SMEs in convincing way the possibilities of

modelling to solve industrial challenges and to identify ways and supporting instruments of

how to facilitate the adoption by SMEs of modelling as a tool.

This session will discuss the specific requirements of SMEs, identify typical barriers preventing

SMEs to adopt materials modelling as a standard tool in their design processes, discuss the

best practices and show cases for successful implementation of modelling by SMEs,

understand how materials modelling may effectively support SMEs and finally outline a

strategy for wider use of modelling by SMEs.


13

Discussion points and questions What prevents SMEs for wider use of modelling?

How to increase awareness of SMEs in modelling capabilities?

Identify the type of SMEs for which modelling would be most useful: by applications

sector, material type, product type, product manufacturing and design, other?

Which type of models are most needed?

How can the translators facilitate the use of modelling by SMEs? Training?

What do the SMEs need to convince them in applying modelling as a tool:

demonstrator, show cases, economic impact of modelling, business case, other?

How can modelling become a practise at SMEs? Obligatory request for modelling data

from material suppliers or end users, other?

Collaborative EU/national initiatives for SMEs?

Session7 THU,April6,2017 10:15‐12:00 Room1

Barriersorbreakthroughs?Towardsbettermodelsinthemetallurgicalandautomotiveindustrialsectors

Chair David Cebon (Granta Design, UK)

Impulse 1 Christos Argyrakis (Rolls‐Royce, UK)

Towards better models in the metallurgical and aviation/automotive industrial sectors

Impulse 2 W. A. Curtin (EPFL, CH)

Towards better models in the automotive industrial sector

Session presenter at Podium Discussion Tanja Graf (Volkswagen, DE)

Introduction

The overall focus of session 7 deals with this issue, namely Model development and model

gaps in the context of (especially) industrial end‐users and their needs. Our underlying

objectives, are

(i) to find ways to identify the gaps between the models that are needed by industrial

stakeholders, and the models that are in fact available, and

(ii) to produce a roadmap for how to remedy these gaps. We want to identify which

are the most urgent fields where model development is needed and what type of

model developments that are needed.

In session 7 we will focus on the automotive and metallurgical industrial sectors, but frequent

references to other areas, similarities, differences and synergies are expected.


14


The following questions summarize the issues for this session. They primarily, but not

exclusively, refer to the industrial sectors targeted in this session (see title). If you want to

bring forward challenges in other sectors, or if your comments are general, please make the

context clear.

• In your opinion, which are the largest barriers hindering the wider use of modelling

in industry? ‒ Hopefully, some of your answers will relate to the lack of adequate

computational models and the need to develop improved ones, as this is the main

topic of Session 7.

• Thus: which applications/materials/phenomena are the most interesting and urgent

for todays (and tomorrow's) model developments to target, i.e. in what directions

should the model development efforts go to meet the needs of industry?

• In your experience, are industrial stakeholders mostly interested in modelling results

expressed as absolute values or as trends, i.e. would they best benefit from

quantitative or descriptive answers?

• Do you foresee applications where discrete models (i.e. electronic, atomistic,

mesoscopic) will become more crucial in the (near) future? (Why is this so?)

• Who should do the needed model development (academic researchers?, industrial

people?). If academic researchers are expected to engage in targeted model

development for industrial needs, how shall it best be organized and financed? Or

differently put: How can we turn the different perspectives and interests of industry

and academia into a win‐win situation?

• Do you discern any important differences between the two sectors in this session in

terms of how they ought to approach the issue of model developments and model

gaps?


Pragmaticapproachestointeroperability:Implementations/realisations/scenariosofpracticalrelevance

Chair Zachary T. Trautt (NIST, US)

Impulse 1 Georg J. Schmitz (Access Technology, DE)

Practical interoperability ‐the view of a software provider

Impulse 2 Joerg Neugebauer (Max‐Planck‐Institut für Eisenforschung, DE)

How to achieve interoperability ‐ a modeller’s perspective

Session presenter at Podium Discussion Borek Patzak (CTU, CZ)


15

Introduction

The issue of interoperability refers to “the ability of computer systems or software to exchange

and make use of information”. In materials modelling there is often the need to link models in

order to address typical industrial problems. Efficient integration and exchange of data is also

becoming more and important in the light of high throughput computation, enabled by

modern hardware.

