Description

On 23 July 2014, an ATR72-500 (B-22810) being operated by TransAsia Airways on a scheduled domestic passenger flight from Kaohsiung to Pengdu Island Airport at Magong as GE222 crashed and caught fire after failing to land following a VHF Omnidirectional Radio Range (VOR) approach to runway 20 at destination conducted in day Instrument Meteorological Conditions (IMC). There were 48 fatalities, including all 4 crew members; 9 of the remaining 10 passengers sustained serious injuries and one sustained minor injuries. Five buildings on the ground were impact/fire damaged and minor injuries were sustained by 5 people.

Investigation

An Investigation was carried out by the Taiwan Aviation Safety Council (ASC). Both the Flight Data Recorder (FDR) and Cockpit Voice Recorder (CVR) were recovered and data from them was successfully downloaded.

It was established that after a cruise at 7000 feet Altimeter Pressure Settings, the aircraft had arrived in the vicinity of the destination but after being informed of low visibility, the crew requested clearance to take up a holding pattern. Then, after almost 20 minutes holding, ATC advised that the visibility had improved to 800 metres and the crew requested an Instrument Landing System (ILS) approach to runway 02. Twelve minutes later, whilst still awaiting that clearance, ATC advised that although visibility for runway 20 had increased to 1600 metres, there was now a thunderstorm overhead and asked what the intentions of the crew were. They again requested an ILS approach to runway 02, but on almost immediately hearing a UNI Airways ATR 72 aircraft requesting and receiving clearance for a VOR approach to runway 20, they also requested such an approach.

Fifteen minutes later (and about 50 minutes after taking up a holding pattern) FDR data showed that descent on the VOR procedure had begun. Three minutes later, TWR issued a landing clearance and advised the crew that the surface wind velocity was 250° at 19 knots. The aircraft was flown through the 330 feet MDA and at a range of around 1.5nm was leveled at approximately 2000 feet. The AP was then disengaged and soon afterwards, "the aircraft began to deviate from the approach course to the left and continued to decrease in altitude". Twenty seconds after the AP was disconnected, the crew decided to initiate a missed approach and advised the TWR of this. A few seconds later after contact with some treetops, the aircraft crashed into buildings in the village of Xixi, located some 750 metres to the northeast of the threshold of Runway 20.

The Investigation sought to establish why no Terrain Avoidance and Warning System (TAWS) activations were recorded on either the CVR and in the EGPWS Non Volatile Memory. It was concluded during simulations of the flightpath flown by the accident aircraft that "the aircraft did not penetrate the terrain envelopes including the TCF envelope, runway field clearance floor (RFCF) envelope and Terrain Awareness “Look-Ahead” envelope for Software Version -011", which was the one installed in the accident aircraft, but that a later version of the software (-022) would have led to activation(s).

It was noted that all METARs issued for Magong throughout the period during which the aircraft was airborne and just after the time of the crash gave a thunderstorm stationary overhead the airport. They also all reported rain, heavy at times, with the reported cloud cover remained unchanged - SCT/200 feet, BKN/600 feet, FEW CB/1200 feet, OVC/1600 feet. The METAR 6 minutes before the time of the crash gave the wind velocity as 220° 11-21 knots and the visibility as 1600 metres. The METAR 4 minutes after the time of the crash gave the wind velocity as 250° 18-28 knots and both the visibility and Runway Visual Range (RVR) for runway 20 as 800 metres. It was concluded that the crew had been unlikely to gain visual contact with the runway environment despite the violation of MDA which had directly led to Controlled Flight Into Terrain (CFIT). Some issues relating to the provision of present weather information to the crews of approaching aircraft, in particular RVR, were attributed to the joint civil /military status of the airport.

The Jeppesen Chart used for the approach is shown below:

The scope of the Investigation was not confined to the specifics of the accident under investigation but sought to examine its wider context in terms of the TransAsia Airways operation and the oversight of it by the State Safety Regulator, the Civil Aeronautics Administration.

