Network for Agent-based modelling of Socio-ecological systems in Archaeology (NASA)
The end goal of this research network for the period 2020-2024 is to gather an international, interdisciplinary group of researchers to collect and compile ABM elements (implementation modules, techniques, approaches, etc.) and organise them as an open modelling library. The targets of this network are: 1) identify and compile crucial modelling topics for the library; 2) collect and develop best practices and modelling guidelines; 3) develop tools for interoperability following the FAIR principles; 4) guarantee sustainability of the library; 5) create a structure for international collaboration resulting in joint publications within the network. Targeted topics to include in the library particularly entail those aspects and processes that are often revisited and effectively remodelled from scratch in most projects using ABM. Some preliminary categories include: land models (terrain, hydrology, vegetation, soil degradation), demography (population models, mortality, fertility), household and group economies (labour availability, investment, production, cooperation and competition). These ABM elements will be curated, documented and formatted so that they can be used as base modules, for more elaborate and case-driven models and research questions. Some of the major challenges for creating this library include: 1) To implement individual modules in a general, context-agnostic and expandable way to be relevant for a wide variety of research questions; 2) To enable a systematic interaction of modules via a standardized interface, so that researchers can recombine them into new models; 3) To cope with different programming languages, simulation platforms, and personal workflows. Whereas NetLogo can be particularly highlighted as the standard programming tool for ABM in archaeology, other languages and platforms such as Repast, Python, R and Java are also in use. In archaeology, this is aggravated by the lack of formal technical training in computer science. In the five coming years, this network will address these challenges by holding internal working meetings and organise international symposia to gather international experts and develop a set of general best practice guidelines, make them available for researchers in various fields, and apply these in an open model library. The library will be hosted in an online archive maintained by the Sagalassos Project at the University of Leuven.