Abstract
Ratings provided by Pilots on workload scales and usability surveys can be biased by subjective differences in perception, experience, skill, emotional state, motivation, and estimation of risk/cost that may be associated with performing a task. Personality dynamics can further compound polarization of issues during pilot debriefings. What if these unwanted effects could be filtered out of pilot data collection and we could cost-effectively access a higher-order, collective ‘pilot brain’ made up of a combined pilot intellect, intuition, and experience to provide more accurate insight into workload and usability? Swarm AI technology was used in a high fidelity pilot simulation event and compared against a traditional methodology for collecting workload and usability survey data. Pilot and Subject Matter Expert workload and usability survey ratings were collected during the event and compared to a post-event pilot swarm. The results of the study showed pilots engaging in collective intelligence were found to be more effective at rating workload, and also more aligned with Subject Matter Expert workload ratings. This initial workload testing suggests that Swarm AI technology and techniques have great potential for usability research by activating the collective intelligence of groups, which can exceed that of the individual performing alone. The usability survey sample was limited, therefore further study is recommended to validate the generalizability of this technology to Likert Scale data.
Get full access to this article
View all access options for this article.