The Investigation formally documented a series of 10 'Findings related to Probable Causes' as follows:

Flight Operations

1. The flight crew did not comply with the published runway 20 VOR non-precision instrument approach procedures at Magong Airport with respect to the minimum descent altitude (MDA). The Captain, as the pilot flying, intentionally descended the aircraft below the published MDA of 330 feet in Instrument Meteorological Conditions (IMC) without obtaining the required visual references.
2. The aircraft maintained an altitude between 168 and 192 feet before and just after overflying the missed approach point (MAPt). Both pilots spent about 13 seconds attempting to visually locate the runway environment, rather than commencing a missed approach at or prior to the MAPt as required by the published procedures.
3. As the aircraft descended below the minimum descent altitude (MDA), it diverted to the left of the inbound instrument approach track and its rate of descent increased as a result of the flying pilot’s control inputs and meteorological conditions. The aircraft’s hazardous flight path was not detected and corrected by the crew in due time to avoid the collision with the terrain, suggesting that the crew lost situational awareness about the aircraft’s position during the latter stages of the approach.
4. During the final approach, the heavy rain and associated thunderstorm activity intensified producing a maximum rainfall of 1.8 mm per minute. The runway visual range (RVR) subsequently reduced to approximately 500 meters. The degraded visibility significantly reduced the likelihood that the flight crew could have acquired the visual references to the runway environment during the approach.
5. Flight crew coordination, communication, and threat and error management were less than effective. That compromised the safety of the flight. The First Officer did not comment about or challenge the fact that the captain had intentionally descended the aircraft below the published minimum descent altitude (MDA). Rather, the first officer collaborated with the captain’s intentional descent below the MDA. In addition, the First Officer did not detect the aircraft had deviated from the published inbound instrument approach track or identify that those factors increased the risk of a controlled flight into terrain (CFIT) event.
6. Neither of the flight crew recognised the need for a missed approach until the aircraft reached the point (72 feet altitude and 0.5 nautical miles beyond the missed approach point) where collision with the terrain became unavoidable.
7. The aircraft was under the control of the flight crew when it collided with foliage 850 meters northeast of the runway 20 threshold, two seconds after the go around decision had been made. The aircraft sustained significant damage and subsequently collided with buildings in a residential area. Due to the high impact forces and post-impact fire, the crew and most passengers perished.
8. According to flight recorder data, non-compliance with standard operating procedures (SOPs) was a repeated practice during the occurrence flight. The crew’s recurring non-compliance with SOPs constituted an operating culture in which high risk practices were routine and considered normal.
9. The non-compliance with standard operating procedures (SOPs) breached the obstacle clearances of the published procedure, bypassed the safety criteria and risk controls considered in the design of the published procedures, and increased the risk of a controlled flight into terrain (CFIT) event.

Weather

1. Magong Airport was affected by the outer rainbands of Typhoon Matmo at the time of the occurrence. The meteorological conditions included thunderstorm activities of heavy rain, significant changes in visibility, and changes in wind direction and speed.

It further documented a series of 27 Findings Related to Risk as follows:

Flight Operations

1. The Captain did not conduct a descent and approach briefing as required by standard operating procedures (SOPs). The first officer did not question the omission of that required briefing. That deprived the crew of an opportunity to assess and manage the operational risks associated with the approach and landing.
2. The Captain was likely overconfident in his flying skills. That might lead to his decision to continue the approach below the minimum descent altitude (MDA) without an appreciation of the safety risks associated with that decision.
3. The results of the fatigue analysis indicated that, at the time of the occurrence, the Captain’s performance was probably degraded by his fatigue accumulated from the multiple sectors/day flown and flight and duty times during the months preceding the occurrence.
4. The TransAsia Airways observation flights conducted by the Investigation Team and the interviews with members of the airline’s flight operations division showed prevalent tolerance for non-compliance with procedures within the airline’s ATR fleet.
5. The non-compliances with standard operating procedures (SOPs) during the TransAsia Airways’ ATR simulator training sessions observed by the Investigation Team were not corrected by the Instructors. The tolerance for or normalisation of SOPs non-compliance behaviours was symptomatic of an ineffective check and training system with inadequate supervision by the airline’s flight operations management.
6. The non-compliance with standard operating procedures (SOPs) was not restricted to the occurrence flight but was recurring, as identified by previous TransAsia Airways ATR occurrence investigations, line observations, simulator observations, internal and external audits or inspections and interviews with TransAsia Airways flight operations personnel, including managers. The non-compliant behaviours were an enduring, systemic problem and formed a poor safety culture within the airline’s ATR fleet.

