The Presence of Behavioral Traps in U.S. Airline Accidents: A Qualitative Analysis

Transcription

1 safety Article The Presence Behavioral Traps in U.S. Airline Accidents: A Qualitative Analysis Jonathan Velazquez School Aeronautics, Inter American University Puerto Rico, Bayamon Campus, Bayamon 00953, Puerto Rico; jvelazquez@bayamon.inter.edu; Tel.: (ext. 2416) Received: 12 April 2017; Accepted: 9 January 2018; Published: 11 January 2018 Abstract: Behavioral traps are accident-inducing operational pitfalls aviators may encounter as a result poor decision making. The Federal Aviation Administration (FAA) identifies existence twelve se negative pilot behaviors. These are: Peer Pressure; Get-There-Itis; Loss Situational Awareness; Descent Below Minimum En Route Altitude (MEA); Mind Set; Duck-Under Syndrome; Getting Behind Aircraft; Continuing Visual Flight Rules (VFR) into Instrument Conditions; Scud Running; Operating Without Adequate Fuel Reserves; Flying Outside Envelope; and Neglect Flight Planning, Preflight Inspections, and Checklists. The purpose this paper was to study nature ir occurrence in airline domain. Four Subject Matter Experts (SMEs) analyzed 34 National Transportation Safety Board (NTSB) accident reports. The SMEs were able to identify many pilot actions that were representative behavioral traps. Behavioral traps were found in all accidents with Loss Situational Awareness and Neglect Flight Planning, Preflight Inspections, and Checklists dominant. Various mes began to emerge, which played important roles in many accidents. These mes included Crew Resource Management (CRM) issues, airline management and fatigue. The findings this study indicated that behavioral traps were prevalent in airline accidents including habitual noncompliance by pilots. Attitude management training is recommended. Keywords: behavioral traps; hazardous attitudes; cognitive biases; pilot error; attitude management 1. Introduction Three out four aviation accidents result from human error [1]. Behavioral traps are operational pitfalls aviators fall prey to as a result bad decision making and inappropriate conduct. Many Aviation Safety Reporting System (ASRS) reports include a statement on lessons learned after an aviation incident. The following are examples some those pilot testimonials that illustrate presence behavioral traps: Aircraft are machines subject to malfunction and we pilots who operate m are humans subject to human error. As a pilot who plans on remaining on line, I have learned a valuable lesson monitor auto-flight system like a hawk. I was making a rushed approach to land. I have learned that when I am rushed is when I really need to take time for checklist. We thought we had a pretty good handle on our fuel state. Anor minute or two fuel and we would have made it safely to airfield. I blame mistake on simple overconfidence. Experience, it seems, is no replacement for doing one s homework. Behavioral traps affect aviation flight safety. Even pilots with a considerable amount experience may fall prey to m and are more likely to accept flights with greater risks, use less available cockpit resources and, in general, are more prone to bad decisions. Safety 2018, 4, 2; doi: /safety

2 Safety 2018, 4, The following section defines behavioral traps, hazardous attitudes and cognitive biases. This explanation is preceded by hazardous attitudes. Similar to cognitive biases and behavioral traps in that y also affect decision making, hazardous attitudes are first introduced to novice pilot before learning behavioral traps. Finally, because very presence se unsafe behaviors is caused by a breakdown team dynamics in cockpit, a brief discussion Crew Resource Management (CRM) is included Defining Behavioral Traps and Or Decision Making Fallibilities Unsafe pilot behaviors have been part FAA literature since foundations Aeronautical Decision Making (ADM) [2]. ADM is defined by FAA [2] as a systematic approach to mental process used by pilots to consistently determine best course action for a given set circumstances. For many future pilots, concept hazardous attitudes is ir first exposure to factors that affect decision making and judgment. Table 1 describes five hazardous attitudes and provides recommended antidote to counteract thought. Before discussing behavioral traps, it is appropriate to take a look at five classical hazardous attitudes, which have received more attention by scholars and FAA. Similar to behavioral traps, se attitudes also affect decision making: Macho, Anti-authority, Impulsivity, Resignation, and Invulnerability [3]. These attitudes go hand-in-hand with twelve behavioral traps described in FAA s [1] Risk Management Handbook. Table 1. Overview hazardous attitudes with appropriate antidote [3]. Hazardous Attitude Anti-authority Impulsivity Macho Invulnerability Resignation Characteristics Pilots with this attitude dislike following rules or having someone else tell m what to do. To se pilots, rules and procedures are a waste time and effort. This attitude belongs to pilots who feel y must do something, anything, and immediately. They seldom take a moment to reflect or evaluate all possibilities. Their actions are result whatever comes first to mind. Macho pilots are risk takers, people overconfident about ir skills and constantly proving that y are better than everybody else. To m, y are best pilots out re. Similar to macho-type pilots, se pilots also take risks but only because in ir mind accidents happen to ors and not to m. They are untouchable. People with this attitude feel y are incapable making a difference. Pilots with resignation-type attitudes are passive and inactive throughout ir flights. To se pilots, when something bad happens it is due to bad luck or fault ors; someone else is responsible. Antidote Follow rules; y are usually right. Not so fast; think first. Taking chances is foolish. It could happen to me. I am not helpless; I can make a difference. The hazardous attitudes share commonalities with behavioral traps. Refer to Table 2 for a summary behavioral traps as defined by FAA. It is likely that at some point in ir careers, veteran aviators may have fallen prey to, or have been tempted by, one or more se tendencies in ir flying pressions [1]. As indicated earlier, behavioral traps and hazardous attitudes share some commonalities. For example, an individual experiencing Anti-authority could fall under behavioral trap Duck-Under Syndrome or Neglect Flight Planning, Preflight Inspections, and Checklists. An aviator exhibiting Macho might experience behavioral trap known as Flying Outside Envelope. Scud Running and Operating Without Adequate Fuel Reserves are indicative hazardous attitude known as Invulnerability. The behavioral traps known as Mind Set and Get-There-Itis are signs a

3 Safety 2018, 4, pilot affected by Impulsivity. Lastly, Getting Behind Aircraft and Peer Pressure both characterize pilots with hazardous attitude identified as Resignation. Table 2. Overview behavioral traps as defined by FAA [4]. Behavioral Trap Peer Pressure Mind Set Get-There-Itis Duck-Under Syndrome Scud Running Continuing Visual Flight Rules (VFR) into Instrument Conditions Getting Behind Aircraft Loss Positional/Situational Awareness Operating Without Adequate Fuel Reserves Descent Below Minimum En Route Altitude (MEA) Flying Outside Envelope Neglect Flight Planning, Preflight Inspections, and Checklists Definition Poor decision-making may be based upon an emotional response to peers, rar than evaluating a situation objectively. A pilot displays Mind Set through an inability to recognize and cope with changes in a given situation. This disposition impairs pilot judgment through a fixation on original goal or destination, combined with a disregard for any alternative course action. A pilot may be tempted to make it into an airport by descending below minimums during an approach. A pilot may believe that re is a built-in margin error in every approach procedure, or a pilot may not want to admit that landing cannot be completed and a missed approach must be initiated. This occurs when a pilot tries to maintain visual contact with terrain at low altitudes while instrument conditions exist. Spatial disorientation or collision with ground/obstacles may occur when a pilot continues VFR into instrument conditions. This can be even more dangerous if pilot is not instrument rated or current. This pitfall can be caused by allowing events or situation to control pilot actions. A constant state surprise at what happens next may be exhibited when pilot is Getting Behind Aircraft. In extreme cases, when a pilot gets behind aircraft, a loss positional or situational awareness may result. The pilot may not know aircraft s geographical location or may be unable to recognize deteriorating circumstances. Ignoring minimum fuel reserve requirements is generally result overconfidence, lack flight planning, or disregarding applicable regulations. The Duck-Under Syndrome, as mentioned above, can also occur during en route portion an Instrument Flight Rules (IFR) flight. The assumed high-performance capability a particular aircraft may cause a mistaken belief that it can meet demands imposed by a pilot s overestimated flying skills. A pilot may rely on short- and long-term memory, regular flying skills, and familiar routes instead established procedures and published checklists. This can be particularly true experienced pilots. Cognitive biases can also distort decision making and lead to an aviation disaster [5]. There are certain pilot cognitive biases that may affect safety flight. Dismukes, Berman and Loukopoulos [6] analyzed 19 major U.S. accidents between 1991 and 2000 in which NTSB identified crew error as a causal factor. Various common cognitive bias mes emerged from this study; nine accidents were result, or influenced by, plan continuation bias, a tendency to remain fixed on pre-determined course action or destination. They [6] also noticed that crews succumbed to increasing workload and were unable to perform tasks well once flight demands intensified. This could be related to confirmation bias, a tendency to focus on information that confirms our beliefs when too much information is available. Finally, four out 19 accidents showed that pilots deviated from explicit guidance or SOPs. The pilot cognitive biases and operational errors found during this [6] study resemble FAA behavioral traps.

