Skip Ribbon Commands
Skip to main content
Safety Recommendation Details

Safety Recommendation A-04-062
Synopsis: On November 12, 2001, about 0916:15 eastern standard time, American Airlines flight 587, an Airbus Industrie A300-605R, N14053, crashed into a residential area of Belle Harbor, New York, shortly after takeoff from John F. Kennedy International Airport, Jamaica, New York. Flight 587 was a regularly scheduled passenger flight to Las Americas International Airport, Santo Domingo, Dominican Republic, with 2 flight crewmembers, 7 flight attendants, and 251 passengers aboard the airplane. The airplane’s vertical stabilizer and rudder separated in flight and were found in Jamaica Bay, about 1 mile north of the main wreckage site. The airplane’s engines subsequently separated in flight and were found several blocks north and east of the main wreckage site. All 260 people aboard the airplane and 5 people on the ground were killed, and the airplane was destroyed by impact forces and a postcrash fire. Flight 587 was operating under the provisions of 14 Code of Federal Regulations (CFR) Part 121 on an instrument flight rules flight plan. Visual meteorological conditions prevailed at the time of the accident.
Recommendation: TO THE FEDERAL AVIATION ADMINISTRATION: Along with developing the guidance recommended in Safety Recommendation A-04-61, evaluate issues concerning the level of automation appropriate to teaching upset training and develop and disseminate guidance that will promote standardization and minimize the danger of inappropriate simulator training.
Original recommendation transmittal letter: PDF
Overall Status: Closed - Acceptable Action
Mode: Aviation
Location: Belle Harbor, NY, United States
Is Reiterated: No
Is Hazmat: No
Accident #: DCA02MA001
Accident Reports: In-Flight Separation of Vertical Stabilizer American Airlines Flight 587, Airbus Industrie A300-605R, N14053
Report #: AAR-04-04
Accident Date: 11/12/2001
Issue Date: 11/10/2004
Date Closed: 7/21/2016
Addressee(s) and Addressee Status: FAA (Closed - Acceptable Action)
Keyword(s): Training and Education,

Safety Recommendation History
From: NTSB
Date: 7/21/2016
Response: We note that, in the final rule, Table A1A, Qualification Performance Standards (QPS) Requirements, entry 2.n, “Upset Prevention and Recovery Training,” states that “the unrealistic degradation of simulator functionality (such as degrading flight control effectiveness) to drive an airplane upset is generally not acceptable unless used purely as a tool for repositioning the FSTD with the pilot out of the loop.” This statement satisfies Safety Recommendation A-04-62, which is classified CLOSED—ACCEPTABLE ACTION.

From: NTSB
Date: 10/2/2014
Response: We have reviewed the Federal Aviation Administration’s (FAA) notice of proposed rulemaking (NPRM) titled “Flight Simulation Training Device Qualification Standards for Extended Envelope and Adverse Weather Event Training Tasks,” published at Federal Register 39462-39753 on July 10, 2014. This recommendation is currently classified “Open—Acceptable Response.” If the revisions proposed regarding simulator standards related to use of FSTDs for upset recovery training become final, we likely will consider this an acceptable action.

