Consistency and consensus in conflict solution performance was investigated in two real-time simulations involving experienced air traffic controllers and trainees. The simulations consisted of participants repeatedly playing a specific en route traffic scenario. Conflict solution performance and consistency was measured by analysing participant’s solutions to a specifically designed conflict. Consensus was determined by comparing participant’s consistent conflict solution behaviour. Four different clusters of consistent behaviour were identified. All participants could be classified as consistent according to one or more of these clusters. Consensus in solving the designed conflict could not be determined between participants. There were, however, smaller groups of participants who solved the designed conflict similarly. Participants varied considerably in terms if when to interact and solve the conflict. This is a potential issue when it comes to determining thresholds for the timing of decision-aid interaction.