Carl Westin

8 Introduction Prequel simulation Conformance design Experiment simulation Collect solution Manual conflict solving Scenarios repeated several times Quantify solutions Determine individual conflict-solving style Define conformal solutions Define nonconformal solutions Script resolution advisories Real-time experiments Analysis & programming Aided conflict solving Replay solutions as advisories Scenarios as in Prequel Same participants as in Prequel Analyze conformance effects F IGURE 1-2: Experimental approach for studying strategic conformance. solutions are analyzed and quantified against a solution parameters framework. The purpose of this stage is to determine each controller’s conflict-solving style (more generally referred to as problem-solving style) for specific conflicts encountered in the prequel simulation. Importantly, determining a controller’s conflict-solving style based on how he/she solves the same conflict repeatedly ascertains, and validates, that the style represents a consistent solution preference, and not a random behavior. The individual conflict-solving styles are then used to script conformal reso- lution advisories to be replayed. A conformal advisory is intended to match a controller’s own solution for a specific conflict, including matching parameters of aircraft choice, resolution type (i.e., heading, speed, or combination), and resolu- tion direction (e.g., left or right heading). In contrast, a nonconformal advisory is intended to deviate from a controller’s own solution. A realistic nonconformal advi- sory can be acquired and validated by using a different solution for the same conflict made by another controller. As such, there is no need to develop and validate an ac- tual conflict resolution algorithm. The same controllers participate in the final experiment simulation , in which they are supported by an automated decision aid for solving conflicts. The decision aid provides either conformal or nonconformal resolution advisories, although con- trollers are led to believe that all advisories are generated by the automation. By all other means, this simulation is identical to the prequel simulation. In addition, the described method provides a novel approach for investigating decision-making strategies in conflict resolution. Previous elicitation methods have generally been based on subjective techniques (e.g., interviews, focus groups, and questionnaires) in combination with static traffic and conflict scenarios. 24, 31, 35, 40 In