Airline Safety Management

1. TransAsia Airways’ inadequate risk management processes and assessments, ineffective safety meetings, unreliable and invalid safety risk indices, questionable senior management commitment to safety, inadequate safety promotion activities, underdeveloped flight operations quality assurance (FOQA) system, and inadequate safety and security office and flight operations resources and capabilities constituted an ineffective safety management system (SMS).
2. The safety risks associated with change within the TransAsia Airways were not assessed and mitigated. For example, the company did not assess or mitigate the safety risks associated with the increase in ATR operational tempo as a result of the recent increase in ATR fleet size and crew shortage that, in turn, elevated crew flying activities and the potential safety risks associated with crew fatigue.
3. Findings regarding non-compliance with standard operating procedures (SOPs) during operations by the TransAsia Airways’ ATR crews had been identified by previous Aviation Safety Council occurrence investigations. The proposed corrective safety actions were not implemented by the airline.
4. TransAsia Airways self-audits were mostly spot checks rather than system audits or system self-evaluations. The self-audits failed to assess and address those safety deficiencies, including standard operating procedures, non-compliance behaviours, lack of standardization in pilot check and training activities and high crew flying activities on the ATR fleet. Such deficiencies had been pointed out in previous occurrences and audits and were considered by senior flight operations managers as problems.
5. The TransAsia Airways annual audit plan did not include an evaluation of the implementation and/or effectiveness of corrective actions in response to the safety issues identified in previous audits, regulatory inspection findings, or safety occurrence investigation recommendations. The airline’s self-audit program was not consistent with the guidance contained in (the applicable Advisory Circular)
6. The TransAsia Airways had not developed a safety management system (SMS) implementation plan. This led to a disorganized, non-systematic, incomplete and ineffective implementation, which made it difficult to establish robust and resilient safety management capabilities and functions.
7. The Civil Aeronautics Administration’s (CAA) safety management system (SMS) assessment team had identified TransAsia Airways’ SMS deficiencies, but TransAsia Airways failed to respond to the CAA’s corrective actions request. That deprived the airline of an opportunity to improve the level of safety assurance in its operations.
8. TransAsia Airways had not implemented a data-driven fatigue risk management system (FRMS) or alternative measures to manage the operational safety risks associated with crew fatigue due to fleet expansion and other operational factors.
9. The ATR flight operation did not include in its team a standards pilot to oversee Standard Operating Procedure (SOP) compliance, SOP-related flight operations quality assurance (FOQA) events handling and standard operations audit (SOA) monitoring before the GE222 accident.
10. Due to resource and capability limitations, the safety and security office was unable to effectively accomplish the duties they were required to undertake.
11. The safety and security office staff was not included in the Flight Safety Action Group. That deprived the airline of an opportunity to identify, analyse and mitigate flight safety risks in flight operations more effectively.
12. The TransAsia Airways’ safety management system was overly dependent on its internal reactive safety and irregularity reporting system (as a means) to develop full awareness of the airline’s safety risks. It did not take advantage of the instructive material from external safety information sources. That limited the capability of the system to identify and assess safety risks.
13. The TransAsia Airways’ flight operations quality assurance (FOQA) settings and analysis capabilities were unable to readily identify those events involving standard operating procedure (SOP) non-compliance during approach and likely other stages of flight. FOQA events were not analyzed sufficiently or effectively, leaving some safety issues in flight operations unidentified and uncorrected. Some problems with crew performance and reductions in safety (that were) indicated in the FOQA trend analyses were not investigated further. Clearly, the airline’s FOQA program was not used to facilitate proactive operational safety risk assessments.

