Evaluating Psychological Aircraft Accident Reports for Differences in the Investigation of Human Factors

Introduction

Analyzing aircraft accidents clearly is an important method for maintaining and improving safety in the aviation domain and is done frequently (e.g. Branford, 2011; De Voogt, 2011; Endsley, 1995a; Goh & Wiegmann, 2002; Van Doorn, 2014; Van Doorn & de Voogt, 2007, 2011). Nevertheless, evaluating these accident reports is equally important. Knowing how research is applied in practice and which methods find application is ultimately essential for improving the expertise of accident investigators and thus flight safety. However, such evaluations seem to be generally neglected, especially in the military aviation domain. Since such military reports are often classified, this is hardly surprising. Yet, we managed to obtain permission for analyzing aircraft accident reports issued by aviation psychologists of the German Armed Forces and aim on bridging that gap.

The aviation psychologist is one member of the aircraft accident investigation board and has the task to investigate the human factors contributing to the accident. Therefore, the aviation psychologist issues an accident report that provides one basis for the work of the accident investigation board.

Our leading question was which human factors for explaining aircraft accidents were applied in those psychological aircraft accident reports and if differences in the identified human factors exist. Therefore, we analyzed their content for established human factors known to be involved in accidents. As a first step, we examined, if guidelines on behalf of aviation-related organizations exist that suggest which human factors should be considered during an accident investigation. Indeed, such guidelines exist for example on behalf of the International Civil Aviation Organization (ICAO) or the International Society of Air Safety Investigators (ISASI). Nevertheless, these were not suitable as a basis for a category system. Therefore, as a second step, we conducted a literature research based on the recommendations by ICAO (1993). We focused on a) identifying particular human factors with relevance for accident incurrence and b) on aircraft accident models/taxonomies. As a result, we decided to focus on the following factors (for a detailed overview of the factors, please refer to the respective literature):1) Mental Workload (e.g. Kahneman, Beatty, & Pollack, 1967; Ruffel Smith, 1979; Young & Stanton, 2001), 2) Situation Awareness (e.g. Endsley, 1988, 1995a, 1995b; Sarter & Woods, 1991, 1995), 3) Decision making (e.g. Billings & Reynard, 1984; Flin, Salas, Strub, & Martin, 1997; Flin et al., 2003; Klein, Orasanu, Calderwood, & Zsambok, 1993; Shapell et al., 2007), 4) Cooperation (e.g. Flin et al., 2003), 5) Leadership and Management (e.g. Flin et al., 2003; Sumwalt & Lemos, 2010), 6) Fatigue (e.g. Caldwell & Caldwell, 2003; Rosekind, Co, Gregory, & Miller, 2000), 7) Stress (e.g. Harris, 2011; Lazarus & Folkman, 1984; NTSB, 2001; Salas, Driskell, & Hughes, 2016), 8) Spatial Disorientation (e.g. Cheung, Money, Wright, & Bateman 1995; Lyons, Ercoline, O’Toole, & Grayson, 2006; Singh & Navathe, 1994)., 9) Human Machine Interaction/Design (e.g. Baxter, Besnard, & Riley, 2007; Billings, 1997; Sarter & Woods, 1995; Sarter, Woods, & Billings, 1997; Sherry, Polson, & Feary, 2002; Rudisill, 1995). Additionally, we decided to focus on the following aircraft accident models: 1) The Human Factors Analysis and Classification System (HFACS; e.g. Shappell & Wiegmann, 1997, 1998, 2000, 2001; Wiegmann & Shapell, 2001, 2004) and 2) the AcciMap approach (e.g. Rasmussen, 1997; Rasmussen & Svedung, 2000; Svedung & Rasmussen, 2002; Vicente & Christoffersen, 2006).