Carl Westin

160 Discussion and recommendations choice between alternative solutions often is based on the alternative with the pos- itive information cue available in memory while ignoring the rest. At the end of this process, the controller derives at a solution, or makes choice randomly if in- formation cues cannot be discriminated. The process follows a simple three-step principle: 20 • Search rule: describes which information is searched for. • Stop rule: describes when to stop searching for information. • Decision rule: describes which rules to apply on the information gathered. Controllers may apply different search, stop, and decision rules when solving con- flicts. As a result, differences in information gathered, and the rules for interpreting or weighing the information, results in different solutions to apply for solving the conflict. Although controllers experienced the same conflict repeatedly, differences in information considered makes each repetition unique form their perspective. If any issues or flaws are realized, the solution is modified or rejected altogether. The controller then determines whether the advised solution conforms to her/his preferences for solving the conflict. In experiments, conformal advisories were ac- cepted because they made sense and complied with the controller’s preference for solving the conflict. The advisory is rejected if the controller arrives at a solution contrasting the advisory, unless the advised solution is considered better than their own for some reason. Since conformal advisories were based on the controller’s ear- lier implemented solution to the same conflict, it was likely that the same solution would be accepted again, provided a similar reasoning occurred. A precondition for this heuristic process to be effective is a somewhat struc- tured problem-solving context. In en-route ATC, real-time operations are faceted by many irregularities and disturbances, notably time deviations between planned and actual flight and added complexity from short-term “ad hoc” traffic. In addition, the weather has a considerable influence on traffic movements and ATC capacity en-route and at airports. Despite these irregularities, however, movements within a sector, and the conflicts that occur have similar patterns. Airspace sections (i.e., sec- tors) are typically structured in terms of entry, exit, and intermediate waypoints that make up conditional airways that aircraft follow. Regular patterns can, for exam- ple, be attributed to general flows of eastern or western transatlantic flights during a working day. In addition, several airlines operate the same routes on a regular basis. Moreover, it can always be argued that external similarity (i.e., product ) is not the same as the similarity of underlying process (i.e., process ). After all, automation algorithms are not visible from the visible result. However, although the underlying process is difficult to determine from only knowing the result, we do so every day