Tag Archives: model validation

The Stone Raft adrift in the Atlantic Ocean

I spent most of last week at the European Union’s Joint Research Centre in Ispra, Italy.  I have been collaborating with the scientists in  the European Union Reference Laboratory for alternatives to animal testing [EURL ECVAM].  We have been working together on tracking nanoparticles and, more recently, on the validity and credibility of models.  Last week I was there to participate in a workshop on Validation and Acceptance of Artificial Intelligence Models in Health.  I presented our work on the credibility matrix and on a set of factors that we have developed for establishing trust in a model and its predictions. I left the JRC on Friday evening and slipped back in the UK just before she left the Europe Union.  The departure of the UK from Europe reminds me of a novel by José Saramago called ‘The Stone Raft‘ in which the Iberian penisula breaks off from the Europe mainland and drifts around the Atlantic ocean.  The bureaucrats in Europe have to run around dealing with the ensuing disruption while five people in Spain and Portugal are drawn together by surreal events on the stone raft adrift in the ocean.

Fake facts & untrustworthy predictions

I need to confess to writing a misleading post some months ago entitled ‘In Einstein’s footprints?‘ on February 27th 2019, in which I promoted our 4th workshop on the ‘Validation of Computational Mechanics Models‘ that we held last month at Guild Hall of Carpenters [Zunfthaus zur Zimmerleuten] in Zurich.  I implied that speakers at the workshop would be stepping in Einstein’s footprints when they presented their research at the workshop, because Einstein presented a paper at the same venue in 1910.  However, as our host in Zurich revealed in his introductory remarks , the Guild Hall was gutted by fire in 2007 and so we were meeting in a fake, or replica, which was so good that most of us had not realised.  This was quite appropriate because a theme of the workshop was enhancing the credibility of computer models that are used to replicate the real-world.  We discussed the issues surrounding the trustworthiness of models in a wide range of fields including aerospace engineering, biomechanics, nuclear power and toxicology.  Many of the presentations are available on the website of the EU project MOTIVATE which organised and sponsored the workshop as part of its dissemination programme.  While we did not solve any problems, we did broaden people’s understanding of the issues associated with trustworthiness of predictions and identified the need to develop common approaches to support regulatory decisions across a range of industrial sectors – that’s probably the theme for our 5th workshop!

The MOTIVATE project has received funding from the Clean Sky 2 Joint Undertaking under the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 754660 and the Swiss State Secretariat for Education, Research and Innovation under contract number 17.00064.

The opinions expressed in this blog post reflect only the author’s view and the Clean Sky 2 Joint Undertaking is not responsible for any use that may be made of the information it contains.

Image: https://www.tagesanzeiger.ch/Zunfthaus-Zur-Zimmerleuten-Wiederaufbauprojekt-steht/story/30815219

 

Same problems in a different language

I spent a lot of time on trains last week.  I left Liverpool on Tuesday evening for Bristol and spent Wednesday at Airbus in Filton discussing the implementation of the technologies being developed in the EU Clean Sky 2 projects MOTIVATE and DIMES.  On Wednesday evening I travelled to Bracknell and on Thursday gave a seminar at Syngenta on model credibility in predictive toxicology before heading home to Liverpool.  But, on Friday I was on the train again, to Manchester this time, to listen to a group of my PhD students presenting their projects to their peers in our new Centre for Doctoral Training called Growing skills for Reliable Economic Energy from Nuclear, or GREEN.  The common thread, besides the train journeys, is the Fidelity And Credibility of Testing and Simulation (FACTS).  My research group is working on how we demonstrate the fidelity of predictions from models, how we establish trust in both predictions from computational models and measurements from experiments that are often also ‘models’ of the real world.  The issues are similar whether we are considering the structural performance of aircraft [as on Wednesday], the impact of agro-chemicals [as on Thursday], or the performance of fusion energy and the impact of a geological disposal site [as on Friday] (see ‘Hierarchical modelling in engineering and biology‘ on March 14th, 2018) .  The scientific and technical communities associated with each application talk a different language, in the sense that they use different technical jargon and acronyms; and they are surprised and interested to discover that similar problems are being tackled by communities that they rarely think about or encounter.

On the trustworthiness of multi-physics models

I stayed in Sheffield city centre a few weeks ago and walked past the standard measures in the photograph on my way to speak at a workshop.  In the past, when the cutlery and tool-making industry in Sheffield was focussed around small workshops, or little mesters, as they were known, these standards would have been used to check the tools being manufactured.  A few hundred years later, the range of standards in existence has extended far beyond the weights and measures where it started, and now includes standards for processes and artefacts as well as for measurements.  The process of validating computational models of engineering infrastructure is moving slowly towards establishing an internationally recognised standard [see two of my earliest posts: ‘Model validation‘ on September 18th, 2012 and ‘Setting standards‘ on January 29th, 2014].  We have guidelines that recommend approaches for different parts of the validation process [see ‘Setting standards‘ on January 29th, 2014]; however, many types of computational model present significant challenges when establishing their reliability [see ‘Spatial-temporal models of protein structures‘ on March 27th, 2019].  Under the auspices of the MOTIVATE project, we are gathering experts in Zurich on November 5th, 2019 to discuss the challenges of validating multi-physics models, establishing credibility and the future use of data from experiments.  It is the fourth in a series of workshops held previously in Shanghai, London and Munich.  For more information and to register follow this link. Come and join our discussions in one of my favourite cities where we will be following ‘In Einstein’s footprints‘ [posted on February 27th, 2019].

The MOTIVATE project has received funding from the Clean Sky 2 Joint Undertaking under the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 754660.

The opinions expressed in this blog post reflect only the author’s view and the Clean Sky 2 Joint Undertaking is not responsible for any use that may be made of the information it contains.