Civil Aeronautics Administration

1. The Civil Aeronautics Administration’s oversight of TransAsia Airways did not identify and/or correct some crucial operational safety deficiencies, including crew non-compliance with procedures, non-standard training practices, and unsatisfactory safety management practices.
2. The development and maintenance of a safety management system (SMS) implementation plan at TransAsia Airways was not enforced by the Civil Aeronautics Administration. That deprived the Regulator of an opportunity to assess and ensure that the airline had the capability to implement a resilient SMS.
3. Issues regarding TransAsia Airways’ crew non-compliance with standard operating procedures (SOPs) and deficiencies with pilot check and training had previously been identified in Aviation Safety Council investigation reports. However, the Civil Aeronautics Administration (CAA) did not monitor whether the operator had implemented the recommended corrective actions; correlatively, the CAA failed to ensure the proper measures for risk reduction had been adopted.
4. The Civil Aeronautics Administration provided limited guidance to its Inspectors to enable them to effectively and consistently evaluate the key aspects of operators’ management systems. These aspects included evaluating organisational structure and staff resources, the suitability of key personnel, organisational change and risk management processes.
5. The Civil Aeronautics Administration did not have a systematic process for determining the relative risk levels of airline operators.

Air Traffic Service and Military

1. The Runway Visual Range (RVR) reported in the Magong aerodrome routine meteorological reports (METAR) and the aerodrome special meteorological reports (SPECI) was not in accordance with the requirements documented in the Air Force Meteorological Observation Manual.
2. The discrepancies between the reported runway visual range (RVR) and Automated Weather Observation System (AWOS) RVR confused TWR controllers about the reliability of the AWOS RVR data.
3. During the final approach, the runway 20 runway visual range (RVR) values decreased from 1,600 meters to 800 meters and then to a low of about 500 meters. This RVR information was not communicated to the occurrence flight crew by ATC. Such information might have influenced the crew’s decision regarding the continuation of the approach.

Finally, it documented a series of 9 Other Findings as follows:

1. The flight crew were properly certificated and qualified in accordance with the Civil Aeronautics Administration and company requirements. No evidence indicated any pre-existing medical conditions that might have adversely affected the flight crew’s performance during the occurrence flight.
2. The airworthiness and maintenance of the occurrence aircraft were in compliance with the extant civil aviation regulations. There were no aircraft, engine, or system malfunctions that would have prevented normal operation of the aircraft.
3. All available evidence, including extensive simulations, indicated that the enhanced ground proximity warning system (EGPWS) functioned as designed.
4. The enhanced ground proximity warning system (EGPWS) manufacturer’s latest generation EGPWS equipment would have provided flight crews with an additional warning if an aircraft encountered similar circumstances to the occurrence flight. Installing the latest EGPWS equipment on the occurrence aircraft would have required approved modifications.
5. According to the Civil Aeronautics Administration (CAA) regulations, a 420 metre simple approach lighting system should have been installed to help pilots visually identify runway #The CAA advised that the Runway End Identification Lights, a flashing white light system, was installed at the runway’s threshold as an alternative visual aid to replace the simple approach lighting system.
6. From the perspective of flight operations, the runway 20 VOR missed approach point (MAPt) was not located in an optimal position. With the same Obstacle Clearance Altitude, if the MAPt had been set closer to the runway threshold, it would have increased the likelihood that flight crews (would be able) to visually locate the runway.
7. During holding, the occurrence flight crew requested the runway 02 instrument landing system (ILS) approach after receiving the weather information that the average wind speed for runway 02 had decreased to below the tailwind landing limit. While the decision for the use of the reciprocal runway was still under consideration by the Magong Air Force Base duty officer, the weather report indicated that the visibility had improved to 1,600 meters, which met the landing visibility minimal requirement for an approach to runway 20. The flight crew subsequently amended their request and elected to use runway 20.
8. At the time of the occurrence, the weather information exchange and runway availability coordination between civil and military personnel at Magong’s joint-use airport could have been more efficient.
9. ATR’s flight data recorder (FDR) readout document contained unclear information. That affected the efficiency of the occurrence investigation.

