Validation of Agent Based Distillation Movement Algorithms

Agent based distillations (ABD) are low-resolution abstract models, which can be used to explore questions associated with land combat operations in a short period of time. Movement of agents within the EINSTein and MANA ABDs is based on a simple attraction-repulsion weighting system and an associated numerical penalty function. The relative simplicity of these ABDs seems to have led to the general acceptance of their associated movement algorithms, without much validation or analysis of behaviour. This paper analyses these movement algorithms and finds unwanted behaviour, and proposes suggestions for improvement to the penalty function based on relative distances. A novel technique based on the concepts underlying spatial estimation is also proposed as an alternative.