From: NTSB
Date: 9/16/2013
Response: We do not agree that the three documents that the FAA referenced in its letter effectively address this recommendation. These documents are described below, along with our evaluation of their relevance to Safety Recommendation A-04-62. 1. Information for Operators (InFO) bulletin 10010, “Enhanced Upset Recovery Training,” which advises of the availability of and encourages the incorporation into training programs of the Airplane Upset Recovery Training Aid (AURTA). We believe the AURTA is a valuable and useful document, and revisions made to it were the basis for our previously classifying Safety Recommendation A-04-61 “Closed?Acceptable Action.” However, our review of the AURTA and the InFO did not find any discussion of the issues discussed in Safety Recommendation A-04-62, as discussed below. 2. Safety Alert for Operators (SAFO) 10012, “Possible Misinterpretation of the Practical Test Standards (PTS) Language ‘Minimal Loss of Altitude,’” which provides guidance on stall recovery training. This SAFO is a valuable document that addresses issues raised in other safety recommendations that we have made regarding stall awareness and recovery training; however, it is not relevant to the issues discussed in Safety Recommendation A-04-62. 3. Advisory Circular (AC) 120-109, “Stall and Stick Pusher Training” which provides best practices on training, testing, and checking of stall warnings, aerodynamic stalls, and recommends recovery procedures. Although this AC discusses the uses and limitations of simulators in stall training, it does not address our specific concerns regarding the use of simulators in upset training. We acknowledge the FAA’s point that a causal factor in many loss-of-control accidents is the pilot’s inappropriate reaction to the first indication of a stall, and both the SAFO and AC address pilot training to avoid this inappropriate reaction. We note that the simulator exercise in the American Airlines Advanced Aircraft Maneuvering Program that contributed to the negative training of the flying pilot in the American Airlines flight 587 accident involved a wake turbulence encounter, not a stall. We believe it is appropriate to reiterate the reason why Safety Recommendation A-04-62 was issued. In the letter that transmitted this recommendation to the FAA, we said this: Regarding simulator training, the Safety Board recognizes that some members of the training community advocate the introduction of upset situations in simulators by having pilots close their eyes or look away as the upset is established, rather than by attempting a simulated context for the onset. The Board also recognizes that some members of the training community believe that advanced simulators should not be used in upset training because the range of simulator [motion] fidelity is relatively narrow and the portrayal of accelerations is not comparable with what could occur during an aggressive upset. Therefore, the Safety Board concludes that the use of lower levels of automation, such as simulators without motion or simple computer screen displays, may be more appropriate to provide the necessary awareness training with less danger of introducing incorrect information. Although the AURTA, InFO 10010, SAFO 10012, and AC 120-109 serve valuable purposes related to stall and upset recovery training, none of them adequately addresses the issue of when the use of lower levels of automation, such as simulators without motion or simple computer screen displays, may be more appropriate. This recommendation is now 8 1/2 years old, and we generally expect action taken to address our recommendations to be completed within 5 years. We also acknowledge that discussions and correspondence with the FAA shortly after this recommendation was issued may have contributed to the belief that these documents would address the recommendation. Therefore, we ask the FAA to reconsider what further actions are necessary to address the issue of when the use of lower levels of automation may be more appropriate. Pending the FAA’s consideration of our comments and completion of the recommended action as clarified here, Safety Recommendation A-04-62 remains classified OPEN—ACCEPTABLE RESPONSE.

From: FAA
Date: 6/10/2013
Response: -From Michael P. Huerta, Administrator: On July 6, 2010, the Federal Aviation Administration (FAA) published Information for Operators (InFO) 10010, Enhanced Upset Recovery Training. The InFO highlights the merits of the FANindustry-produced Airplane Upset Recovery Training Aid (AURTA) and provides a Web link to the AURTA. The InFO also encourages incorporation of applicable sections of the AURIA into training programs. The InFO can be found at: visit/aviation_industry/airline_operators/airline_safety/info/all_infos/me dia/2010/InFOIOOIO.pdf. We also published Safety Alert for Operators (SAFO) 10012, Possible Misinterpretation of the Practical Test Standards (PTS) Language "Minimal Loss of Altitude," on July 6, 2010. This SAFO clarified the meaning of the approaches to stall evaluation criteria, as they relate to "minimal loss of altitude" in the Airline Transport Pilot PTS. It also recommended evaluation criteria for recovery from a stall or approach-to-stall that do not mandate a predetermined value for altitude loss. Instead, the SAFO recommends consideration of a multitude of external and internal variables affecting recovery altitude. This SAFO can be found at: visit/aviation_industry/airline _operators/airline_ safety/safo/all_ safos/media/20 1 0/SAFO 1 0012. pdf. In our initial response letter dated March 1, 2005, we discussed our work under the Safer Skies initiative and with the Commercial Aviation Safety Team (CAST) to identify and implement the most effective methods of flightcrew training for recovery from loss of control (LOC) using existing flight simulation devices. CAST work continues, but the CAST report mentioned in the 2005 letter has not yet been issued. However, based on FAA and aviation industry work done outside of CAST, we issued Advisor Circular (AC) 120-109, Stall and Sticker Pusher Training, on August 6, 2012. A growing causal factor in LOC accidents is the pilot's inappropriate reaction to the first indication of a stall or stick pusher event. Issuance of this AC is intended to more quickly address LOC accidents and training to prevent these types of accidents. AC 120-109 provides best practices on training, testing, and checking of stall warnings, aerodynamic stalls, and stick pusher activations and recommends recovery procedures. The AC includes sections on stall training philosophy, training methodology, and sample training scenarios. The AC also includes an appendix on flight simulation training devices (FSTDs), which include both full flight simulators and flight training devices. This appendix discusses FSTD capabilities and FSTD evaluation recommendations. The AC is responsive to concerns expressed in the Board's August 3, 2005, letter. In that letter, the Board cautioned the FAA to not limit its efforts to addressing only the most effective uses of flight simulation devices. The AC provides best practices and guidance for training, testing, and checking for pilots, within existing regulations, and does not limit the use of FSTDs. The AC can be found at: http://www .faa. gov I documentLibrary /medial Advisory_ Circular/ AC%20 120-1 09. pdf. I believe the FAA has effectively addressed this safety recommendation by issuing InFO 10010, SAFO 10012, and AC 120-109. Therefore, I consider our actions complete and plan no further action on this recommendation.