Interoperability can be achieved by different means and on different levels, from syntactic

(data format) definitions that enable ‘import/export’ via integrated platforms that operate on

the basis of a certain ‘data model’ to semantic level interoperability. The long‐term goal of

achieving cross‐domain interoperability based on a semantic framework (see also Session 5)

will take some time to achieve. There are, however, a number of pragmatic approaches to

interoperability under way. These may be based on limited metadata sets, on particular file

formats such as HDF5 and/or limited to certain domains.


How do you deal with integration and interoperability at the moment?

What current integration and interoperability implementations do you use?

Which integration and interoperability environments are you familiar with?

What are the strengths and weaknesses of current implementations?

What are your top three requests for future interoperability developments?

What would be the practical next steps in addressing integration and interoperability?

What promising developments are you aware of?


Translationinpractice:Whatwecanlearnfrombigcompanies?

Chair Tom Verbrugge (DOW Benelux, NL)

Impulse 1 Jan Stolk (DSM, NL)

Translation in a large company: From business opportunity to science and back

Impulse 2 Ali Karimi (Continental, DE) From materials to performance tyres using computational modelling

Session presenter at Podium Discussion Katya Vladislavleva (DataStories International, BE)

Introduction

Translators are key players at the interface between manufacturers on the one hand and

software owners and modellers on the other hand. Translators support the usage of materials

modelling in industrial R&D to the same level as experiments are used today. The main topics

handled by the Translators are a significant reduction of the language gap between


16

manufactures and modellers, highlighting the role of key performance indicators (KPI) on top

of materials modelling and the integration of business oriented measures directly into the R&D

process.

The role of Translators can be and is performed by different stakeholders, including R&D staff

in large enterprises, application scientists in software companies, scientists in research

institutions as well as individual consultants.

In some cases large companies, with their broad internal expertise, perform and implement

translation process internally. In other cases the involvement of external translators is needed.

In both scenarios the translation practices of the large companies can be used to outline the

requirements for Translators and can be used as show cases for transition methodologies.


What does a company need as internal translation process? Expertise of the internal

translators?

Benefits and drawback of translation inside a company and the use of external

translators?

Can a translation process inside a company be adopted by other companies?

What are the requirements of large companies for external translators?


Translation&traininginpractice:Whatwecanlearnfromsoftwareowners?

Chair Erich Wimmer (Materials Design, FR)

Impulse 1 Johannes Fraaije (CULGI, NL)

Experience in translating and training with the CULGI software

Session presenter at Podium Discussion Erich Wimmer (Materials Design, FR)

Introduction

The adoption of materials modelling by industrial organizations is driven by the economic

benefits resulting from this technology. To this end, translation and training play a key role.

Here, translation can be understood as the process of analysing an industrial problem,

formulating a modelling approach, performing or guiding the modelling and simulation tasks,

extracting the relevant information from the simulations, and translating this information back

into specific recommendations of economic value to the end‐user.


17

Software owners who want to see their software being used by industry are thus motivated to

provide training and to stimulate translation.


In your experience, what are the most successful training events?

What lessons were learned from these training sessions?

What should be avoided?

Where is the „bottleneck“ in transfer of knowledge?

How do the users of your software achieve successful translation, i.e. adding economic

value for their organizations?

How much effort can end‐users expect to master the software and realize full capability

of the modeling?

Who can play the role of the neutral advisor?

What are your best success stories?

What were cases that went wrong and why?

What is currently missing in training activities for SME’s?

What are the differences in training and translation resources between commercial big

and small software companies?


