Reducing Pilot Error: Automate or Cooperate?

John Plessis

(Swansea University)

Over the last three decades there has been a rapid increase in flight deck automation. This has resulted in pilots becoming systems managers attempting to monitor mode-rich systems that possess both high levels of autonomy and authority. Consequently, the ability of the pilot to maintain situational awareness has become increasingly complex, expecially when the agency of the automation enables it to be both silent and powerful. It has been arguedthat increasing the level of automation is the means by which pilot error is reduced. However, a key question remains: what constitutes a pilot error? This talk will explore pilot error in terms of a more constructivist approach whereby error is explored systematically rather than casually - it is about uncovering systematic patterns whereby the successful construction of safety degrades. Using this constructivist view of pilot error, it will be suggested that increasing pilot automation co-operation, rather than simply increasing automation, may provide a useful way forward in reducing pilot error.

Friday 16th June 2017, 15:00
Board Room (Faraday 314)
Department of Computer Science