An architecture for automated group formation within remote laboratories

Group/Team formation has been a well studied field in numerous contexts, (i.e. business teams, project teams, educational teams etc.) but have barely been considered within the scope of remote laboratories. Formation of educational groups in traditional labs/classes often occurs in an ad-hoc fashion where students are assigned to groups mostly without any particular constraints or regard to the group composition that is most likely to lead to optimal educational outcomes. This same ad hoc approach has typified the formation of groups within current remote laboratory environments that involve collaborative groups in remote laboratory settings. There is typically no arbitration for allocating group members to a specific group to perform a particular experiment. In this paper, we consider an approach to automated group formation that continuously analyses group performance and uses this to build rules regarding optimal group composition. These rules can be subsequently used to allocate students to groups that are more likely to have higher performance