From: NTSB
Date: 4/30/2012
Response: Notation 8406: The National Transportation Safety Board (NTSB) has reviewed the Federal Aviation Administration’s (FAA) notice of proposed rulemaking (NPRM) titled “Pilot Certification and Qualification Requirements for Air Carrier Operations,” which was published at 77 Federal Register (FR) 12374 on February 29, 2012. The notice proposes to create new certification requirements for pilots in air carrier operations, including requiring that first officers in 14 Code of Federal Regulations (CFR) Part 121 operations hold an airline transport pilot (ATP) certificate and type rating for the aircraft to be flown; allowing pilots with an aviation degree or military pilot experience but fewer than 1,500 hours total time as a pilot to obtain an ATP certificate with restricted privileges; and requiring at least 1,000 flight hours in air carrier operations to serve as pilot in–command (PIC) in Part 121 air carrier operations. The notice also proposes to modify the requirements for obtaining an ATP certificate with an airplane category multiengine class rating or type rating to require 50 hours of multiengine flight experience and completion of a new FAA-approved ATP certificate training program that would include academic training and training in a flight simulation training device. According to the NPRM, these changes would help to ensure that pilots entering an air carrier environment have the training and aeronautical experience necessary to adapt to a complex, multicrew environment in a variety of operating conditions. The NPRM cites the 2009 Colgan Air accident near Buffalo, New York, as an event that focused public, congressional, and industry attention on flight crew experience requirements and training for conducting Part 121 air carrier operations. In February 2010, the FAA published an advance notice of proposed rulemaking (ANPRM), titled “New Pilot Certification Requirements for Air Carrier Operations” (75 FR 6164, February 8, 2010) that sought input on current Part 121 eligibility, training, and qualification requirements for seconds-in-command (SICs). The current NPRM is based on comments in response to the ANPRM, input received from an aviation rulemaking committee established in July 2010, and statutory requirements for modifying ATP certification outlined in the Airline Safety and Federal Aviation Administration Extension Act of 2010 (Public Law 111-216). Adding to that foundation, the NPRM states that the FAA conducted a study of 61 NTSB investigation reports from fiscal year (FY) 2001 through FY 2010 (31 Part 121 accidents and 30 Part 135 air carrier accidents, with 107 fatalities, 28 serious injuries, and 44 minor injuries). The study showed that the accidents examined involved pilot deficiencies in aircraft handling, including stall and upset recognition and recovery, high altitude training, active pilot monitoring skills, effective crew resource management (CRM), stabilized approaches, operations in icing conditions, and hypoxia training. The NPRM asserts that the changes to air carrier pilot qualification would address, in part, 21 NTSB safety recommendations in the following areas: Safety Issue Recommendations Training flight crews to respond to sudden, unusual, or unexpected aircraft upsets: A-96-120, A-04-62, A-07-3, and A-09-113 Developing and conducting stall recovery training and providing stickpusher familiarization training for pilots of stickpusher-equipped aircraft: A-10-22 and -23 Training in high altitude operations: A-07-1 and -2 Training and guidance for rudder use in transport-category aircraft: A-02-2 Airport situational awareness: A-07-44 Stabilized approach concept: A 01 69 and A-08-18 Landing performance calculations: A-07-59 and A-08-41 CRM training: A-03-52 Pilot monitoring duties: A-10-10 Requirements for flight crewmember academic training regarding leadership and professionalism: A-10-15 Training in icing conditions: A-07-14 Hypoxia awareness training: A 00 110 Training in landing and taking off in crosswinds with gusts: A 10-110 and -111 The NTSB is generally supportive of the proposed rule as it relates to many of the issues previously identified in our safety recommendations. Specific comments on several areas of the NPRM follow. Academic Credit To Reduce Flight Experience Requirements Although the NTSB has not made recommendations for flight hour minimums for air carrier pilots (instead focusing its recommendations on specific procedures and training, needed regulations, and needed guidance to crews and operators), we stated in our comments on the ANPRM that: Ensuring a high level of knowledge, skills, and professionalism for flight crewmembers is essential, but total flight hours or an airline transport pilot certificate does not necessarily equate to the level of knowledge, skills, and professionalism required for consistently safe flight operations. The comments went on to state that, “the NTSB recognizes the value of academic training for air carrier pilots, but the NTSB also believes that academic training is not a substitute for practical experience.” An important tenet in the recent NPRM is the concept that, “in certain