No abstract

No abstract

Agent-Based Modelling has been used for social simulation because of the several benefits it entails, including the capacity to improve conceptual clarity, enhance scientific understanding of complex phenomena, and contribute policy-relevant insights through simulation experiments. Social models are often constructed by inter-disciplinary teams that include subject-matter experts with no programming skills. These experts are typically involved in the creation of the conceptual model, but not the verification or validation of the simulation model. The Overview, Design concepts, and Details (ODD) protocol has emerged as a way of presenting a model at a high level of abstraction and as an effort towards reproducibility of Agent Based Models (ABMs). However, this popular protocol does not improve the involvement of experts because it is typically written after a model has been completed. This paper reverses the process and provides non-programming experts with a user-friendly and extensible tool called ODD2ABM for creating and altering models on their own. This is done by formalizing ODD using concepts abstracted from the NetLogo language, enabling users to generate NetLogo code from an ODD description automatically. We verified the ODD2ABM tool with three existing NetLogo models and assessed it with criteria developed by other researchers.

There are many different notions of models in different areas of science that are often not aligned, making it difficult to discuss them across disciplines. In this study, we look at the differences between physical models and mental models as well as the difference between static and dynamic models. Semiotics provides a philosophical underpinning by explaining meaning-making. This allows for identifying a common ground between models in different areas. We use examples from natural sciences and linguistics to illustrate different approaches and concepts and to find commonalities. This study distinguishes between systems, models, and descriptions of models. This distinction allows us to understand the commonalities of mental and physical models in different areas.

No abstract