Proceedings of the Human Factors and Ergonomics Society Annual Meeting · 1996 · 78 citations · 4 references
EngineeringSystems EngineeringErgonomicsPerformance MeasurementSchedulingCognitive WorkloadFlight Management SystemsFlight PerformanceSituation AssessmentTask AnalysisWorkload AnalysisAviation WorkloadCrew Performance AnalysisCrew Workload ManagementOperator Workload ManagementPrioritizationWorkloadWorkload AssessmentWorkload Management
Task management on the flight deck is routinely performed but often inconsistently, and it is especially critical during non‑normal situations. The study aimed to analyze pilot task‑management behavior in non‑normal situations. Pilots were observed in a full‑workload simulation to discern their task‑management strategies. Four strategies—Aviate‑Navigate‑Communicate‑Manage Systems, Perceived Severity, Procedure Based, and Event/Interrupt Driven—were identified, and the combination of Perceived Severity for personal workload with Aviate‑Navigate‑Communicate‑Manage Systems for monitoring proved most effective, yielding fewer errors and faster responses.
Task management (TM) is always performed on the flight deck, although not always explicitly, consistently, or rigorously. Nowhere is TM as important as it is in dealing with non-normal situations. The objective of this study was to analyze pilot TM behavior for non-normal situations. Specifically, the study observed pilots' performance in a full workload environment in order to discern their TM strategies. This study identified four different TM prioritization and allocation strategies: ‘Aviate-Navigate-Communicate-Manage Systems’; ‘Perceived Severity;’ ‘Procedure Based;’ and ‘Event/Interrupt Driven.’ Subjects used these strategies to manage their personal workload and to schedule monitoring and assessment of the situation. The ‘Perceived Severity’ strategy for personal workload management combined with the ‘Aviate-Navigate-Communicate- Manage Systems’ strategy for monitoring and assessing appeared to be the most effective (fewest errors and fastest response times) in responding to the novel system failure used in this study.
4