circumstances, the combination of focused academic training and structured flight training can substitute for actual flight experience” (p. 12379). The NTSB concurs with the FAA’s acknowledgement that there may be multiple pathways to becoming a qualified air carrier pilot. However, there remain unresolved issues for how academic credit should be applied, including student performance within an accredited academic program and the type of degree conferred. These issues are not addressed in the NPRM and require more evaluation before this proposal is implemented. It is essential that the content and rigor involved in academic training be clearly defined and, most importantly, appropriate resources allocated to conduct evaluation and oversight of these alternative methods of qualification. ATP Certification Training Program The NPRM discusses the establishment of an FAA-approved ATP certificate training program for a multiengine class ATP or type rating. The proposed training program outlined under section 61.154 would include 24 hours of classroom training and 16 hours of simulator training (8 in a full flight simulator of at least Level C standards) and is intended to provide pilots with the core knowledge and understanding in areas critical to operating high performance aircraft in a complex and high altitude environment. The training would be provided by an authorized training provider and would be required to be completed before a pilot would be eligible to take the ATP knowledge test. Issued as part of the NPRM, draft Advisory Circular (AC) 61-ATP, “Airline Transport Pilot Certification Training Program for Airplane Category Multiengine Class Rating or Type Rating,” contains an outline of the curriculum topics and objectives for both the classroom and simulator training making up this training program. The AC is intended for use by training providers when developing the program and by the FAA when reviewing and approving the programs. Many of the topics contained in the draft AC address issues from NTSB safety recommendations; in fact, the FAA notes that most of the 21 recommendations cited in the NPRM are addressed, in part, by the proposed amendments and advisory material. Although the NTSB concurs with the FAA’s assessment that, in most cases, the topics addressed will serve to partially satisfy the action requested in existing recommendations, the amount of specificity provided in the proposed rule and AC does not allow a comprehensive review of the degree to which the FAA’s proposed actions would satisfy the intent of the NTSB’s recommendations. In some instances, neither document provides evidence that a recommendation topic is addressed. The NTSB notes that recent safety recommendations in this area have focused on attempts to improve crew response to in-flight emergencies, including task prioritization and training. While AC 61-ATP does include a classroom training objective named “differences between emergency and non-normal checklist procedures and checklists,” the guidance on emergency procedures should be made more explicit to incorporate the issues identified in these NTSB recommendations. CRM is another topic relevant to previous NTSB recommendations and outlined in AC 61-ATP. However, the list of proposed topics in the AC does not explicitly refer to the importance of first officer assertiveness, which is an issue addressed in Safety Recommendation A-11-39. This recommendation is not cited in the NPRM, but the NTSB believes that it is within the scope of the draft advisory material and suggests amending the AC to include information consistent with Safety Recommendation A-11-39 to help support this important aspect of CRM. The NTSB is encouraged that the NPRM proposes to centralize the process for approving ATP certification training programs. Specifically, the NPRM states that only authorized training providers can administer the training required under section 61.154. These providers can be certificate holders providing training and operating under Parts 141, 142, 121, or 135, and each provider must receive approval of their ATP certification training program by the FAA Air Transportation Division (AFS-200). The NTSB notes that, theoretically, centralization should help to ensure standardization of these programs, but suggests that additional guidance documentation with more specific and robust detail about the content of the proposed training is necessary to provide a solid foundation on which the FAA can evaluate the program content (and to assist training providers to develop courses likely to receive FAA approval). For example, additional detail, such as cross-referencing material from draft AC 120-STALL, would be appropriate in the discussion of stall training in AC 61-ATP. In addition, the FAA will need to provide the appropriate oversight resources to these programs—not only in their initial approval but also to conduct ongoing oversight to demonstrate that the content delivered is consistent with the approved program. The rigor with which these programs are implemented and overseen will determine their ultimate influence on improving safety in air carrier operations. Pilot-in-Command Requirements