Repositoriesofmaterials(modelling)dataandknowledge:designanduse

Chair David Cebon (Granta Design, UK)

Impulse 1 Zachary T. Trautt (NIST, US)

Making Materials Data Discoverable, Accessible, Interoperable and Reusable: Efforts at NIST and Beyond

Impulse 2 Saulius Grazulis (Vilnius University / COD, LT) Bridging experimental and computed data: experiences in standardising and implementing a community API

Session presenter at Podium Discussion David Cebon (Granta Design, UK)

Introduction

A significant amount of information and knowledge is being produced in labs all around Europe

and the world, but most of it remains confined within the silos of each community, or even

worse, within the realms of each individual research group or lab. Repositories of materials

models, software tools, materials relations (such as interatomic potentials or constitutive


18

equations), knowledge and parameter data (also from experiments) is a promising avenue to

enable more efficient use of materials modelling and accelerated utilisation by industry.

Repositories can also provide promising routes for the traceability and validation of results of

materials modelling. As industry increasingly looks to the simulation of material properties

across all length scales and models, the demand for validated data, both in terms of input to

models and as results will only increase and will bring with it an increasing need for

interoperability of data, meta‐data and interfaces allowing better access to the curated data.

Recent actions, such as EuDAT and OpenAir in Europe aim to catalogue every possible

repository, yet it remains difficult to find those that are related to materials modelling and

how to exploit the data stored in them.


What exists in access and retrieval mechanisms and in interoperability mechanisms?

What are the technology barriers for interoperability? For example: Interoperable

Interfaces/APIs; Metadata schema (combining experimental characterisation and

modelling data); other technology issues.

What are the key stakeholder barriers that remain to be addressed? For example:

Examples of best practice for data and information sharing; Expertise requirements

(Information and Communication Technologies, materials, physics, data management,

use of tools…); Concerns regarding proper provision for IP and privacy issues;

Mechanisms for promoting participation and collaboration.

How can we promote interoperable Interfaces/APIs to stimulate the proliferation of

interoperable databases?

How to better stimulate participation in the catalogue of all the databases that exists?

What are the benefits of data and information sharing to each stakeholder (Material

modellers, experimentalists, manufacturers, software owners, translators, etc.)

How do we overcome societal barriers and concerns regarding sharing (for free or

against payment) of data?

What measures need to be taken to guarantee traceability of data, intellectual property

rights protection and quality of data?


SuccessstoriesandhowtowidensuccessthroughoutEuropeanindustry

Chair Rudy Koopmans (Koopmans Consulting, CH)

Impulse 1 Rudy Koopmans (Koopmans Consulting, CH) SME’s can benefit from Materials Modelling too!

Impulse 2 Massimo Noro (Unilever, UK) Harnessing the power of predictive computation


19

Session presenter at Podium Discussion Rudy Koopmans (Koopmans Consulting, CH)

Introduction

The European Science & Technology community dealing with Materials Modelling is

considered to be globally leading and spearheading the advancements in this field. Still the

perception exists that the materials modelling landscape with all its players is very diverse,

scattered over many and different type of organisations, and shows little visible impact as to

facilitating European industrial growth. In particular, examples are difficult to come by where

materials modelling in all its aspects was critical for securing a commercial success, let alone a

quantification of the materials modelling “return on investment”. Furthermore, materials

modelling is often perceived as a tool in support of product or process research and

development making it seem “a bridge too far” to address the challenges at hand. A critical

understanding is therefore required as to which various materials modelling techniques are

available, where they can be found, who needs to “operate” them, how much do they cost,

and what can they actually bring? The EMMC‐CSA is accelerating this understanding by seeking

insight into the industrial needs and finding ways to make materials modelling an integrated

part of good industrial practices.


What are the hurdles/challenges for a broad and successful implementation of

materials modelling in industry and particularly SME’s in view of limited available

success stories?

Why is it difficult or easy to successfully implement materials modelling in industry and

SME’s and demonstrate this as such with publishing success stories?

What are the companies’ and industrial stakeholders’ expectations from materials

modelling in order to visibly display success stories for a larger audience?