4 Safety 2018, 4, Crew Resource Management and Behavioral Traps CRM is an FAA-mandated pressional training provided by air carriers to assist captains and first ficers in ir use all resources (e.g., human, hardware and stware). CRM is epitome or ultimate expression teamwork between flight crewmembers prior, during and after a flight. Good CRM practices are predicated on following checklists, Standard Operating Procedures (SOPs), conducting good preflight action, maintaining open communication and engaging in proper flight planning to prepare for unexpected events during flight. Initially, concept was known as cockpit resource management; however, CRM programs evolved to include flight attendants, maintenance personnel, dispatchers and ors [7]. The current definition includes all groups routinely working toger with flight crew who are involved in decision making processes required for safe operation flight. CRM training is one way to address challenge improving human/machine relationship by using checklists, resources and written procedures and accompanying interpersonal activities [1]. Advisory Circular e [8] is ficial FAA manuscript that provides guidance to air carriers on implementing CRM training. The major topics within a typical CRM training program are: (a) communications processes; (b) decision behaviors; (c) team building and maintenance; (d) workload management; (e) and situation awareness. Unfortunately, FAA does not provide specific guidelines towards attitude management training, nor does it provide much information to counteract hazardous attitudes, behavioral traps or various cognitive biases pilots are usually confronted with. Attitude management is defined by FAA [1] as ability to recognize hazardous attitudes in oneself and willingness to modify m as necessary through application an appropriate antidote thought. The most recent generation CRM focused on Threat and Error Management (TEM) [9]. During TEM, aircrews apply risk management strategies to avoid, trap, and mitigate errors [10]. However, for pilots, recognizing self-attitudes or personality threats that are hazardous to flight safety is not easy, albeit a necessary task during CRM Aims and Objectives This research sought to reveal presence behavioral traps in U.S. Title 14 Code Federal Regulations (CFR) Part 121 airline domain. The study led to a greater understanding how behavioral traps affect team dynamics in cockpit and a specific understanding how behavioral traps affect aeronautical decision making and ultimately flight safety. In addition, knowledge se behavioral traps in crews can influence portions CRM training to include hazardous behavior identification and modification techniques. Several studies [6] have suggested presence many unsafe pilot attitudes during airplane accidents that impair judgment. However, no published study had examined presence behavioral traps within crew operated flights. Instead, studies concerning unsafe pilot behavior have been mostly limited to single-pilot and/or General Aviation (GA) domain. Wher it is behavioral traps or or decision making imperfections (e.g., hazardous attitudes, cognitive biases), it is important to continue to explore limitations human performance so that appropriate safety interventions may be revised (i.e., CRM) or, if necessary, devised anew. 2. Materials and Methods This study utilized archival research methods to explore behavioral traps contributing to flight crew accidents. This research drew from population 34 NTSB U.S. Aviation Accident Reports (AARs) and factual reports attributed to flight crew error from 1991 to The reports were used to explore exclusively commercial Part 121 flight crew-related accidents. Purposive, also known as judgmental, sampling was used to study only those NTSB accident reports where flight crew error was a causal factor.

5 Safety 2018, 4, There are various reasons for selecting time frame specified. First, beyond 1991, NTSB has consistently generated factual reports in its analysis aviation accidents. Second, years 1991 to 2013 were selected because vast majority factual reports had already been upgraded from preliminary to final status. The accident reports were downloaded from Embry-Riddle Aeronautical University s website collection NTSB AARs. Third, beginning in 1991, CRM training had been established successfully at most U.S. airlines and was maturing [9]. An analysis behavioral traps during this period, albeit to a very limited degree, (indirectly) highlighted successes or shortcomings with such a training program Sources Data The data collected from NTSB had to conform to following criteria: U.S. 14 CFR Part 121 airline accidents that were partly or wholly attributed to flight crew error. In addition, accident must have involved death, serious injury or substantial damage to aircraft. Excluded from consideration were accidents with undetermined causes and those that were attributed to sabotage, suicide or criminal activity such as hijacking. Accidents attributed to maintenance issues are only included if improper crew decisions contributed to accident Data Collection A team composed four Certified Flight Instructors (CFIs) served as Subject Matter Experts (SMEs) and coded data. Furrmore, se SMEs also had an Airline Transport Pilot (ATP) certificate. The possession a flight instructor certificate and an ATP certificate ensured that all SMEs have (1) been exposed to commercial operations and (2) taught concepts hazardous attitudes and behavioral traps to students. All SMEs became thoroughly familiar with behavioral traps by receiving necessary instruction on se unsafe pilot behaviors before data were coded and analyzed. Familiarization training and evaluation sessions included case studies with examples behavioral trap categorization techniques. After training, aircraft accidents were randomly assigned to SMEs such that at least two different SMEs independently analyzed each accident. The SMEs used a subset FAA-defined list behavioral traps (see Table 2) to classify unsafe pilot behaviors. In past, Jeppesen [11] had categorized list behavioral traps among: (1) General Aviation (GA); (2) instrument-rated; and (3) commercial pilots. Because U.S. Part 121 airline flights are conducted under Instrument Flight Rules (IFR), this study used commercial and instrument behavioral traps. For ease understanding and because similarities in ir definitions, behavioral traps Descent Below Minimum En Route Altitude (MEA) and Duck-Under Syndrome were merged into a new one called Unauthorized Descent Below an IFR Altitude. Likewise, this study will follow Jeppesen [11] recommendation to combine behavioral traps Loss Positional/Situational Awareness with Getting Behind Aircraft. This combination resulted in behavioral trap called Loss Situational Awareness. In summary, following list behavioral traps was studied: (1) Peer Pressure; (2) Get-There-Itis; (3) Unauthorized Descent Below an IFR Altitude; (4) Loss Situational Awareness; and (5) Neglect Flight Planning, Preflight Inspections, and Checklists. Having said this, SMEs were alerted to lookout for presence or GA or VFR behavioral traps such as Flying Outside Envelope. Table 3 illustrates behavioral traps under consideration. Using narrative data obtained from NTSB as first step in analysis, SMEs used AARs and Factual Reports to assemble a chain events for each accident. It was necessary for both SMEs to agree on what constituted an event, sequence events, pilot actions associated with those events and how events affected decision making and CRM. Next, SMEs examined reports for evidence pilot actions related to behavioral traps. Again, both SMEs had to agree on what constituted a pilot action and if that action was reflective a behavioral trap. Any incongruities among SMEs were dealt with through integrative sessions. These sessions allowed ideas and notes

6 Safety 2018, 4, to be cross-compared with or SME that shared same NTSB report. This final act led to new observations and/or linkages, which resulted in revisions in data collection process. Table 3. Justification for inclusion or exclusion behavioral traps for this study. GA, General Aviation. Behavioral Trap Peer Pressure Mind Set Get-There-Itis Duck-Under Syndrome Scud Running Continuing Visual Flight Rules (VFR) into Instrument Conditions Getting Behind Aircraft Loss Positional/Situational Awareness Operating Without Adequate Fuel Reserves Descent Below Minimum En Route Altitude Flying Outside Envelope Neglect Flight Planning, Preflight Inspections, and Checklists Determination One commercial behavioral traps according to Jeppesen [11]. Therefore, it was included in this study One VFR or GA behavioral traps according to Jeppesen [11]. Therefore, it was not included in this study. One commercial behavioral traps according to Jeppesen [11]. Therefore, it was included in this study One instrument behavioral traps according to Jeppesen [11]. Because Part 121 operations are conducted under IFR, it was included in this study. However, this concept was merged with Descent Below Minimum En Route Altitude (see below) into a new behavioral trap called: Unauthorized Descent Below an IFR Altitude. One VFR or GA behavioral traps according to Jeppesen [11]. Therefore, it was not included in this study. One VFR or GA behavioral traps according to Jeppesen [11]. Therefore, it was not included in this study. Jeppesen merged this commercial behavioral trap with Loss Situational Awareness (see below) and for good reason. The FAA [1] explains that, in extreme cases, when a pilot gets behind aircraft, a loss positional or situational awareness may result. One commercial behavioral traps according to Jeppesen [11]. Therefore, it is included in this study One VFR or GA behavioral traps according to Jeppesen [11]. Therefore, it was not included in this study. One instrument behavioral traps according to Jeppesen [11]. Because Part 121 operations are conducted under IFR, it is included in this study. However, this trap has been merged with Duck-Under Syndrome (see above) into a new behavioral trap called: Unauthorized Descent Below and IFR Altitude. One VFR or GA behavioral traps according to Jeppesen [11]. Therefore, it was not included in this study. One commercial behavioral traps according to Jeppesen [11]. Therefore, it was included in this study The goal this research project was to discover what behavioral traps may exist in Part 121 accident world and how y manifest. Although behavioral traps were used as a priori codes, coding process SMEs included thorough review each NTSB report for key text identifying operational errors that led to behavioral traps mselves. The NTSB reports were independently coded and subsequently cross-checked to develop consistency in coding (inter-rater consensus). The coded text passages within NTSB document, although done manually by pilot experts, was subsequently entered into a qualitative data analysis stware called QSR NVivo Treatment Data The treatment data included descriptive statistics demonstrating prevalence each behavioral trap across aviation accidents and most prominent contributing factors, as well. However, with NVivo, data transitioned beyond descriptions coded text to an analysis associations, comparisons and pattern identifications. Comparing and contrasting data may

7 Safety 2018, 4, reveal systems, relationships and processes that could not be discovered in manual coding stage. When investigating relational patterns, researchers explore specific connections between pairs codes in data, some by building a data narrative and ors by examining relationships between categories within data [12] Reliability and Validity Data To assess quality qualitative portion a study, researchers may employ four tests named: credibility, dependability, transferability and compatibility [13]. Some se tests were more applicable to this archival design than ors. Credibility refers to accuracy; descriptions events or aviation accidents and antecedents must be plausible and recognizable. Credibility was achieved by including or investigators in study, namely SMEs. Dependability is more suited to interviews or observational research and refers to extent to which anor researcher, with similar training and rapport with participants, makes same observations. Although not entirely applicable to archival studies, SMEs still cross-checked ir coding process against each or and, in similar fashion, gained dependability. Transferability refers to generalizability study findings to or settings, populations and context. This test quality is usually one weakness studies involving qualitative methods; however, sufficient details regarding methodology procedures allow readers to assess this. Transferability is akin to external validity. The interpretations and conclusions this study could be applicable to most U.S. air carrier environments, resulting in good external validity [14]. Finally, confirmability refers to objectivity data or how much anor researcher agrees with meaning data. This was achieved by three methods: (1) a team (all SMEs) coded and categorized narratives NTSB reports; (2) SMEs had a coding scheme from which to work (i.e., subset FAA list behavioral traps); and (3) inter-coder consensus: SMEs reconciled any differences found during coding process by comparing ir work against each or. The behavioral traps are well-defined and well-known FAA terms. The NTSB accident reports fered an accessible account that included facts, findings causal factors and recommendations. The focus this research was on human (pilot) experience as recorded by NTSB accounts; archives provided deep insights that were not possible with quantitative methods. The SMEs all had similar backgrounds as pressional pilots and flight instructors and were exposed, through flight experience and pressional training, to concepts unsafe behaviors by pilots. 3. Results 3.1. Descriptive Statistics As described in previous section, for each case, assigned SMEs determined which behavioral traps were present. Every SME found a minimum one behavioral trap and a maximum four throughout analysis. The average number behavioral traps was two with a standard deviation In addition, during coding process, researcher asked SMEs to identify actions representative behavioral traps and any contributing factors that may have influenced outcome flight. Figure 1 shows frequency with which behavioral traps were present in all aviation accidents. Figure 2 displays most prevalent traps in fatal accidents. Please use following descriptors for understanding: LSA = Loss Situational Awareness, Neg = Neglect Flight Planning, Preflight Inspections, and Checklists, PP = Peer Pressure, Get = Get-There-Itis, Des = Unauthorized Descent Below an IFR Altitude. The behavioral traps Loss Situational Awareness and Neglect Flight Planning, Preflight Inspections, and Checklists were overwhelmingly dominant, even throughout fatal accidents. Peer Pressure, Get-There-Itis and Descent Below an IFR Altitude were each found in six accident reports

