Agent-based modelling (ABM) is a form of simulation that enables us to model a process from the ground up, testing different theories and hypotheses. For EIDOS, it provides one way that we can study how and to what extent different urban mechanisms contributed to a city’s decline or resilience.

ABM, using the open source platform NetLogo, will be used to test what variables contribute to urban decline and resilience. The results will be analysed and interpreted based on our archaeological and historical understanding. 

For example, we could question: how extreme can environmental change be for a city to survive and at what point will it begin to decline? One way to consider this is to look at a variable such as precipitation or annual rainfall. The model might test varying degrees of precipitation, which leads to changes in agricultural growth, and which ultimately can affect a city’s total population over time. By running a model that changes annual precipitation over time we can gain an idea of how little or how much rain is needed for a city to maintain a specific population level. However, one of the benefits of agent-based modelling is that we can consider multiple interacting variables. In addition to changing precipitation for a single city, we could also add another variable that considers how a trade network affects a city’s population due to increased access to foodstuff. If a city has a strong trade network connected to many cities, then less access to local food may not have a significant effect on the city’s population. However, if the same environmental change affects multiple cities and causes part of the trade network to disappear, then this too could impact a city’s ability to gain the necessary resources for the local population. Using ABM, we can run the model changing parameters and seeing what variables have the greatest impact on the model. It also allows us the ability to test different assumptions and hypotheses, helping us to determine what variables contribute most to a city’s ability to persist or not.