What improvements are needed/recommended for enhancing (if any) success in

industry and disseminate these successes?

Is there sufficient awareness of materials modelling success?

How should or can success stories be quantified?

How can success stories be better disseminated?

Does dissemination of success stories have an impact on materials modelling use?

Who are the audiences/stakeholders/champions to address for disseminating success

stories?

Is cost a real issue hindering materials modelling implementation, use, and publishing

success stories?


20


TheneedforMarketplaceservices:Requirementsfromstakeholders(modeldevelopers,translators,softwareownersandendusers)

Chair Nicola Marzari (EPFL/MARVEL, CH), Georg J. Schmitz (Access Technology, DE)

Impulse 1 Alejandro Strachan (Purdue University, US)

Making data and simulation tools accessible and useful ‐ the nanoHUB experience, approach and impact

Impulse 2 Sergio Lopez (Scientific Computing & Modelling, NL)

Fortissimo: an HPC hub to enhance the global competitiveness of European manufacturing SMEs

Impulse 3 Thomas Göhler (MTU Aero Engines, DE)

Materials modelling marketplaces as engines for accelerated industrial innovation and ICME Breakthroughs

Introduction

There is an emerging need to utilize what the information age offers in terms of connectivity

and abundance of information to boost materials modelling related activities allowing rapid

and more integrated development of novel materials and processes. Materials Modelling

Marketplace is a platform for the exchange of information, software and services covering all

aspects of materials modelling and its industrial application.

Materials Modelling Marketplace would foster an ecosystem integrating all stakeholders in a

more sustainable and efficient manner. Marketplaces can act as simple “match making” site

between model and solution providers offering their software and skills by demonstrating

certain applications and solutions/competences, and on the other side “clients” or SME’s

looking for a solution to a specific problem. Marketplace platforms can furthermore provide

more complex services ranging from targeted consultation to software as a service (SaaS),

utilising therefore the full pallet of technological advancements. It is focused on exploring

possibilities of, and offering novel solutions based on information technology to the entire

material modelling community.


How often do you use online forums, search (“google”, other), or other online

resources for consultation or finding information about how to apply materials

modelling (as modeller, translator) or to find collaborations (as translator, researcher)

or potential customers (as software owner)?


21

As a student/educator: What should be the added value from your point of view of

such a Materials Modelling Marketplace in terms of finding the right information,

tools, and collaboration, etc. ? (Optional: please specify a use case scenario).

As a translator/consultant/software or data owner (please indicate role!): What should

be the added value from your point of view (in terms of increased sales, outreach,

marketing…) of such a Materials Modelling Marketplace platform, and for what tasks

and in what way would you envisage utilising such a platform?

Would you like to see all aspects from translation to implementation and running a

simulation (i.e., from the analysis of your problem to application and results) available

on a marketplace platform? And if so how often would you use it (provided your case

is covered by the services)

As a modeller, software or data owner, what conditions/requirements/provisions are

needed before you start trading on such a platform?

Do you expect feedback on software, services, and other “goods” offered on the

marketplace (like, Amazon user feedback for e.g.) will provide added value to both end

users (“customers”) and service/software providers (“merchants”)? If so, how can this

feedback be utilized in your opinion for enhancing confidence and quality of the

marketplace?

What would you expect a Translation1 service on a Marketplace to look like? Would

you expect a database of translators indicating their field of expertise (type of

material, application sector), organisation (research centre, university, software

owner, and consultant), and other required information?

Would you be willing or interested to interact online with Translation experts and

under what conditions?

How can such platforms contribute to education of R&D in industry to better utilise

materials modelling?

