Migration of control in distributed intelligence systems

A methodology and associated models are described that allow designers of information processing and decision systems to investigate the effects their designs have on the properties of decision-making organizations. Specifically, a quantitative model of a property that becomes very important when distributed decision making is considered, the migration of control, is described. An effort is made to determine how the migration occurs and which decision-making node becomes dominant. An attempt is also made to develop rules for redesigning the protocols of the organization or restructuring it to avoid migrations. The model and the computational procedure are illustrated by an example in command and control. The analysis shows that the presence of distributed decision making gives rise to dynamic phenomena that need novel approaches and concepts for their description