8 Safety 2018, 4, Safety 2018, 4, 2 Safety 2018, 4, or Flight fewer. Planning, The prominence Preflight Inspections, Loss Situational and Checklists, Awareness PP in fatal Peer accidents Pressure, is unsurprising Get Get-There-Itis, given that it Flight has been Planning, Des Unauthorized well established Preflight Inspections, Descent that Below situational and an IFR Altitude. awareness Checklists, ten PP = implicated Peer Pressure, in aviation Get = accidents. Get-There-Itis, Des = Unauthorized Descent Below an IFR Altitude. Behavioral Traps Total LSA Neg PP Get Des LSA Neg PP Get Des Figure 1. Frequency count all behavioral traps found. LSA Loss Situational Awareness, Figure 1. Frequency count all behavioral traps found. LSA = Loss Situational Awareness, Neg Neglect Flight Planning, Preflight Inspections, and Checklists, PP Peer Pressure, Neg = Neglect Flight Planning, Preflight Inspections, and Checklists, PP = Peer Pressure, Get Get-There-Itis, Des Unauthorized Descent Below an IFR Altitude. Get = Get-There-Itis, Des = Unauthorized Descent Below an IFR Altitude. Figure 2. Behavioral traps in fatal accidents only. Figure 2. Behavioral traps in fatal accidents only. The behavioral traps Loss Situational Awareness and Neglect Flight Planning, Preflight Inspections, Aviation The behavioral and accidents traps Checklists are generally Loss were overwhelmingly Situational result aawareness series dominant, simultaneous and Neglect even throughout or consecutive Flight Planning, fatal circumstances Preflight accidents. that Inspections, Peer each Pressure, add and operational Checklists Get-There-Itis risk; were and seldom overwhelmingly Descent is a single Below isolated dominant, an IFR cause Altitude identified. even throughout were each During found fatal in analysis accidents. six accident coding Peer Pressure, reports process Get-There-Itis or fewer. performed The prominence by and SMEs, Descent Loss various Below Situational mes an IFR Awareness began Altitude to emerge, were each in fatal which found accidents played in six is unsurprising significant accident roles reports given inthat many or fewer. it accidents. The prominence has been These well topics Loss established were airline Situational that management, Awareness situational awareness CRM in fatal issues, accidents ten fatigue is implicated and unsurprising a former in behavioral given that aviation accidents. trap it called has been Flying well Outside established Envelope. that situational This latter trap awareness was left is out ten implicated current study in due aviation to Aviation Jeppesen s accidents. accidents [11] categorization are generally behavioral result traps series among commercial, simultaneous instrument-rated or consecutive and circumstances general Aviation aviation accidents that each pilots. are add operational However, generally during result risk; seldom training, a is single series SMEs isolated were simultaneous cause alerted identified. to or existence consecutive During all circumstances that each add operational risk; seldom is a single isolated cause identified. During analysis behavioral traps coding andprocess were told performed to flag m by if SMEs, y saw various ir mes presence began among to Part emerge, 121 which pilots. Figure analysis coding process performed by SMEs, various mes began to emerge, which played 3significant shows roles distribution in many accidents. se contributory These topics factors. were For airline easemanagement, understanding, CRM issues, please observe played significant roles in many accidents. These topics were airline management, CRM issues, fatigue and following: former behavioral Airline = trap airline called management Flying Outside issues, CRM Envelope. issues This = lack latter or trap poorwas practice left out Crew fatigue and a former behavioral trap called Flying Outside Envelope. This latter trap was left out Resource current study Management, due to Jeppesen s Envelope [11] = categorization behavioral trap called behavioral Flying Outside traps among Envelope commercial, and Fatigue current study due to Jeppesen s [11] categorization behavioral traps among commercial, instrument-rated = accidents and where general fatigue aviation was pilots. a contributing However, factor. during training, SMEs were alerted to instrument-rated and general aviation pilots. However, during training, SMEs were alerted to existence all behavioral traps and were told to flag m if y saw ir presence among Part 121 existence all behavioral traps and were told to flag m if y saw ir presence among Part 121 pilots. Figure shows distribution se contributory factors. For ease understanding, please pilots. Figure 3 shows distribution se contributory factors. For ease understanding, please

9 Safety 2018, 4, observe following: Airline = airline management issues, CRM issues = lack or poor practice Crew Resource Management, Envelope = behavioral trap called Flying Outside Envelope and Fatigue = accidents where fatigue was a contributing factor. Safety 2018, 4, Contributing Factors Airline = 16 CRM issues = 31 Envelope = 9 Fatigue = 7 Figure Or contributing factors factors in connection in connection with with Part 121 Part accidents. 121 accidents. CRM, Crew CRM, Resource Crew Management. Resource Management Qualitative Data The following section describes, in comprehensive fashion, how each unsafe behavior is manifested within pilots aviation accidents. For sake avoiding repetition similar pilot actions, only most representative examples (i.e., (i.e., NTSB NTSB report report passages) are are illustrated for for top top two two behavioral traps traps found, found, as as well well as as for for remaining traps Peer Pressure, Unauthorized Descent Below an IFR Altitude and and Get-There-Itis. Because an an unexpected trap trap had had a presence a presence in Part in 121 Part accidents, 121 accidents, results results are also are shown also shown for additional for additional behavioral behavioral trap Flying trap Outside Flying Outside Envelope. The Envelope. pilot actions The pilot representative actions representative behavioral traps behavioral are illustrated traps are using illustrated tables that using aretables immediately that are followed immediately by word followed frequency by word queries frequency calledqueries tags or called word clouds. tags or word clouds. A tag or word cloudis isa a visual representation textual data data highlighting importance most most commonly-used words words within within a source a source (e.g., (e.g., document, document, interview). interview). Each Each cloud cloud shows shows most most frequently-used frequently-used words words by increasing by increasing its font itssize font and sizeplacing and placing those words those words nearer nearer center center cloud. cloud. Word Word tag clouds or tag clouds are very areuseful very for useful quickly for quickly perceiving perceiving most prominent most prominent term and term its and relative its relative prominence prominence compared compared to ors to used orswithin used within a source. a source. The relative The relative font size font indicates size indicates which which words words were were coded coded most most commonly commonly throughout sources. sources. This This last last action was performed to investigate common or or emerging mes within traps traps mselves and and contributory factors factors that arose that arose when when SMEs SMEs began began to analyze to analyze data. data The Top Behavioral Traps Neglect Flight Planning, Preflight Inspections, and and Checklists Checklists was was found found in 27 in (79%) 27 (79%) 34 cases. 34 Acases. closer A examination closer examination this behavioral this behavioral trap reveals trap that reveals morethat thanmore one action than one is being action accounted is being for. accounted Any occasion for. Any where occasion where pilot would pilot deliberately would deliberately or unconsciously or unconsciously bypass a procedure, bypass a procedure, checklist, inspection checklist, inspection or flight planning or flight process, planning process, SME team would SME team codewould this behavior code this under behavior this node under within this NVivo. node within Table 4NVivo. lists illustrations Table 4 lists pilot illustrations actions that pilot exemplify actions Neglect that exemplify Flight Planning, Neglect Preflight Flight Inspections, Planning, Preflight and Checklists Inspections, found and inchecklists aviationfound accidents. in aviation accidents. Figure 4 shows a frequency query tag cloud illustrating prominence specific words within sources analyzed: NTSB AARs and NTSB factual reports. The most frequently-used words for this behavioral trap were captain, airplane, procedures, approach and checklist. Because this behavioral trap takes more than one action into consideration, most frequently-used words align well with concept or behavioral trap itself. The word captain might signal that he/she is flight crewmember that is initiating or prompting behavioral trap. The word airplane might be used in connection to airplane procedures or airplane checklists, which are also top words found in query. Finally, inclusion word approach might suggest that it is during approach and landing phase where most manifestations this trap are occurring.