for Air Carrier Operations The NPRM proposes primarily experience-based requirements for new PICs in air carrier operations. However, the NTSB has previously issued safety recommendations addressing the need for a specific leadership training course for upgrading captains. Although the NPRM cites Safety Recommendation A-10-15 and describes it as applicable to leadership and professionalism training, it addresses only the latter topic. The NPRM does not mention Safety Recommendations A-10-13 and -14, which were issued with -15, but the NTSB believes that a leadership training course for upgrading captains is within the scope of the proposed rulemaking and that section 121.436 should be amended to include a specific requirement for such a course. In addition to the requirements already outlined in section 121.434, the NTSB has recommended that Part 135 pilots who need a type rating for the aircraft they fly be required to have a minimum level of initial operating experience. Given the applicability of the NPRM to Part 135 pilots who are engaged in air carrier operations, the NTSB believes it would be appropriate to incorporate similar experience requirements for these pilots as exist for Part 121 pilots. The NTSB supports the use of simulators in training environments and notes that the training program outlined in the NPRM specifies that training on topics such as low energy states/stalls and upset recovery techniques will be conducted in a Level C or higher full-flight simulator. Simulators, regardless of their fidelity, are dependent on their physical limits of motion, as well as the efficacy of the available computer programs (which are often limited in issues of upset training because of the lack of flight test data at the extreme areas of the flight envelope). Simulators are not always adequate in portraying upsets and stalls and may inadvertently introduce negative training. Consistent with Safety Recommendation A-04-62, the FAA should allow flexibility in determining what level of simulation or automation is appropriate for specific training. Summary Observations This NPRM addresses many training issues applicable to becoming an air carrier pilot, including some critical issues demonstrated in recent accident history to be responsible for accidents. The NTSB is encouraged that its recommendations were considered in the development of this proposed rule, especially as the issue areas relate to the core content to be provided to new entrant pilots through the ATP certification training program. However, the intent of our recommendations in this area is for all pilots to receive training in these topics. Therefore, it is important that air carriers provide equally robust training in these topic areas for their current air carrier pilots on a recurrent basis. The NTSB appreciates the opportunity to comment on this NPRM.

From: NTSB
Date: 8/3/2005
Response: On December 15, 2004, FAA staff met with Safety Board staff to discuss this safety recommendation. At that meeting, Board staff explained that this recommendation applies to the full range of flight simulation devices from the simplest to the most complex. The FAA indicates that it has initiated research and development on the issue in this recommendation. Under its Safer Skies agenda and in coordination with the Commercial Aviation Safety Team, the FAA indicates that it is identifying and implementing methods of flight crew training for recovery from loss of control using existing flight simulation devices. The Safety Board believes these efforts are responsive to this recommendation but cautions the FAA not to limit these efforts to addressing only the most effective uses of flight simulation devices. The Board believes that the FAA should include consideration of the necessity and/or advisability of including flight simulator activities as part of loss of control training. Pending the results of the FAA's research and development activities investigating flight crew training for recovery from loss of control, Safety Recommendation A-04-62 is classified OPEN -- ACCEPTABLE RESPONSE.

From: FAA
Date: 3/1/2005
Response: Letter Mail Controlled 3/17/2005 12:28:54 PM MC# 2050112 - From Marion C. Blakey, Administrator: On December 15, 2004, the FAA staff met with the Board's staff to discuss this safety recommendation. The Board's staff explained that it was considering flight simulation devices comprising flight training devices ranging from Level 1 flight training devices at the low end, to Level D full flight simulators at the high end. The Board asked that the FAA address this safety recommendation with that understanding. The FAA has initiated research and development that addresses the shared concern of industry, the Board, and the FAA. Under its Safer Skies agenda and in coordination with the Commercial Aviation Safety Team, the FAA is identifying and implementing the most effective methods of flightcrew training for recovery from loss of control using existing flight simulation devices. The team conducting the work believes that it will make a preliminary report by January 1, 2007. I will provide periodic updates of the FAA's progress on this effort.