Session14 FRI,April7,2017 10:15‐12:00 Room3

Traininginpractice:Efficientexploitationofmodelsfrommodel‐developerspointofview


Impulse 1 Anders Engström, (Thermo‐Calc Software AB, SE)

Challenges and success criteria for successful exploitation of scientific models ‐ Experiences from Thermo‐Calc Software

1 Translators are defined as experts who can translate an industrial problem into materials modelling

cases to be simulated


22

Impulse 2 Ole Swang (SINTEF, NO)

Electronic modelling case: VCM production

Session presenter at Podium Discussion Pietro Asinari (Politecnico di Torino, IT)

Introduction

Model developers are often academic groups or members of R&D facilities which create

models focusing on a specific research goal. There have been a lot of such scientific research

projects successfully accomplished and they already offer wide spectra of applicable modelling

tools ready for industrial use. The remaining task relates to translation and transfer

respectively the way of communication between model developers and end‐users as well as

the required support towards final implementation into the value chain. Model developers idealize the end‐user requirements, identify required materials data and

create a capable model. The level of model suitability to real application cases then depends

on the quality of input data and on the remaining level of complexity. Thus, the successful use

of especially novel models beyond the scientific community is related to the translation of the

results to respective end‐users. At this stage training and translation is mandatory since it

provides key information for decision making and transfers long‐lasting competences to end‐

users. Consequently, the session focusses on the aspects of training with that purpose and

figures out, from model developers point of view, how integration of materials modelling and

required training should look like. Based on existing experience best‐practice for training will

be discussed.


How to bridge the language gap between end‐users and model developers? Is training

necessary also for model developers in order to understand the requirements of

industry?

Is training really as crucial as discussed? Is it considered as an important event by the

end‐user? If yes, which significant (model) aspects should be trained?

In your (auditorium) experience, what are the types of most successful/useful training

events (workshops, webinars etc.)?

What is more important for the training: Simplification of the real engineering

scenarios or the explanation of simplifying formulations describing complex scenarios

to end‐users? (from use‐case to model or from model to use‐case)

Should training also include “scientific” understanding / translation of the modelling

results?

Should filtering of industrially relevant information (from model input until post‐

processing) be trained separately?


23


Industrialdeploymentofmaterialsmodellingsoftware:Currentpractice,constraintsandbarrier

Chair Volker Eyert (Materials Design, FR)

Impulse 1 Gilles Dennler (IMRA Europe, FR)

Accelerated discovery of new energy materials by high throughput ab‐initio computations and experimental validation

Session presenter at Podium Discussion Volker Eyert (Materials Design, FR)

Introduction

Traditionally, the majority of discrete materials modelling codes have been developed in an

academic environment, i.e. at universities and in government laboratories. Also, despite the

growing importance, discrete materials modelling not yet had the widespread distribution and

use in industry as, e.g. continuum and product design modelling. Distribution and use of

software was thus mainly among academic groups. This development and distribution practice

had a strong impact on functionality, development priorities, software engineering, validation,

licensing models, support, and sustainability. Different perspectives by the mostly academic

developers and industrial users turn out as a major obstacle of wider industrial deployment of

materials modelling software. The present session aims at analysing current practices of

software deployment and identifying constraints and barriers to pave the way for improved

practices, which would foster the application of materials modelling software in industrial

contexts.


As a user, what type of software is currently used in your organization, i.e. commercial,

free software, in‐house developments, other?

As a software developer, what license model are you using for your software?

What is the experience in terms of functionality, documentation, robustness, training,

support, interoperability? Which of those criteria are most important?

Which type of software in your opinion is most successful and why?

Which requirements are currently not met?

From the viewpoint of software owners, which licensing and distribution models are

currently practiced and which have the highest industrial impact?

What is the biggest barrier from the perspective of software users and software

owners?

Which other aspects and ideas do you see related to the present session topic?


24


IntegrationofmaterialsmodellingintoBusinessDecisionSupportSystems

Chair Teodoro Laino (IBM, CH)

Impulse 1 Barbara Holtz (Dassault Systemes BIOVIA, UK) Materials Modelling in the Context of Business Decisions

Impulse 2 Katya Vladislavleva (DataStories International, BE) Storytelling in advanced analytics: facilitating interaction and insights for effective research

Session presenter at Podium Discussion Teodoro Laino (IBM, CH)

Introduction

Intelligent, well‐informed decisions have to be made in a timely manner by industry on the

choice of materials and processing needed for enhancing existing and creating novel market

differentiating products. Indeed, the selection of an optimal material for an engineering design

or manufacturing process from among available, existing options based on technological and

business commercial attributes is a relatively well established decision making process in

industry.