10 Safety 2018, 4, Table 4. Neglect Flight Planning, Preflight Inspections, and Checklists. Accident Report Although airplane speed was within target range, airplane did not meet FedEx s criteria for a stabilized approach because its rate descent was greater than FedEx s recommended 1000 feet per minute (fpm). Investigation revealed that flight crew did not perform several appropriate checklists and interrupted an emergency fire-related checklist. Had captain complied with standard operating procedures in response to flap anomaly, unstabilized approach, stick shaker, and terrain awareness and warning system warning and initiated a go around maneuver, accident likely would not have occurred. The captain returned about 10 min after ficer, and neir m performed a walk around inspection airplane, nor were y required to do so by USAir procedures. He said that a norm existed for first ficer to make hydraulic system configuration changes; he was aware that this was not standard operating practice, which assigned task to pilot not flying at all times. He said he conducted his cockpit according to standard operating practice, because he was new to airplane, and he did not expect first ficers to configure hydraulic pumps. The abbreviated briefing was contrary to company policy, and Safety Board notes that it is prudent for pilots to fully conduct taxi briefings according to standard operating procedures. The first ficer stated that he thought that pilots were required to (and should) check landing distances with a contaminated runway. He said that he believed 4000 feet was required landing distance but indicated that y did not check landing distance charts. The reason first ficer retracted flaps and suggested raising gear could not be determined from available information, se actions were inconsistent with company stall recovery procedures and training. I ll just do a quick procedure turn headed back in, so I m not going to straighten out on thing, localizer, just teardrop and come right back around and land. The FO [First Officer] simply responded OK. The captain actively moved power levers from flight idle gate into beta range for undetermined reasons. Operation propellers in beta range while in flight is prohibited by airplane flight manual. About 1912:02, captain transmitted a burp over ARTCC [Air Route Traffic Control Center] radio frequency that would have been heard by or pilots and air traffic controllers. An unknown voice on radio frequency responded to captain s burp, stating, nice tone, and [cockpit voice recorder] CVR recorded accident pilot s chuckling. About 1912:53, captain talked about deliberately dropping a flight manual on a passenger whose foot had intruded into cockpit. The first ficer engaged in banter with captain, and both pilots used informal, nonstandard terminology during flight. The captain was not authorized under COM to allow first ficer to fly airplane. The captain told investigators that he was not familiar with section COM that indicated that he was not supposed to share flying duties with first ficer. Peer Pressure was found in 6 (18%) 34 cases. Peers are important social components life. However, friends, colleagues, and associates may cloud judgment. The desire to conform to ors, to be accepted, and to be right are fundamental needs human beings. Peer Pressure can be obvious or subtle, verbal or non-verbal, intentional or unintentional, and its origin may be personnel or organizational [15 20]. Peer Pressure is a behavioral trap that affects decision making. Table 5 lists illustrations pilot actions that Peer Pressure.

11 Safety 2018, 4, Safety 2018, 4, Figure 4. Tag cloud helps visualize word query for Neglect Flight Planning, Preflight Inspections, Figure 4. Tag cloud helps visualize word query for Neglect Flight Planning, Preflight and Checklists. Inspections, and Checklists. Table 5. Peer Pressure. Peer Pressure was found in 6 (18%) 34 cases. Peers are important social components life. However, friends, colleagues, and associates may cloud judgment. The desire to conform to ors, Accident Report to be accepted, and to be right are fundamental needs human beings. Peer Pressure can be obvious Fifty feet, first ficer stated, I m gonna go around. The captain stated, No, no, no I or subtle, verbal or non-verbal, intentional or unintentional, and its origin may be personnel or got it The first ficer responded, You got airplane As first ficer said organizational [15 20]. world Peer airplane. Pressure Theis automatic a behavioral voice said trap thirty. that affects The captain decision tookmaking. control and Table landed 5 lists illustrations pilot actions airplane. that Peer Pressure. Following application power, airspeed began increasing. At 0435:40, first ficer asked, should Table I go 5. around, and Peer captain Pressure. replied, no, and n stated, keep descending. Accident Report The Fifty captain feet, became first ficer overly stated, reliant I m on gonna first go ficer. around. This contributed The captain to stated, runway No, no, incursion. The captain knew re was something wrong, he even questioned but no I got it The first ficer responded, You got airplane As first ficer said acknowledged FO s instructions. world airplane. The automatic voice said thirty. The captain took control and After landed hearing airplane. wear observation, captain commented, we re not getting in...we don t have an ILS [instrument landing system]. The first ficer responded, I Following application power, airspeed began increasing. At 0435:40, first know...go all this [expletive] way. Well, let s try it. The captain responded, yeah, we ll ficer asked, should I go around, and captain replied, no, and n stated, try it. About 30 s later, captain said, I don t want to...go all way out here for nothing keep descending. tonight, and moments later said, I ll be so happy when we have an ILS everywhere The captain we became go. The overly firstreliant ficer concurred, first and ficer. This captain contributed continued, to I thought runway we were incursion. gonnathe havecaptain it easyknew tonight. re was something wrong, he even questioned but acknowledged FO s instructions. The first ficer told Safety Board investigators that his goal after recognizing that flaps After were hearing not extended wear was observation, to get captain captain to initiate commented, a go-around. we re Thirty not getting seconds before in...we touchdown, don t have an ILS first[instrument ficer stated landing wantsystem]. to take it around? The first ficer and responded, captain I replied know...go no all that s this alright. [expletive] * keep way. your Well, speed let s uptry here it. about The captain uh. When responded, captain yeah, denied we ll first try ficer s it. About request 30 s to later, go around captain and said, told him I don t to keep want histo...go speedall up, way firstout ficer here did for nothing challenge tonight, captain s and moments statement. later said, He also I ll did be not so happy question when we captain have to an ILS determine everywhere hiswe reason(s) go. The forfirst continuing ficer concurred, approach. and The captain first ficer continued, stated that I thought re was we no were time gonna for discussion have it easy with tonight. captain because approach was so fast. The first ficer s failure to question captain s decision to continue approach was The first ficer told Safety Board investigators that his goal after recognizing that inconsistent with CRM training he had received that emphasized importance sharing flaps were doubts not with extended or was crewmembers to get captain and quickly to initiate resolving a go-around. conflicts. Thirty seconds before touchdown, first ficer stated want to take it around? and captain During replied no course that s alright. performing * keep your missed speed approach up here about procedure, uh. When first ficer captain acted, without challenge, to a command from captain to down, push it down. denied first ficer s request to go around and told him to keep his speed up, first ficer did not challenge captain s statement. He also did not question captain to determine his reason(s) for continuing approach. The first ficer stated that re was no time for discussion with captain because approach was so

12 Safety 2018, 4, fast. The first ficer s failure to question captain s decision to continue approach was inconsistent with CRM training he had received that emphasized importance sharing doubts with or crewmembers and quickly resolving conflicts. Safety 2018, 4, 2 During course performing missed approach procedure, first ficer acted, without challenge, to a command from captain to down, push it down. Figure Figure 5 5 shows shows a a frequency frequency query query tag tag cloud cloud illustrating illustrating prominence prominence specific specific words words within within sources sources analyzed. analyzed. For For Peer Peer Pressure, Pressure, most most frequently-coded frequently-coded words words were were captain, captain, first, first, ficer, ficer, stated stated and and airplane. airplane. Once Once again, again, captain captain was was one one top top words. words. As previously As previously mentioned, mentioned, word captain word might captain signal might that signal he/she that ishe/she flight is crewmember flight crewmember that is initiating that is initiating or prompting or prompting behavioral trap. behavioral However, trap. captains However, might captains also might be pressured also be pressured by ir first by ir ficer first colleague. ficer colleague. Because Because re is a relationship re is a relationship developing between developing se between two crewmembers, se two crewmembers, it is not surprising it not that surprising any instance that any Peer Pressure instance will Peer result Pressure in manifestations will result in by manifestations captain or by first captain ficer, or hence first ficer, tophence two words top found. two The words inclusion found. The word inclusion stated must word be due stated to must fact be that due any to example fact that Peer any Pressure example begins Peer or ends Pressure with abegins verbal or statement ends with by a verbal one statement se flight by crewmembers. one se flight Thecrewmembers. result is an action The result that changes is action airplane s that changes status; hence, airplane s word status; airplane. hence, word airplane. Figure Figure Tag Tag cloud helps visualize word queryfor forpeer Pressure. Get-There-Itis is is known in some in some textbooks textbooks and and ficial ficial FAA FAA documents documents as Get-Home-Itis. as Get-Home-Itis. During Get-There-Itis, During Get-There-Itis, pilot s pressure pilot s pressure comes comes from from within within (i.e., (i.e., him/herself), and obstinacyis is specifically related to to arriving at destination. Table 6 lists several illustrative instances Get-There-Itis. This behavioral trap was found in five (15%) 34 cases. Table Get-There-Itis. Accident Report The cockpit voice recorder indicated that flight crew had discussed wear and The cockpit voice recorder indicated that flight crew had discussed wear and needed to expedite Approach. The captain stated we got to get over re quick. needed to expedite Approach. The captain stated we got to get over re quick. The pilots failed to conduct a prestart checklist properly and, subsequently, failed to observe The pilots failed illuminated to conduct light on a prestart annunciator checklistpanel. properly A second and, subsequently, opportunity to failed detect to observe status illuminated pitot heat light knob on was annunciator annunciator panel. panel Acheck second just opportunity before takef. to detect In this case, status first pitot ficer heat called knob checklist was items annunciator without panel captain s check request just before and takef. without In this case, first ficer called checklist items without captain s request and without using normal challenge and response procedures as airplane was being taxied into using normal challenge and response procedures as airplane was being taxied into position for takef. The pilots appeared to be rushed, and re is no evidence that position for takef. The pilots appeared to be rushed, and re is no evidence that first ficer actually observed annunciator panel. first ficer actually observed annunciator panel. The captain stated man we re almost speed heat here, two sixty four or two sixty three The captain sixty stated two he man said we re gosh, almost we gonna come speeddown. heat here, two sixty four or two sixty three... sixty two he said gosh, we gonna come down. The flight crew ignored present wear conditions and continued approach to land even during unestablished approach, Captain took flight controls at last moment when it was too late to correct or execute a go around. The Captain fixated on landing airplane with a disregard for any alternative course action such as performing a go-around.

