Essentials of Modelling

Simulation models are simplified but useful representations of a part of reality. Many modelling projects deal with global issues such as climate change, sustainability, and the management of socio-ecological or socio-technical systems. These issues are highly inter- and transdisciplinary in nature, involving domain knowledge from different disciplines and expert opinions and views from stakeholders.

The development of simulation models involves multiple steps. Each of these steps involves a plethora of aspects to consider that do not directly relate to the mathematical and computational side of modelling. Think of establishing common definitions, a model ontology (does something mean the same thing to everybody?), settling on a model purpose (why do we build the model?), conceptualizing the model to match the model purpose, and considering what data are available or required for the modelling. Moreover, the rise of GenAI may offer opportunities – and pitfalls – to consider in model development and application.
 

This course is intended to make PhD candidates who are working in a modelling project aware of these essential aspects of the modelling work, and to teach them how to deal with common problems. Modelling aspects that we consider in this course will include:
•    How to involve stakeholders?
•    How to establish a shared model ontology?
•    How to balance scientific model credibility and technical detail with relevance and societal views and needs?
•    How to (not) use GenAI in modelling?
 

Set-up
The course consists of (guest) lectures, discussions, and a lot of group work. During the week, the participants will work in teams on a (fictional) modelling challenge, bringing theory to practice. At the end of the week, teams will present their suggested solutions for addressing their modelling challenge.

This course is primarily aimed at PhD candidates who are in the start-up phase of a modelling project, though more experience modellers may benefit from it as well. We purposefully aim to involve people from different scientific backgrounds. The emphasis is not on the mathematical, computational, and statistical aspects of modelling per se, but on the elements of the modelling process besides the mathematical aspects. You will learn about scoping a model, and to think critically about the choices in modelling. The set-up is to have PhDs cooperate with peers from other disciplines to reflect on the foundational design, requirements and purpose of their own modelling projects.