Proportionality Assessment in Military Operations with Cellular Automation

Abstract

In this research a Cellular Automation (CA) approach to model the proportionality assessment in military operations is introduced by capturing spatial-temporal dynamics of this complex decision-making process. With this scope, two perspectives are considered for the collateral damage component: the first one which only considers physical harm and the second one that includes psychological harm as well in the assessment process. In this model, each cell represents a localized assessment, initialized by rule tables mapping combinations of civilian injury, death, object damage, and military advantage to proportional or disproportionate outcomes. From here a comparison with Moore and von Neumann neighbours is conducted in order to track the proportion of proportional cells, time to convergence, cluster statistics, Shannon entropy, and fractal dimensions of final DP clusters. Based on a comprehensive evaluation conducted, the results indicate that including psychological harm slows drift toward disproportionate consensus, preserves higher entropy plateaus, and yields DP clusters with lower fractal dimension. This implies resilience from the proportional judgments. Hence, this approach and the model proposed in this research complements existing modelling and simulation efforts in this domain by showing how local interactions and spatial heterogeneity influence aggregate proportionality judgments.