Efficient decision strategies on the flight deck

Ute Fischer, Judith Orasanu, & Mike MontalvoNASA Ames Research CenterMoffett Field, CADecision making in aircraft crews becomes most pertinent when abnormal situationsoccur, such as system malfunctions, or bad weather conditions. In these circumstances,workload is high: Crews have to determine the nature of their problem, and what course ofaction they should take. Moreover, additional tasks have to be performed aside fromstandard procedures. On the other hand, we know from psychological research that humanscan attend to and interpret only a limited amount of information at any given point in time(e.g., Broadbent, 1982; Miller, 1956). In light of these constraints, both crew productivityand flight safety depend critically on efficient crew problem solving and decision making.But what constitutes efficient problem solving and decision making behavior?Efficient behavior is generally understood as “productive of desired effects; esp.:productive without waste” (Webster's 3rd). We need to modify this generic definition inseveral ways to be able to apply it to aeronautical decision making. Orasanu et al. (1993) listSituation Assessment, Risk Assessment, and Time Assessment as integral parts of decisionmaking in the cockpit. Accordingly, efficient cockpit decision making behavior could becharacterized as follows: Efficient crews gather crucial information that is needed to make asoptimal a decision as possible under the given circumstances, and they are doing so withoutwasting mental energy, time and money. The important phrase in this context is “under thegiven circumstances.” The efficiency of decision making behavior cannot be determinedindependent of situational constraints; i.e., the nature of the problem, the risk involved, timeconstraints, and crew size (Orasanu, 1993).However, to date we know little about how efficient crew problem solving and decisionmaking behavior changes as a function of different situational constraints. Previous work(Mosier, 1991; Predmore, 1992) has focused on crew performance in the context of a specificproblem type. The present study differs from these analyses in two respects. First, wedescribe how aircraft crews cope with different types of problems. And second, we attemptto discern characteristics of optimal task management behavior. For this reason, wecontrasted the behavior of more effective and less effective crews. Underlying this approachis the assumption that crews who commit few operational errors also demonstrate optimalproblem solving and decision making behavior.MethodOur data were drawn from a two-day B-727 full mission simulation study conducted byChidester et al. (1990). The study was comprised of two scenarios which confronted thesubjects with various system malfunctions and/or bad weather conditions. Twenty-threethree-member commercial airline crews participated as subjects.The performance of the crews were videotaped and their verbal interactions transcribed.Crew performance data were collected from two sources (1) expert observation, and (2)video coding of operational errors such as poor airplane control and omissions of requiredprocedures. Operational errors were further classified in terms of their severity (seeChidester et al. for further details).From the total of 23 crews, we selected 20 crews - 10 crews per scenario - for ouranalyses. For each scenario, we took the five crews that had committed the fewestoperational errors (= more effective crews), and the five crews with the highest number ofoperational errors (= less effective crews). Not all crews were consistent in their