However, the incorporation of materials modelling in the business decision process is

hampered by the lack of standardised integration schemes into existing and future BDSS

systems, decision makers in industry have to rely on their R&D departments, if available, to

develop their own custom tools to pre‐process input files and build initial models, then convert

simulation results into a suitable post‐processing applications to extract relevant materials and

process attributes including key performance indicators (KPI’s) that are necessary for the BDSS.


What are the opportunities for integration of materials modelling in BDSS?

How can available data (e.g., publications, patents, internal documents) optimise the

design and parameterization of specific materials simulations? How can these data

complement those areas not or insufficiently (e.g., only qualitatively) covered by

materials simulations?

What types of query could you envisage a non‐expert asking of a BDSS system with

regard to materials modelling?


25


SynergywithnationalModellingTranslationActivities

Chair Javier Llorca (IMDEA, ES), Ole Swang (SINTEF, NO)

Impulse 1 Javier Llorca (IMDEA, ES) Task force on Translation

Impulse 2 Lula Rosso (VTT Technical Research Centre of Finland, FI) Synergies with national programmes, trends and opportunities for modelling translators

Session presenter at Podium DiscussionLula Rosso (VTT Technical Research Centre of Finland, FI)

Introduction

Industrial “end‐users” especially SMEs, often have insufficient or even lack of expertise in

modelling that prevents them from integrating modelling and simulation software into their

industrial development workflows. There is a need to identify ways and supporting

instruments of how to bridge the “language gap” between industrial stakeholders and

scientific modellers.

Translators, who have the ability to ‘translate’ industrial problems into materials modelling

cases to be simulated, could play a crucial role in this process.

Within EMMC a Europe‐wide network of translators is created and translation methodologies

based on best practices are being developed. But the last step, the actual implementation, is

often a local operation between an expert and SME.

Last year the EC initiated a Task Force on Translation, with the goal to stimulate a synergy

between the translation activities at national and at European level (via EMMC).


What are the possibilities to utilise or initiate regional/national funding for

cooperation between SMEs and research centres/technical schools/industrial

campuses?

How to involve relevant national, regional and local authorities and organizations in

translation activities?

What type of show cases on translation need to be collected to demonstrate to

national/regional bodies the need of the SMEs for translation?

Could large companies facilitate the use of modelling by SMEs?

What type of joined EMMC and national translation activities would be most

useful/efficient: e.g. workshops, training events, other?


26


Industrialdeploymentofmaterialsmodellingsoftware:Thewayforward

Chair Georg J. Schmitz (Access Technology, DE)

Impulse 1 Roger Assaker (E‐Xstream, LU) Digimat, The multi‐scale material modelling platform – From Academic Research to Industrial ICME

Impulse 2 Rob Meier (DSM, NL) How to improve the impact of modelling in industry: on the needs for tools and appropriate software

Session presenter at Podium Discussion Georg J. Schmitz (Access Technology, DE)

Introduction

Successful industrial deployment of materials modelling software is multifaceted and many

aspects regarding the specific models, training and translation and the integration and

interoperability of software are discussed in other sessions. Here the focus will be on aspects

such as software engineering and quality, robustness, licensing models, user documentation,

continuous support and resolution of issues.

