Notes from the field: Practical considerations for agent-based and continuous representations of system dynamics

In this paper, we consider how to leverage the complementary characteristics of agent-based and stock-flow representations to maximize the utility of systems modelling at acceptable cost and complexity. The increasing capability of tools available for dynamic modelling renders a wide range of structures that could be implemented in models of complex systems. However, attendant with this analytic capacity is the challenge of choosing between alternative model structures. To assist dynamic modellers in designing projects and selecting techniques appropriate for the problem at hand, this paper poses a set of practical considerations for implementing agent-based and continuous representations that encode differential equations using stock-flow structures. Drawing from our collective experiences implementing system dynamics models with different structural forms, we offer recommendations for maintaining a systemic perspective while being open to a multi-method modelling process.