Institution news
Hrvoje will be presenting at the upcoming Simulation and Modelling conference in late September, For a full agenda and to book please visit the website.
Please could you briefly explain your role, involvement, and experience with regards to the
simulation & modelling industry and this event?
Hrvoje Stojic (HS): Machine learning researchers at Secondmind drive the development of the innovative
technology that our products are based on. As a researcher I have been the main contributor
to the machine learning solutions that underpin Secondmind for System Design, seeing it
through the initial idea stage to a solution that is being successfully used in the automotive
industry. There I leveraged advanced machine learning methods to deliver a step-change in
system design optimization and exploration, taking set-based design to new and essential
levels of efficiency and impact.
What, in your experience, has been the biggest roadblock for the industry over the past 2-3
years?
HS: With the caveat that I'm a newcomer to the industry, my impression is that competitive and
regulatory pressures have driven the industry to design more complex systems. As a result,
they are becoming more difficult to understand and develop - which is where machine
learning tools like ours could fit in and greatly facilitate an engineer's job.
What key topics are you excited to discuss at this year's conference?
HS: Broadly speaking, I'm excited to discuss how machine learning can be successfully used for
helping to solve important problems in engineering. More narrowly, I will describe in some
technical detail how we at Secondmind are using data-efficient machine learning to
significantly enhance set-based design capabilities. I will discuss our solutions for tackling
designs of complex systems that can parallelise costly simulations to speed up the design
process and make exploration of high dimensional design spaces much more efficient.
What would you say are the areas of innovation across the UK simulation & modelling
industry?
HS: One key area I have observed - and I'm likely to be biased here - is using machine learning
methods for solving applied problems in engineering. This is happening across engineering
disciplines, from tackling predictive maintenance problems in mechanical engineering, to
optimizing chemical processes and improving yields in chemical engineering.
Who else are you most interested in hearing from on the programme?
HS: I'm interested in hearing from Jorge Lopez Martinez who gave a fascinating talk on battery
development at Williams Advanced Engineering last time around. I'm also keen to hear from
Nawal Prinja, a nuclear energy expert at Jacobs that is also knowledgeable about the use of
machine learning in engineering.
Why is it important for engineers to join this conference?
HS: To expose themselves to randomness. That is, to explore what's happening at the frontier, in
fields that are outside of their comfort zone but still relevant, and get inspiration to help solve
problems in their own work. And for an opportunity to engage with speakers or other
attendants, to speed up their learning journey, and again increase their productivity.
The Simulation and Modelling 2024 conference will take place on 25-26 September 2024 in Birmingham
The event is a comprehensive showcase of the latest techniques and technologies available to practitioners and will provide a crucial forum to address common challenges in model development, complexity, fidelity and speed.
Bringing together simulation practitioners and design expert from multiple engineering sectors, attendees will benefit from fresh perspectives and lessons learned from simulation projects across a wide variety of applications.
Key areas for 2024 include multiphysics applications, the use of digital twins, emerging standards, machine learning, AI and data analysis. For a full agenda and to book please visit the website.