S13: From paper to production: turning computational mechanics research into engineering software
Co-chairs:
Paulo R. Refachinho de Campos* (Dassault Systèmes UK)
Nicola Bombace (AMD UK)
Callum J. Runcie (University of Glasgow)
Keywords: Engineering Software, High-Performance Computing, Open-Source Software, Verification and Validation, AI-Assisted Development, Research-to-Production.
ABSTRACT
Computational mechanics research continually produces new methods, algorithms and modelling approaches with the potential to improve engineering analysis. Yet the journey from a successful research implementation to software that is reliably used by engineers and researchers presents challenges that extend well beyond demonstrating a method in a publication.
This Mini-Symposium brings together developers and users of commercial, open-source and established research software to showcase computational mechanics technologies that have progressed beyond proof-of-concept implementations and to discuss the journey that enabled their adoption. Contributions may introduce the software, its capabilities and applications, while sharing the technical and practical experience behind turning research ideas into tools used by a wider community.
The session will explore questions such as: What separates a promising research implementation from dependable engineering software? What challenges emerge when a method moves beyond controlled academic examples to real users and applications? How are requirements such as robustness, verification and validation, computational performance and scalability, maintainability and long-term evolution addressed in practice? How are emerging technologies, including AI-assisted development and evolving CPU, GPU and HPC architectures, changing the way computational mechanics software is designed, developed and deployed? And what can academia, software developers and the wider computational mechanics community learn from successful and unsuccessful attempts to bridge this gap?
By providing a forum for commercial developers, open-source communities, research software engineers, academics and HPC specialists to present established software and exchange experience, the Mini-Symposium aims to strengthen the connection between computational mechanics research and practical engineering impact. It will also expose students and early-career researchers to the often-unseen challenges involved in transforming computational research into software that can be deployed, maintained and trusted.