International workshop on machine learning and data analytics in advanced metals processing

A workshop to explore the use of machine learning techniques in alloy design and metals processing will be held on Mon 22nd and Tues 23rd May 2017 in Manchester, UK.

The workshop will bring together experts in computational materials design, machine learning and high-performance computing with those in metallurgy and materials characterisation to consider the particular challenges in exploiting modern computational methods in designing new metallic materials and to explore opportunities and exchange ideas for major collaborative projects. Please see below for further information about the background to and motivation for the workshop. Further information is also to be found on the workshop website:

https://sites.google.com/view/machinelearningandmetals/home

Confirmed invited speakers:

Richard Anderson (Hartree Centre, STFC, Daresbury)

Gareth Conduit (University of Cambridge)

Gabor Cysani (University of Cambridge)

Douglas Kell (University of Manchester)

Nicola Marzari (École Polytechnique Fédérale de Lausanne, EPFL)

Joerg Neugebauer (Max Planck Institut fuer Eisenforschung, MPIE)

Dave Rugg (Rolls-Royce)

Hauke Springer (Max Planck Institut fuer Eisenforschung, MPIE)

Iain Todd (University of Sheffield)

Registration:

Thanks to funding from the University of Manchester, registration is free of charge and includes refreshments, lunch on both days and a workshop dinner on the evening of the 22nd. There is space in the programme for contributed talks and boards will be provided to exhibit posters for viewing during breaks.

To register please fill in this form on the workshop website:

https://sites.google.com/view/machinelearningandmetals/registration

Deadline for registration is Friday 14th April - places are limited to a total of 40 attendees. In the event that the workshop is over-subscribed, attendees will be selected to achieve a balance of representation across fields and institutions.

Venue:

School of Materials (MSS Tower), University of Manchester, Manchester, UK

(The School of Materials is a 5 minute walk from Manchester Piccadilly Station, which is 15 minutes by train from Manchester Airport.)

Outline programme:

A detailed programme will be published soon.

Monday 22nd May: 10:00 - 17:30

Tuesday 23rd May: 08:30 - 17:00

For further information, please contact Chris Race ([email protected]).

With best wishes,

Chris Race and Philip Withers.

Background to workshop

The new Sir Henry Royce Institute (SHRI) for Advanced Materials will provide £235m of capital investment over the next five years. One of nine core areas in the SHRI is Advanced Metals Processing. Computationally guided materials design is widely seen as a primary means of improving the efficiency and rate of discovery of new, improved materials. Key aspects of such projects are the generation (or accumulation) and subsequent management of large quantities of property and performance data. Significant progress has been made in these areas. The most challenging step in such projects is the successful use of such data - how are we to make useful predictions about new candidate materials? How are we to determine which materials might be the best candidates for some proposed application?

Machine learning and data analytics are finding wide application in materials chemistry and biology, but the field of metallurgy presents some particular challenges to the successful use of machine learning, for example:

1) The useful properties of an alloy are rarely dependent solely (and often not even mostly) on the properties of a perfect single crystal (or the basic unit cell). Rather it is the microstructure that plays the determining role. 

2) The microstructure of a final, useable alloy is arrived at via an often complicated series of processing steps, with the evolution at different length and time scales being complex and interdependent.

3) Characterisation of the microstructure of materials is still far from automated and so is costly and time consuming. Raw data for machine learning are therefore expensive and coverage can be sparse. 

4) Brute force, high accuracy simulation of microstructural features is, except in special cases, beyond current computational capabilities. Consequently we need to think not just about modelling methods, but also about the design of high-throughput small-scale processing and high-throughput experimental evaluation/screening.

5) We do not yet have an agreed basis for quantifying/representing a materials microstructure for input into machine learning methods.

6) We do not have the databases necessary to embark on machine learning approaches.