Successful software for materials modelling has an expected lifetime of many decades. This

long‐term nature requires a sound foundation in terms of software engineering and quality as

well as legal and business aspects. This includes extended design principles as, e.g. strict

modularity of codes as well as clear guidelines for software development and documentation

especially in large team efforts. During its lifetime software (and its licensing) must be

adaptable not only to additional functionality and features but also different types of

integration, hardware and delivery. Along with these changes, software owners need to adapt

their business models to changing licensing and delivery scenarios, e.g. resulting from different

types of hardware architecture (in case of per CPU licensing), cloud deployment, hosting (SaaS)

etc. Hence license and delivery models need to be carefully thought through to ensure a

sustainable exploitation by both academic and industrial end‐users.


What are the expectations of end‐users with respect to the use and the conditions of

use of modelling software?

What are the three most important barriers for future successful software deployment

in industry?

What actions should be taken to overcome them?

What aspects should be taken into account when defining standards for software

development in materials modelling?


27

Should materials modelling software get ready for use in regulatory purposes and

how?

What are most suitable forms of future software ‘packaging’? For example Functional

components versus monolithic codes, integrated software packages versus individual

codes (with standard APIs; see also the Interoperability Session 5), generic modelling

codes vs task specific Apps.

How should software be deployed in future to enable stronger integration, e.g. with

data/repositories?

What licensing and distribution models should be adopted from an industrial end user

perspective?

What licensing and distribution models will best support the long term software

quality and sustainability?

What business models could/should software owners adopt?

Which other aspects and ideas do you see related to the present session topic?


EconomicimpactofMaterialsModelling

Chair Hein Koelman (DOW Benelux, NL), Gerhard Goldbeck (Goldbeck Consulting, UK)

Impulse 1 Christa Court (University of Florida, US) Modelling Impacts

Impulse 2 Albert Konter (M2i ‐ Materials Innovation Institute) Choices for materials modelling in Industrial Product Innovation

Session presenter at Podium DiscussionChrista Court (University of Florida, US)

Introduction

A recent report on the Economic Impact of Materials Modelling provided an overview of

approaches and methods for impact assessment and presented analysis of an industry survey.

Qualitative indicators of impact were identified including: more efficient and targeted

exploration; deeper understanding; broader exploration; R&D strategy development; source of

property data; trouble shooting; performance optimisation; intellectual property protection;

value chain benefits; improved communication and collaboration between R&D and

production; more efficient upscaling and market introduction; more compelling marketing.

Quantitative, microeconomic performance indicators include financial metrics such as net

present value (NPV), return on investment (ROI), and internal rate of return (IRR). Where

sufficient data are available, these analyses could be extended to a more in depth cost‐benefit

analysis.


28

On a company internal level, a case study has demonstrated the effectiveness of defining a

performance metrics for a corporate modelling R&D function. Furthermore, maturity models

could be developed for materials modelling, in order to provide a “consistent measuring stick

that businesses can use to assess the current maturity level of their operations and identify key

areas for improvement.”

Discussionpointsandquestions What benefits can be derived/expected from materials modelling?

Discuss how benefits of materials modelling can be quantified. What are the best

options and what is required to put them into practice?

How can SMEs assess the affordability and benefit of utilising materials modelling?

What are the barriers and implementation hurdles hindering/limiting impact?

What are best practices? Would a maturity model be useful for assessment?

EMMC International Workshop 2017April 5-7, 2017 - Vienna / Austria

OVERVIEW

This event has received funding via the EMMC-CSA project from the European Union‘s Horizon 2020 research and innovation programme under Grant Agreement No 723867

OWT MOORENO MOOREERHT MOORSESSION 10

Translation & training in practice: What we can learn from software owners?Chair: Erich Wimmer (Materials Design, FR)

SESSION 11Repositories of materials (modelling) data and

knowledge: design and useChair: David Cebon (Granta Design, UK)

SESSION 12Success stories and how to widen success

throughout European industryChair: Rudy Koopmans (Koopmans Consulting, CH)

Impulse presentationby Johannes Fraaije (CULGI, NL)

Impulse presentation #1by Zachary T. Trautt (NIST, US)

Impulse presentation #2by Saulius Grazulis (Vilnius University / COD, LT)

Impulse presentation #1by Rudy Koopmans (Koopmans Consulting, CH)