13 Safety 2018, 4, The flight crew ignored present wear conditions and continued approach to land even during unestablished approach, Captain took flight controls at last moment when it was too late to correct or execute a go around. Safety 2018, 4, 2 The Captain fixated on landing airplane with a disregard for any alternative course action such as performing a go-around. Passages such as as we got to to get re over quick, pilots appeared to to [be] [be] rushed, man man we re we re almost speed heat here, crew ignored present wear and continued approach, and and finally, captain fixated on on landing airplane with disregard for any alternative course are are all all signs signs Get-re-Itis. Figure 6 shows that most frequently-coded words for for this this behavioral trap trap were were captain, captain, flight, flight, airplane, crew, crew, approach, wear and and landing. The The stubbornness reaching destination destination airport airport at all at costs, all costs, even even if thisif means this means disregarding disregarding wear wear conditions, conditions, by by both both crew members crew members resulted resulted in inclusion inclusion words captain, words captain, wear wear and crew. and It seems crew. It asseems it is during as it is approach during andapproach landing phase and landing wherephase this behavioral where this behavioral trap is most trap routinely is most routinely manifested; manifested; hence, hence, words approach words and approach landing. and landing. Figure 6. Word frequency tag cloud for Get-There-Itis. Figure 6. Word frequency tag cloud for Get-There-Itis. Unauthorized Unauthorized Descent Descent Below Below an IFR an Altitude IFR Altitude was found was found in five (15%) in five (15%) 34 reports. 34 A reports. combination A combination previously-named previously-named behavioral traps behavioral Descent traps Below Descent MEA Below and MEA Duck-Under and Duck-Under Syndrome, Syndrome, concept is evidenced concept is when evidenced a pilotwhen descends a pilot below descends minimum below altitudes minimum during altitudes during en route phase en route flight or phase during flight approach or during stage before approach obtaining stage visual before contact obtaining with visual environment, contact with eir due to environment, a belief that re eir isdue a built-in to a belief margin that re error is a inbuilt-in every margin approach error procedure in every or approach a refusal to procedure admit that or landing a refusal cannot to admit be safely that accomplished landing cannot andbe a safely missedaccomplished approach must and be a missed initiated. approach Table 7 must highlights be several initiated. illustrations Table 7 highlights pilot actions several that illustrations exemplify pilot behavioral actions trap that known exemplify as Unauthorized behavioral Descent trap Below known an as IFRUnauthorized Altitude. Descent Below an IFR Altitude. Table Table Unauthorized Descent Belowan anifr IFR Altitude. Accident Accident Report Report The airplane continued to descend at 1500 fpm and passed through and continued The airplane continued to descend at 1500 fpm and passed through and continued below desired below glidepath. desired glidepath Captain descended below 3000 feet prior to establishing airplane on final approach Captain descended below 3000 feet prior to establishing airplane on final approach course, course, contrary to directions on approach plate and establish requirements [ ] contrary to directions on approach plate and establish requirements [... ] captain did not have captain runway did not environment have runway in sight environment when he told in sight first when ficer he told continue first ficer approach. to continue approach. The captain, while acting as pilot flying, descended below MDA [minimum descent altitude] after supposedly having approach lights in sight. The airplane struck trees and crashed just short runway The MDA for localizer (glideslope out) approach to runway 28 was 429 feet [above ground level] agl. No CVR evidence or post-accident interview information indicated that eir crewmember had runway environment in sight by that altitude. The captain initially did not recognize descent below MDA, and he failed to react immediately when he was alerted to altitude deviation by first ficer.

14 and crashed just short runway. The MDA for localizer (glideslope out) approach to runway 28 was 429 feet [above ground level] agl. No CVR evidence or post-accident interview information indicated that eir crewmember had runway environment in sight by that altitude. Safety 2018, 4, 2 The captain initially did not recognize descent below MDA, and he failed to react immediately when he was alerted to altitude deviation by first ficer. Descending below glide path, glideslope or or minimum descent altitude (MDA) without establishing visual contact with airport environment are areall allexamples Unauthorized Descent below an anifr IFRAltitude. Figure 77demonstrates most frequently coded words within this thisbehavioral trap. These words were airplane, MDA, approach, captain, descended and andrunway. Because crew commands airplane to toan analtitude that thatis isdangerously low, low, it itis isnot notsurprising to tosee see word airplane among most frequently-found words. As Aswith previous behavioral traps, word captain appears once again again within within most most common common words. words. It is It is captain captain that hasthat final has authority final authority as to operation as to operation flight. Theflight. appearance The appearance word descended word descended is no surprise no considering surprise considering that nature that nature trap itself is trap a descent itself below is a descent an altitude. below Because an altitude. objective Because objective crew is to make crew visual to contact make visual with contact runway with or airport runway environment, or airport it environment, not surprisingit tonot seesurprising word runway to see among word runway top words among in this query. top words Finally, MDA this query. standsfinally, for minimum MDA stands descentfor altitude. minimum Thisdescent altitude altitude. was repeatedly This altitude descended was repeatedly through without descended adequate through visual without contactadequate with runway; visual contact hence, with MDA as runway; one hence, mostmda common as one words. most common words. Figure 7. Word frequency tag cloud for Unauthorized Descent Below an IFR Altitude. Figure 7. Word frequency tag cloud for Unauthorized Descent Below an IFR Altitude. Loss Situational Awareness was found in 25 (74%) 34 cases. The FAA [1] explains that, Loss Situational Awareness was found in 25 (74%) 34 cases. The FAA [1] explains that, in in extreme cases, when a pilot gets behind aircraft, a loss positional or situational awareness extreme cases, when a pilot gets behind aircraft, a loss positional or situational awareness may may result. The pilot may not know aircraft s geographical location, or may be unable to recognize result. The pilot may not know aircraft s geographical location, or may be unable to recognize deteriorating circumstances. Coding this behavior under this node within NVivo involved recognizing deteriorating circumstances. Coding this behavior under this node within NVivo involved any signs spatial, geographic, operational or procedural disorientation. Situational awareness recognizing any signs spatial, geographic, operational or procedural disorientation. Situational includes full appreciation not only aircraft s physical position in space and time, but awareness includes full appreciation not only aircraft s physical position in space and time, correct procedures and ability to plan appropriate responses to real aircraft situation. Table 8 but correct procedures and ability to plan appropriate responses to real aircraft situation. shows some most illustrative examples Loss Situational Awareness. Table 8 shows some most illustrative examples Loss Situational Awareness. Table 8. Loss Situational Awareness. Accident Report Investigation determined that pilots have generally had little exposure to, and refore may not fully understand, effect large rudder pedals inputs in normal flight or mechanism by which rudder deflections induce roll on a transport category airplane. An unidentified voice in cockpit stated, a.... we are f course. In a post-accident interview, first ficer stated that he thought airplane was stabilized until about 400 feet above field elevation, at which point airplane drifted to right.

15 Safety 2018, 4, Table 8. Cont. Accident Report A review first ficer s medical record from his personal psychiatrist revealed that, in July 2001, he began seeing psychiatrist for treatment various anxiety-related symptoms. The psychiatrist prescribed alprazolam to treat first ficer condition. Common side effects alprazolam include drowsiness and light-headedness. Flight crew s failure to monitor and maintain a minimum safe airspeed while executing an instrument approach in icing conditions, which resulted in an aerodynamic stall at low altitude. About 1 2 min later, first ficer stated, guess we turn here. When captain expressed some doubt about this left turn, first ficer replied, Near as I can tell. Man, I can t see out here. A lack proper crew coordination, including a virtual reversal roles by DC9 pilots, which led to ir failure to stop taxing ir airplane and alert ground controller ir positional uncertainty in a timely manner before and after intruding onto active runway. He believed that snow had all but stopped and was more concerned about amount vehicular traffic, such as sweepers and plows, than he was about snowfall. Captain asked, what do you think?, and first ficer responded, I can t see [expletive]. About 2 s later, as airplane continued to descend, captain stated, yeah, oh re it is. Approach lights in sight. Almost immediately, GPWS annunciated two hundred feet. According to first ficer, captain reached up to overhead panel as GPWS was alerting. The captain did not recall doing this and stated that he had interpreted GPWS alerts as a high sink rate warning. The Captain decided to continue to land from an unstable approach without realizing gear was up and flaps were up. The result was a wheels up landing at Houston airport. About 1 min later, first ficer stated, something s messed up with this thing, and, at 0039:07, he asked why is this thing? At 0041:21, first ficer stated that control wheel felt funny. He added, feels like I need a lot force. It is pushing to right for some reason. I don t know why...i don t know what s going on. The first ficer n repeated twice that it felt like he needed a lot force. The CVR did not record captain responding to any se comments. The reason captain did not recognize impending onset stick shaker could not be determined from available evidence but that first ficer s tasks at time low-speed cue was visible would have likely reduced opportunities for her timely recognition impending event; failure both pilots to detect this situation was result a significant breakdown in ir monitoring responsibilities and workload management. The flight crew did not monitor fuel quantity gauges or respond properly to airplane s changing handling characteristics. Flight crew s failure to detect and remove ice contamination from wings was a causal factor in this accident. Phrases such as doubt, positional uncertainty, failure to detect, I don t know why, I don t know what s going on and did not recall doing are all examples not being aware. The tag cloud represented by Figure 8 shows that most frequently-coded words during analysis were captain, airplane, approach, first and ficer. These words have also frequently appeared as most common words with previous behavioral traps. Once again, it seems as though it is during approach and landing phase where crew confusion regarding airplane status is mostly being exemplified.

16 Safety 2018, 4, approach and landing phase where crew confusion regarding airplane status is mostly being exemplified. Safety 2018, 4, Figure 8. Word frequency tag cloud for Loss Situational Awareness. Figure 8. Word frequency tag cloud for Loss Situational Awareness. Flying Outside Envelope can range from pilot assuming an inappropriate level performance capability a particular aircraft, intentionally exceeding aircraft limits limits assuming re re is is a margin a margin safety safety built built into into aircraft aircraft or an or an overestimation pilot s pilot s flying flying skills skills that that causes causes flight flight to exceed to exceed aircraft s aircraft s structural structural and/or and/or aerodynamic limits. limits. In In any any case, case, pilot pilot allows allows or or causes causes aircraft aircraft to to exceed exceed its its design limitations. See See Table 99 for for occurrences this behavioral trap, which was found in seven (21%) 34 accidents. Table 9. Flying Outside Envelope. Accident Accident Report Report The probable cause this accident was in-flight separation vertical stabilizer as a The probable cause this accident was in-flight separation vertical stabilizer result loads beyond ultimate design that were created by first ficer s unnecessary as a result loads beyond ultimate design that were created by first ficer s and unnecessary excessive rudder and excessive pedal inputs. rudder pedal inputs Continuation Continuation approach approach to a to landing a landing when when company s company s max max crosswind crosswind was was exceeded exceeded and and use use reverse reverse thrust thrust greater greater than than engine engine per per ratio ratio after after landing. landing. The The excessive excessive vertical vertical and and lateral lateral forces forces on on right right main main landing landing gear gear during during landing exceeded landing those exceeded that those gear that was designed gear was to designed withstand to withstand resulted and in resulted fracture in outer fracture cylinder and outer collapse cylinder and right collapse main landing right gear. main landing gear The The captain s captain s calculations calculations showed showed airplane airplane outside outside acceptable acceptable weight weight and balance and balance limits limits. Table 9 reveals that re were many ways that a crew exceeded airplane design limits. This was exemplified by excessive rudder inputs, operations beyond company s maximum crosswind Table 9 reveals that re were many ways that a crew exceeded airplane design limits. This was limit, too much landing force on only one main landing gear and operations past acceptable exemplified by excessive rudder inputs, operations beyond company s maximum crosswind limit, weight and balance limitations. Figure 9 shows us word frequency tag cloud for this behavioral too much landing force on only one main landing gear and operations past acceptable weight and trap. The most commonly-found words were airplane, landing, captain, accident, approach, balance limitations. Figure 9 shows us word frequency tag cloud for this behavioral trap. The most company and exceeding. Because two se accidents occurred during approach and landing commonly-found words were airplane, landing, captain, accident, approach, company and exceeding. phase, it is not surprising to see both terms in list most frequently-used words. In order to Because two se accidents occurred during approach and landing phase, it is not surprising to operate outside envelope, a company or manufacturer limit must have been exceeded. Hence, see both terms in list most frequently-used words. In order to operate outside envelope, words company and exceeded. Finally, it should be no surprise that se actions all resulted in a company or manufacturer limit must have been exceeded. Hence, words company and exceeded. an accident. Therefore, query also provided word accident as one top words. Finally, it should be no surprise that se actions all resulted in an accident. Therefore, query also provided word accident as one top words.

