System dynamics applications for defence combat modelling: preliminary insights from a literature exploration

Abstract

System dynamics (SD) encompasses a suite of conceptual and simulation methods used to study complex system behaviour over time. The structure of an SD model represents the cause-effect relationships and feedback loops in the system. SD models include stocks and flows to map system structure. Stocks, also known as accumulations, characterise the system state and change in response to flows (also known as rates). For example, the number of armoured vehicles involved in a conflict could be represented as a stock that changes in response to the attrition rate. SD causal mapping and conceptual models help teams generate ideas and facilitate shared understanding by developing a coherent representation of the problem. The causal representation of SD models provides a useful means for stakeholders to understand dynamic problems, reason about the effect of changes, and identify potential leverage points. The use of simulation methods to examine dynamic behaviour over time enables assessing alternative decisions under a range of scenarios, evaluating the veracity of different views and providing evidence linking assumptions and outcomes.SD has been used to analyse a range of different defence decisions and topics. In this paper, we focus on combat modelling as an important research and application area in defence modelling. Given the lack of synthesis on the use of SD in combat modelling, this paper provides a preliminary exploration of the topic to identify the most salient points and set the groundwork for future research. We review and synthesise published SD applications in combat modelling to understand how SD models have been used to analyse alternative combat force designs and strategies. The review provides a starting point for researchers who want to understand the SD combat modelling literature and those who want to build SD models to examine combat dynamics.The paper is organised around addressing four research questions: (i) why use SD for combat modelling, (ii) where does SD combat modelling fit within the broad SD defence modelling domain, (iii) what recurrent themes and operationalisation can be found in SD combat models, and (iv) what is the SD combat modelling state of the art? We conclude by identifying promising future research directions, such as: applying SD to explain the impact of new technologies involving the integration of autonomous systems to combat.