Impulse presentation #2by Massimo Noro (Unilever, UK)

1445 - 1500

1645 - 1800

1800 - 1900

1930 - 2230

Time

0900 - 0930

0930 - 1000

1000 - 1015


Training in practice: Efficient exploitation of models from model-developers point of view

Chair: Pietro Asinari (Politecnico di Torino, IT)

SESSION 15Industrial deployment of materials modelling

software: Current practice, constraints and barrier

Chair: Volker Eyert (Materials Design, FR)

SESSION 16Integration of materials modelling into Business Decision Support Systems

Chair: Teodoro Laino (IBM, CH)

Impulse presentation #1by Anders Engström (Thermo-Calc Software, SE)

Impulse presentation #2by Ole Swang (SINTEF, NO)

Impulse presentationGilles Dennler (IMRA Europe, FR)

Impulse presentation #1by Barbara Holtz (Dassault Systemes BIOVIA , UK)

Impulse presentation #2by Katya Vladislavleva (DataStories International, BE)

1200 - 1300


Synergy with national Modelling Translation Activities

Chair: Javier Llorca (IMDEA Material Institute, ES) Ole Swang (SINTEF, NO)

SESSION 18Industrial deployment of materials modelling

software: The way forward

Chair: Georg J.Schmitz (Access Technology, DE)

SESSION 19Economic impact of Materials Modelling

Chair: Hein Koelman (DOW Benelux, NL)

Gerhard Goldbeck (Goldbeck Consulting, UK)

Impulse presentation #1by Javier Llorca (IMDEA Material Institute, ES)

Impulse presentation #2by Lula Rosso

(VTT Technical Research Centre of Finland, FI)

Impulse presentation #1by Roger Assaker (E-Xstream, LU)

Impulse presentation #2by Rob Meier (DSM, NL)

Impulse presentation #1by Christa Court (University of Florida, US)

Impulse presentation #2by Albert Konter

(Mi2 - Materials Innovation Institute, NL)

1445 - 1500

1500 - 1600

1600 - 1610 Recommendations of the Industrial Advisory Board (IAB) to the EMMC1610 - 1615

ROOM ONE

Conclusion Workshops as Podium discussion Part 1 (Sessions 1-9, 11)

ROOM ONE

DAY 2 - April 6, 2017

ROOM THREE

1300 - 1445

Impulse presentations each 15 min.

Coffee break

1500 - 1645


Joint Session on Marketplaces 13The need for Marketplace services: Requirements from stakeholders (model developers, translators, software owners and end users)

Chair: Nicola Marzari (EPFL / MARVEL, CH) and Georg J. Schmitz (Access Technology, DE)

Impulse presentation #1by Alejandro Strachan (Purdue University, US)

Impulse presentation #2by Sergio Lopez (Scientific Computing & Modelling, NL)

Impulse presentation #3by Thomas Göhler (MTU Aero Engines, DE)

Conclusion and Goodbye by Nadja Adamovic (TU Wien, AT)

Industrial Advisory Board Meeting (only for members of the IAB) Chair: Rudy Koopmans (Koopmans Consulting, CH)Dinner at Restaurant "Lusthaus"

(1900 Lobby / 1915 Transfer)

DAY 3 - April 7, 2017

Plenary Talk 5Challenges and opportunities for industrial materials modelling

by Tanja Graf (Volkswagen, DE)

Plenary Talk 6The role and impact of materials modelling in accelerating innovation in an industrial context

by Hein Koelman (DOW Benelux, NL)

Coffee break

ROOM ONE

1015 - 1200


Lunch

1300 - 1445


Coffee break

Conclusion Workshops as Podium discussion Part 2 (Sessions 10, 12, 14-19)

EMMC International Workshop 2017 April 5-7, 2017 - Vienna / … · 2018-08-10 · EMMC...

Documents

Transcript of EMMC International Workshop 2017 April 5-7, 2017 - Vienna / … · 2018-08-10 · EMMC...