17 Safety 2018, 4, Safety 2018, 4, Figure 9. Tag cloud for Flying Outside Envelope. Figure 9. Tag cloud for Flying Outside Envelope Additional Additional Contributing Contributing Factors Factors Such Such as as Airline Airline Management, Management, CRM CRM Issues, Issues, and and Fatigue Fatigue During During analysis, analysis, SMEs SMEs discovered discovered that that many many accidents accidents were were also also result result factors factors outside outside five five behavioral behavioral traps traps under under study. study. These These factors factors included included airline airline management, management, CRM CRM issues issues and and fatigue. fatigue. Airline Airline management management causes causes are are shown shown in in Table Table CRM CRM issues issues are are shown shown in in Table Table fatigue fatigue are are presented presented in in Table Table A word word frequency frequency query query within within NVivo NVivo explored explored se se issues issues furr. furr. Thus, Thus, Figures Figures display display word word frequency frequency tag tag clouds clouds for for each each contributing contributing factor, factor, respectively. respectively. Passages such as airplane was dispatched to ANC with left engine thrust reverser inoperative, company did nottable teach10. its pilots bounced airline landing management recovery as a techniques, contributing Maintenance factor. personnel use an inappropriate manual engine star procedure, and use QRH instead manual are all Accident Report indications airline management operational deficiencies that contributed to many airplane accidents. According to flight plan and release documents, airplane was dispatched to ANC James Reason [21] created term organizational accident to allude to those aviation mishaps that [airport code for Anchorage, Alaska] with left engine thrust reverser inoperative. were caused by latent failures within organization. These organizational errors when left alone Executive airline s manager training and standards stated that, before accident, and unattended to, company may lead did to not a teach catastrophic its pilots event bounced such landing as an recovery accident. techniques. The manager also The most commonly-used stated that he would words not found want to inconduct word bounced frequency landing query recovery for training airline management in issues were airplane, simulator procedures, because it company was very difficult and inspection. to demonstrate. Many company management issues, such as lack, or bad, Maintenance procedures personnel and inspections, use an inappropriate may lead tomanual airplane engine experiencing star procedure, a catastrophic which led to event09-03 such as aviation uncommanded accident; hence opening inclusion left engine air words turbine company, starter valve, procedures, and subsequent inspections, left and airplane. engine fire. The very existence The checklist behavioral deviations traps and signals or pilot a breakdown procedural indeficiencies CRM. When noted crews by fail FAA to recognize, during a do not monitor, exchange special inspection, flight controls which abruptly, included fail numerous to communicate en route inspections on ir situation about one and month are before unable accident, suggest that problems identified in this accident regarding improper to manage ir workload, a lack CRM is present. In some se passages, reader may detect checklist procedures were systemic at COA [Continental Airlines]. If pilots fail to adhere to a sense urgency or emotional intensification such as when expletives are used or when abnormal procedures during en route inspections by FAA inspectors, y most likely behave in a discussions are present. similar manner when no inspector is present. The words most The first commonly ficer stated coded that within when he CRM and issues his classmates were captain, questioned first, ficer, absence airplane and approach. Unsurprisingly, [manual], se Flight words Safety have International come up again simulator as instructors most frequently-used informed m words. that ValuJet Given that CRM is an issue wanted team m dynamics to use inside QRH [quick cockpit, reference handbook] words airplane, like a Bible first, ficer for abnormal and captain were96-07 expected. The procedures. word approach, The first ficer once again, indicated might that he indicate and his aclassmates pattern stopped CRMir breakdown first simulator in busiest, and arguably session most and critical, called phase company flight, to get that an ficial is determination approach and as landing to what guidance stage. y should use for abnormal and emergency procedures during routine flight operations; he stated that ValuJet management advised m to use QRH instead manual. The DC-9 Operations Manuals were basically developed by Ryan from airplane s previous owner s Operations Manuals, and certain purported Ryan practices were not incorporated into m. The requirement to conduct an exterior inspection airplane at

18 Safety 2018, 4, Table 10. airline management as a contributing factor. Accident Report According to flight plan and release documents, airplane was dispatched to ANC [airport code for Anchorage, Alaska] with left engine thrust reverser inoperative. Executive airline s manager training and standards stated that, before accident, company did not teach its pilots bounced landing recovery techniques. The manager also stated that he would not want to conduct bounced landing recovery training in simulator because it was very difficult to demonstrate. Maintenance personnel use an inappropriate manual engine star procedure, which led to uncommanded opening left engine air turbine starter valve, and subsequent left engine fire. The checklist deviations and or pilot procedural deficiencies noted by FAA during a special inspection, which included numerous en route inspections about one month before accident, suggest that problems identified in this accident regarding improper checklist procedures were systemic at COA [Continental Airlines]. If pilots fail to adhere to procedures during en route inspections by FAA inspectors, y most likely behave in a similar manner when no inspector is present. The first ficer stated that when he and his classmates questioned absence [manual], Flight Safety International simulator instructors informed m that ValuJet wanted m to use QRH [quick reference handbook] like a Bible for abnormal procedures. The first ficer indicated that he and his classmates stopped ir first simulator session and called company to get an ficial determination as to what guidance y should use for abnormal and emergency procedures during routine flight operations; he stated that ValuJet management advised m to use QRH instead manual. The DC-9 Operations Manuals were basically developed by Ryan from airplane s previous owner s Operations Manuals, and certain purported Ryan practices were not incorporated into m. The requirement to conduct an exterior inspection airplane at intermediate stops was one those practices not incorporated. In fact, preflight inspection requirement in Ryan DC-9 manual clearly indicated that exterior inspections were required only on originating flights or after airplane had been left unattended. Table 11. CRM issues as contributing factors. Accident Report Proper CRM was not present. The captain never made a comment regarding deviations or helped First Officer before landing. The captain commented about flap problem, neir crewmember discussed a procedure or checklist to address it. The flight crew s poor communication and failure to follow operating procedures regarding flap asymmetry, showed lack proper Crew Resource Management and Negligence as a Flight Crew during approach. The captain changed autopilot mode from previously briefed prile approach to vertical speed mode, initially setting vertical descend rate to about 700 fpm, n increasing it to 1000 fpm; however, he did not brief first ficer about autopilot mode change A lack proper crew coordination, including a virtual reversal roles by DC9 pilots, which led to ir failure to stop taxing ir airplane and alert ground controller ir positional uncertainty in a timely manner before and after intruding onto active runway. The captain did not adequately manage his crew resources when he failed to call for checklist or to monitor and facilitate accomplishment required checklist items.

19 Safety 2018, 4, Table 11. Cont. Accident Report The first ficer told Safety Board investigators that his goal after recognizing that flaps were not extended was to get captain to initiate a go-around. Thirty seconds before touchdown, first ficer stated want to take it around? and captain replied no that s alright, keep your speed up here about uh. When captain denied first ficer s request to go around and told him to keep his speed up, first ficer did not challenge captain s statement. He also did not question captain to determine his reason(s) for continuing approach. The first ficer stated that re was no time for discussion with captain because approach was so fast. The first ficer s failure to question captain s decision to continue approach was inconsistent with CRM training he had received that emphasized importance sharing doubts with or crewmembers and quickly resolving conflicts. The flight engineer brought to captain s attention airspeed deviation but captain never corrected; neir first ficer nor flight engineer called for a go-around. During course performing missed approach procedure, first ficer acted, without challenge, to a command from captain to down, push it down. The flight crew ignored present wear conditions and continued approach to land even during unestablished approach, captain took flight controls at last moment when it was too late to correct or execute a go around. Shortly reafter, cockpit voice recorder CVR revealed comments by captain on first ficer s flying technique, such as If you re gonna fly that slow you gotta have more flaps, and [unintelligible words] still don t have enough flaps for this speed...add power...you re not on glidepath...bring it up to glidepath, and You re not even on [expletive] localizer at all. At 03 13, captain stated Okay, we re gonna have to go around...cause we re not anywhere near localizer... anywhere near it. Inexplicably, first ficer reacted to stick shaker by immediately deciding that captain should be flying and abandoning control airplane to captain without warning or proper coordination. About 1 min later, first ficer stated, something s messed up with this thing, and, at 0039:07, he asked why is this thing? At 0041:21, first ficer stated that control wheel felt funny. He added, feels like I need a lot force. It is pushing to right for some reason. I don t know why...i don t know what s going on. The first ficer n repeated twice that it felt like he needed a lot force. The CVR did not record captain responding to any se comments. The reason captain did not recognize impending onset stick shaker could not be determined from available evidence but that first ficer s tasks at time low-speed cue was visible would have likely reduced opportunities for her timely recognition impending event; failure both pilots to detect this situation was result a significant breakdown in ir monitoring responsibilities and workload management. Failure crew to recognize and recover from an unusual attitude after experiencing spatial disorientation or an attitude indicator failure during second missed approach. The flight crew did not monitor fuel quantity gauges or respond properly to airplane s changing handling characteristics. Flight crew s failure to detect and remove ice contamination from wings was a causal factor in this accident. All se NTSB excerpts signal a lack proper sleep or operations outside a normal sleep schedule. Or signs fatigue could be simple misspeaks, as when talking over radio or discernable yawns on CVR. According to FAA [1], a fatigued aviator cannot perform at a level corresponding with operational requirements, and fatigue itself can be as incapacitating as drug use.

