Final Report

Dates and location

21 May — 23 May 2025, Gothenburg, Sweden

Chairperson

Mohsen Mirkhalaf

Co-chairperson

Iuri Rocha

Conference fees

Student fee - early bird: 4000.00 €
Student fee: 5000.00 €
Delegates fee - early bird: 5000.00 €
Delegates fee: 6000.00 €

What other funding was obtained?

Support for the conference venue

What were the participants offered?

Welcome gifts (umbrella, notebook ...), Coffee breaks, Lunches, and Banquet Dinner

Applicants (members)

Mahmoud Mousavi

Applicants (non members)

Tamadur Albaraghtheh
Tito Andriollo
Rui Barreira
Cemal Basaran
Antoine Benady
Magnus Ekh
Mohammed El Fallaki Idrissi
Mirja Granfors
Jacob Gunnarsson
Zhe Han
Fleur Hendriks
Nora Kovacs
Alex Lech
Runze Li
Marina Maia
Gagan Kaushik Manyam
Foudil Meraghni
Knut Andreas Meyer
Mohsen Mirkhalaf
Luca Patrignani
Ivan Ruiz Cozar
Ruben Schmeitz
Ronak Shoghi
Joep Storm
Nasrin Talebi
Petter Uvdal
Li Zheng
Li Zheng

Scientific report

Combining classical principles of mechanics and physics of solids with cutting-edge data science techniques has resulted in very accurate and efficient data-driven approaches. Data-driven models have a very high capacity to harness vast volumes of data generated in material science, engineering, and physics to uncover hidden patterns, relationships, and insights. During the last decade, the following have seen key developments, and further enhancements are ongoing:

Efficient and accurate surrogate modelling: By leveraging existing data and employing Machine Learning (ML), very accurate and remarkably efficient models have been developed. This is particularly important for micro-structurally heterogenous materials for which high-fidelity full-field micromechanical simulations are required for an accurate prediction.
Accelerated material design: The integration of data science techniques allows to identify optimal material properties for specific applications much more rapidly. This can lead to the development of innovative materials with enhanced performance, durability, and functionality.
Discovery of constitutive laws: Data-driven models have enabled discovery of constitutive laws from force and displacement information readily extractable from experiments. In comparison with classical approaches, this method dramatically reduces the required effort and time to derive constitutive laws.

Despite the great progress, challenges persist, including (i) the need for robust data sets, (ii) the interpretability of complex models, and (iii) the integration of physics-based constraints. Overcoming these challenges will pave the way for a deeper understanding of material behavior and the realization of more efficient and sustainable engineering solutions.

In this colloquium, we examined the developments and challenges outlined above by bringing together distinguished plenary and keynote presentations from leading scholars, who provided expert insights into the latest progress at the intersection of mechanics, physics, and data science. In addition to the invited lectures, the scientific program featured a series of thematically organized sessions that facilitated deeper engagement with focused topics, including:

Probabilistic Machine Learning,
Surrogate Modelling,
Learning from Physics and Data,
Image-Based Deep-Learning Models and Tools,
Machine-Learning-Assisted Material Design and Manufacturing.

These sessions provided a platform for rigorous discussion of both foundational principles and cutting-edge techniques, enabling participants to explore how data-driven frameworks can harness datasets to uncover hidden patterns and relationships, enhance predictive capability, and accelerate innovation in material science and engineering.

Number of participants from each country

Country	Participants
Sweden	10
Netherlands	5
Switzerland	4
Germany	3
France	2
United Kingdom	2
Denmark	1
United States	1
Portugal	1
Total	29