20 Safety 2018, 4, Table 12. fatigue as a contributing factor. Accident Report Safety 2018, 4, The pilots were flying ir sixth flight day and had flown about 6 h and 14 min in intermediate 14 h and 31 stops minwas duty one time those when practices accident not incorporated. occurred. CVR In fact, recorded preflight a yawn on inspection first ficer s requirement channel. in Ryan DC-9 manual clearly indicated that exterior inspections were Each required pilot made only on an inappropriate originating flights decision or after to use airplane crew room had been to obtain left unattended. rest before accident flight. Passages such There as wereairplane several obvious was dispatched misspeaks to byanc both pilots with (drift left vs. engine crab, and thrust 25 degrees reverser inoperative, company flaps vs. did 23not degrees teach flaps) its pilots that may bounced have indicated landing recovery some degree techniques, fatigue. Maintenance Notwithstanding fact that crewmembers flight 805 were air cargo operations personnel use an inappropriate manual engine star procedure, and use QRH instead veterans and had adapted to se types disrupted work/sleep schedules many times, manual are all indications this experience airline didmanagement not make moperational immune to deficiencies possible adverse that contributed effects fatigue to many or airplane accidents. James ir ability Reason to function [21] created effectively. term organizational accident to allude to those aviation mishaps that The were captain caused reported by latent that hefailures receivedwithin only about 1organization. h sleep during These night organizational before errors when left alone accident and unattended and, as a result, to, may askedlead to first a ficer catastrophic to be vent flyingsuch pilotas foran accident. flight. Safety 2018, 4, 2 Figure 10. Tag cloud for airline management contributing factors. Figure 10. Tag cloud for airline management contributing factors The most commonly-used words found in word frequency query for airline management issues were airplane, procedures, company and inspection. Many company management issues, such as lack, or bad, procedures and inspections, may lead to airplane experiencing a catastrophic event such as aviation accident; hence inclusion words company, procedures, inspections, and airplane. Table 11. CRM issues as contributing factors. Accident Report Proper CRM was not present. The captain never made a comment regarding deviations or helped First Officer before landing. The captain commented about flap problem, neir crewmember discussed a procedure or checklist to address it. The flight crew s poor communication and failure to follow operating procedures regarding flap asymmetry, showed lack proper Crew Resource Management and Negligence as a Flight Crew during approach. The captain changed autopilot mode from previously briefed prile approach to vertical speed mode, initially setting vertical descend rate to about 700 fpm, n increasing it to 1000 fpm; however, he did not brief first ficer about autopilot Figure 11. Word frequency tag cloud for CRM issues. mode change A lack proper crew coordination, including a virtual reversal roles by DC9 The words most commonly coded within CRM issues were captain, first, ficer, airplane and pilots, which led to ir failure to stop taxing ir airplane and alert ground approach Unsurprisingly, se words have come up again as most frequently-used words. controller ir positional uncertainty in a timely manner before and after intruding Given that CRM is an onto issue active team runway. dynamics inside cockpit, words airplane, first, ficer and captain were expected. The captain The word did approach, not adequately once manage again, his might crew indicate resources a pattern when he failed CRM to breakdown call for 98-03

21 Safety 2018, 4, Safety 2018, 4, Figure 12. Tag cloud for fatigue as a contributing factor. Figure 12. Tag cloud for fatigue as contributing factor. The most commonly-coded words within fatigue were accident, fatigue, sleep, crew and room. Some excerpts from Table clearly show pilots suffering from from fatigue and and using using crew crew room room to sleep to sleep or get or get adequate rest. rest Discussion Neglect Flight Planning, Preflight Inspections and Checklists was most widespread behavioral trap. trap. This This unsafe unsafe behavior behavior was was identified identified in 79% in 79% accidents accidents studied. Loss studied. Situational Loss Awareness Situational came Awareness in a close came second in a close placesecond with representation place with representation in 74%. Peer Pressure in 74%. Peer was found Pressure in 18% was found accidents, in 18% while accidents, trapswhile Get-There-Itis traps Get-There-Itis and Unauthorized and Unauthorized Descent BelowDescent an IFR Altitude Below an were IFR both Altitude present were inboth 15% present accidents. in 15% accidents Neglect Flight Planning, Preflight Inspections, and Checklists 4.1. Neglect Flight Planning, Preflight Inspections, and Checklists A review NTSB excerpts presented in Results Section confirms many prior discoveries review NTSB excerpts presented in Results Section confirms many prior discoveries in that pilots, even air carrier aviators, might have a general disregard for rules or procedures and in that pilots, even air carrier aviators, might have general disregard for rules or procedures and underutilize many resources at ir disposal [6,15 17]. The findings this study also align with [18], underutilize many resources at ir disposal [6,15 17]. The findings this study also align with [18], where willful violations were present in 35% regular air carrier flights observed, and study where willful violations were present in 35% regular air carrier flights observed, and study conducted by Velazquez, Peck and Sestak [19], where Neglect Flight Planning, Preflight Inspections, conducted by Velazquez, Peck and Sestak [19], where Neglect Flight Planning, Preflight and Checklists was most dominant trap. Inspections, and Checklists was most dominant trap. If airline flight operations are so highly scripted [6], why are pilots unwilling to follow rules If airline flight operations are so highly scripted [6], why are pilots unwilling to follow rules and and established procedures? An explanation is that Part 121 pilots may experience a phenomenon established procedures? An explanation is that Part 121 pilots may experience a phenomenon called called habitual noncompliance [15]. Highly qualified pilots who routinely fly toger under repetitive habitual noncompliance [15]. Highly qualified pilots who routinely fly toger under repetitive circumstances may constantly betray ir own Flight Operations Manual (FOM) processes. The NTSB circumstances may constantly betray ir own Flight Operations Manual (FOM) processes. The and Goglia [15] advocate installation cockpit cameras to help ensure that pilots conduct NTSB and Goglia [15] advocate installation cockpit cameras to help ensure that pilots conduct mselves under established protocols [16]. mselves under established protocols [16] Loss Situational Awareness 4.2. Loss Situational Awareness The leading behavioral trap in fatal accidents was Loss Situational Awareness. This is not The leading behavioral trap in fatal accidents was Loss Situational Awareness. This is not surprising because concept involves more than knowledge aircraft s geographical or spatial surprising because concept involves more than knowledge aircraft s geographical or spatial position. It also comprises pilot s consciousness different elements affecting overall position. It also comprises pilot s consciousness different elements affecting overall status aircraft. These elements include wear, aircraft condition, crewmember state and status aircraft. These elements include wear, aircraft condition, crewmember state and mission or flight progress. If passengers are being transported, y also form part expansive mission or flight progress. If passengers are being transported, y also form part expansive definition situational awareness [4]. Thus, any sign spatial, geographic, operational or procedural definition situational awareness [4]. Thus, any sign spatial, geographic, operational or procedural disorientation would be coded under Loss Situational Awareness. As opposed to

22 Safety 2018, 4, disorientation would be coded under Loss Situational Awareness. As opposed to previous behavioral trap Neglect Flight Planning, Preflight Inspections and Checklists under Loss Situational Awareness, crew may not be cognizant danger. Sadly, this may explain it being leading prevalence behavioral traps under fatal accidents Peer Pressure Human beings have a natural desire to conform to ors, to be accepted [20]. As stated earlier, Peer Pressure can be verbal, or non-verbal, obvious or subtle, intentional or unintentional, and its origin may be personnel or organizational [20]. In all but one case examined during this study, it was captain flight who was source Peer Pressure for first ficer. A look at all word frequency queries associated with behavioral traps studied revealed that word captain is within top five most commonly-found words. This finding suggests that first ficers are automatically disengaging or suppressing ir own arguments for sake acceptance. This lack assertiveness is furr explained in a following section. Finally, although re were instances managerial factors that contributed to accidents, no overwhelming evidence was found that airlines provided organizational pressure to crews ill-fated flights Get-There-Itis As mentioned earlier, as flight progresses, pilots desire to continue gets stronger [6,20]. This tendency was exemplified in Table 10 where four out five cases Get-There-Itis occurred during approach and landing phase. This finding confirms what Dismukes et al. [6] called plan continuation bias, a failure crew to discontinue an approach when it becomes inappropriate or dangerous to do so. Interestingly, word frequency tag cloud for Get-There-Itis, and for five out six behavioral traps studied, suggests that behavioral traps occur mostly in approach and landing phase flight because words approach and/or landing appear as top common words. This is not surprising considering that majority aviation accidents, including commercial, occur during approach and landing phase flight Unauthorized Descent below an IFR Altitude A look at Figure 9 reveals that word MDA or Minimum Descent Altitude was among most frequently found within documented sources. This finding initiated a search back into NTSB excerpts found in Table 11 to find out wher or not all instances Descending Below an IFR Altitude were indeed associated with non-precision approaches or approaches where no vertical guidance is available. The conclusion was a resounding yes. All cases this behavioral trap were associated with non-precision approaches. These types instrument approaches add complexity to approach and landing phase flight, more so if approach was originally a precision approach, and due to technological difficulties, crew was left with a different approach at last minute CRM Issues The third overall factor contributing to accidents was lack CRM practices. This finding is not surprising considering most prevalent behavioral trap across all cases was Neglect Flight Planning, Preflight Inspections, and Checklists. In addition, re was presence or CRM-rescinding traps such as Peer Pressure, Unauthorized Descent Below an IFR Altitude, Get-There-Itis and additional discovered trap Flying Outside Envelope. CRM is epitome or ultimate expression teamwork between flight crewmembers prior to, during and after a flight. Good CRM practices are predicated on following checklists, SOPs, conducting good preflight action and engaging in proper flight planning to prepare for unexpected events during flight. However, as seen throughout this study, crews are falling under habitual noncompliance, and first ficers are demonstrating a lack assertiveness. Broome [21] believes pilots are rejecting CRM. Is it time to refresh or revamp CRM training?

23 Safety 2018, 4, Though CRM has evolved through many generations to point that crews today are aware that best strategy is to manage threats and errors, it looks as if CRM training lacks an important component called attitude management training. Attitude management is defined as ability to recognize hazardous attitudes in oneself and willingness to modify m as necessary [3]. Unfortunately, FAA CRM training guidance [8] does not provide any direction on attitude management training, nor does it provide much information about hazardous attitudes, behavioral traps or various cognitive biases with which pilots are confronted. The FAA has recently published a Notice Proposed Rulemaking (NPRM) to increase airline pilot pressionalism and ensure pilots adhere to SOPs. With this new NPRM, now would be a logical moment to define and introduce attitude management training as a formal component to CRM training. Line-Oriented Flight Training (LOFT) has been preferred CRM training method for years. However, results this study confirm many findings [6,22] suggesting that this scenario-based training tool may not be applied effectively and continuously. Dismukes et al. [6] cites inadequate knowledge or experience provided by training and/or guidance as a factor in 37% NTSB accidents between 1991 and In or words, pilots were not given adequate instruction about problems known by some sectors industry to exist or, found mselves in challenging situations for which y had received training, but experience received from that training was inadequate fidelity to actual situation, inadequately detailed, or incomplete [6]. 5. Conclusions and Recommendations 5.1. Conclusions This paper accomplished many firsts and contributed considerably to understanding how negative behaviors (specifically behavioral traps) are present in airline operations. No published study had tackled behavioral traps in air carrier operations until now. Secondly, this paper revises Jeppesen s [11] categorization behavioral traps among GA, instrument-rated and commercial pilots. For example, it was discovered that behavioral trap Flying Outside Envelope is not exclusive to GA pilots; airline pilots also exceed airplane operational tolerances. Finally, study also makes public how flight crews might be practicing CRM and tells story captains preeminence. A look at all word frequency queries associated with behavioral traps revealed that word captain is within top five most commonly-found words. Could it be that captains are overpowering ir fellow first ficers? Are today s captains still being overly dominant? If so, how effective is CRM if it is evident that first ficers are automatically disengaging or suppressing ir own arguments for sake acceptance? As indicated earlier, Flying Outside Envelope should be included in commercial category behavioral traps [11] among commercial, instrument-rated and GA pilots due to its presence in 21% accidents analyzed. With exception Peer Pressure, all or behavioral traps mainly occur in approach and landing phase flight. This finding coincides with phase flight responsible for majority commercial aviation accidents. Finally, Unauthorized Descent Below an IFR Altitude was completely related to non-precision approaches or instrument approaches without vertical guidance in approach design. These instrument approaches add a level complexity to operation as opposed to those precision approaches where an electronic vertical guidance is available Recommendations Recommendations for Furr Study Because this study focused on U.S. 14 CFR Part 121 crew-related accidents, any future studies can focus on Part 135 commercial and air taxi operators. In addition, while information contained in National Aeronautics and Space Administration s ASRS accounts is self-reported by pilots,

24 Safety 2018, 4, valuable information can be retrieved from se incident reports to continue to understand unsafe pilot behaviors, wher y are defined as hazardous attitudes or behavioral traps. Because Neglect Flight Planning, Preflight Inspections, and Checklists was top behavioral trap, additional research should focus on reasons for customary noncompliance and pilot motivation. While NTSB and Goglia [15] are advocating for cockpit cameras to be installed in air carrier operations, perhaps a better approach would be to scientifically review flight data recorders. This assessment should be routinely accomplished by airlines in a non-punitive way [16] Recommendations for Industry The creation CRM and alike programs does not always guarantee absence unsafe pilot behaviors [23 27]. However, effective crew performance depends on both technical priciency and interpersonal skills. One main objectives behind FAA s CRM training has always been to focus on crew member attitude and effectual teamwork. Because FAA believes attitudes can be changed or modified through training [1], from standpoint accident prevention, education and training focused on top behavioral traps would likely prove to have highest payf. This recommendation is especially true considering that Neglect Flight Planning, Preflight Inspections and Checklists is prevailing trap. Additional focus should be placed on: (1) captain s authority and ability to identify and mediate unsafe behaviors; (2) first ficer s ability to be assertive and combat Peer Pressure; and (3) approach and landing phase. The lack first ficer assertiveness and preeminence captains should be addressed in training and even investigated in future studies. This former action could be done through cognitive debiasing training and/or scenario-based training during LOFT sessions where additional focus is on interpersonal skills flight crewmembers. Many behavioral traps exist in airline operation. The understanding pilot attitudes and ir role in team dynamics or impact on CRM requires furr study. Finally, attitude management training is recommended as an added component in CRM training. Acknowledgments: The author this research project wishes to express gratitude to following airline transport pilots who contributed ir expertise to data collection portion this study. These individuals were: Kevin Roman, Oswart Mora, Omar Carle, and Pablo Ortiz. Conflicts Interest: The author declares no conflict interest. References 1. Federal Aviation Administration (FAA). Risk Management Handbook; Government Printing Office: Washington, DC, USA, Federal Aviation Administration (FAA). Aeronautical Decision Making (Advisory Circular 60-22); Government Printing Office: Washington, DC, USA, Sanderson, J. Private Pilot; Englewood, CO, USA, Federal Aviation Administration (FAA). Aviation Instructor s Handbook; Government Printing Office: Washington, DC, USA, Murata, A.; Nakamura, T.; Karwowski, W. Influence Cognitive Biases in Distorting Decision Making and Leading to Critical Unfavorable Incidents. Safety 2015, 1, [CrossRef] 6. Dismukes, R.K.; Berman, B.A.; Loukopoulos, L.D. The Limits Expertise: Rethinking Pilot Error and Causes Airline Accidents; Ashgate Publishing Limited: Aldershot, UK, Block, E.E.; Sabin, E.J.; Patankar, M.S. The structure safety climate for accident free flight crews. Int. J. Appl. Aviat. Stud. 2007, 7, Federal Aviation Administration (FAA). Crew Resource Management Training (Advisory Circular AC e); Government Printing Office: Washington, DC, USA, Helmreich, R.L.; Merritt, A.C.; Wilhelm, J.A. The evolution crew resource management training in commercial aviation. Int. J. Aviat. Psychol. 1999, 9, [CrossRef] [PubMed]

25 Safety 2018, 4, Velazquez, J.; Bier, N. SMS and CRM: Parallels and opposites in ir evolution. J. Aviat./Aerosp. Educ. Res. 2015, 24, [CrossRef] 11. Jeppesen. Flight Instructor; JeppDirect: Englewood, CO, USA, Bazeley, P. Qualitative Data Analysis: Practical Strategies; Sage: Thousand Oaks, CA, USA, Lincoln, Y.S.; Guba, E.G. Naturalistic Inquiry; Sage: Beverly Hills, CA, USA, Creswell, J.W. Educational Research: Planning, Conducting, and Evaluating Quantitative and Qualitative Research; Pearson Education, Inc.: Upper Saddle River, NJ, USA, Goglia, J. Torqued: Is Gulfstream IV Crash Corporate Aviation s Wakeup Call? Available online: (accessed on 1 December 2015). 16. Rapp, R. Bedford and Normalization Deviance. Available online: /12/normalization--deviance/ (accessed on 12 December 2015). 17. Veillette, P.R. Watching and waning. Bus. Commer. Aviat. 2006, 98, Klinect, J.R.; Wilhelm, J.A.; Helmreich, R.L. Threat and error management: Data from line operations safety audits (LOSA). In Proceedings 10th International Symposium on Aviation Psychology; The Ohio State University: Columbus, OH, USA, Velazquez, J.; Peck, A.G.; Sestak, T. Behavioral traps in crew-related aviation accidents. In Proceedings 18th International Symposium on Aviation Psychology; Wright State University: Dayton, OH, USA, May 2015; pp Kern, T. Flight Discipline; McGraw-Hill Pressional: Columbus, OH, USA, Broome, D.P. Accident reduction through crew resource management. J. Aviat./Aeros. Educ. Res. 2011, 20, [CrossRef] 22. Wagener, F.; Ison, D.C. Crew resource management application in commercial aviation. J. Aviat. Technol. Eng. 2014, 3, [CrossRef] 23. Cook, J.S. Analysis Hazardous Pilot Behaviors and Causal Factors in Part 121 and Part 135 Aircraft Accidents and Incidents. Master s Thesis, Retrieved from ProQuest Dissertations and Theses database, (UMI No ). Arizona State University, Tempe, AZ, USA, Mosier, K.L.; Fischer, U.; Cunningham, K.; Munc, A.; Reich, K.; Tomko, L.; Orasanu, J. Aviation decision making issues and outcomes: Evidence from ASRS and NTSB reports. In Proceedings Human Factors and Ergonomics Society Annual Meeting; Sage: Los Angeles, CA, USA, October 2012; pp [CrossRef] 25. Shappell, S.; Detwiler, C.; Holcomb, K.; Hackworth, C.; Boquet, A.; Wiegmann, D.A. Human error and commercial aviation accidents: An analysis using human factors and classification system. Hum. Factors J. Hum. Factors Ergon. Soc. 2007, 49, [CrossRef] [PubMed] 26. Wetmore, M.; Lu, C. The effects hazardous attitudes on crew resource management skills. Int. J. Appl. Aviat. Stud. 2006, 6, Reason, J. Human Error; Cambridge University Press: Cambridge, UK, by author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under terms and conditions Creative Commons Attribution (CC BY) license (