A total of 29 Safety Recommendations were made as a result of the Investigation as follows:

• that TransAsia Airways should implement effective safety actions to rectify the multiple safety deficiencies previously identified by the Aviation Safety Council investigations, internal and external Civil Aeronautics Administration audit and inspection findings, and deficiencies noted in this report to reduce the imminent safety risks confronting the airline. [ASC-ASR-16-01-010]

• that TransAsia Airways should conduct a thorough review of the airline’s safety management system and flight crew training programs, including crew resource management and threat and error management, internal auditor training, safety management system (SMS) training and devise systematic measures to ensure:
  • Flight crew check and training are standardized;
  • All flight crews comply with standard operating procedures (SOPs);
  • Staff who conduct audits receive appropriate professional auditor training;
  • All operational and senior management staff receive SMS training, including thorough risk assessment and management training; and
  • Proportional and consistent rules, in accordance with a “Just Culture”, are implemented to prevent flight crew from violating the well-designed SOPs and/or being engaged in unsafe behavior. [ASC-ASR-16-01-011]

• that TransAsia Airways should conduct a rigorous review of the safety management system (SMS) to rectify the significant deficiencies in:
  • Planning;
  • Organizational structure, capability and resources;
  • Risk management processes and outputs;
  • Flight operations quality assurance (FOQA) limitations and operations, including inadequate data analysis capabilities
  • Safety meetings;
  • Self-audits;
  • Safety performance monitoring, including risk indices;
  • Safety education; and
  • Senior management commitment to safety. [ASC-ASR-16-01-012]

• that TransAsia Airways should rectify the human resources deficits in the flight operations division and the safety and security office, including:
  • Crew shortages;
  • Inadequate support staff in the Flight Standards and Training Department, including insufficient standards pilots and crew to conduct operational safety risk assessments; and
  • Safety management staff with the required expertise in flight operations, safety and flight data analytics, safety risk assessment and management, human factors, and safety investigations. [ASC-ASR-16-01-013]

• that TransAsia Airways should review and improve the airline’s internal compliance oversight and auditing system and implement an effective corporate compliance and quality assurance system to ensure that oversight activities provide the required level of safety assurance and accountability. [ASC-ASR-16-01-014]

• that TransAsia Airways should implement an effective safety management process, such as a data-driven fatigue risk management system (FRMS), to manage the flight safety risks associated with crew fatigue]. [ASC-ASR-16-01-015]

• that TransAsia Airways should provide flight crew with adequate fatigue management education and training, including the provision of effective strategies to manage fatigue and performance during operations. [ASC-ASR-16-01-016]

• that TransAsia Airways should implement an effective change management system as a part of the airline’s safety management system (SMS) to ensure that risk assessment and mitigation activities are formally conducted and documented before significant operational changes are implemented, such as the introduction of new aircraft types or variants, increased operational tempo, opening new ports, and so on. [ASC-ASR-16-01-017]

• that TransAsia Airways should implement a more advanced flight operations quality assurance (FOQA) program with adequate training and technical support for the FOQA staff to ensure that they can exploit the analytical capabilities of the program. As such, the FOQA staff can more effectively identify and manage the operational safety risks confronting flight operations. [ASC-ASR-16-01-018]

• that TransAsia Airways should implement an effective standard operating procedures (SOPs) compliance monitoring system, such as the line operations safety audit (LOSA) program, to help identifying threats to operational safety and to minimize the associated risks. The system should adopt a data-driven method to assess the level of organizational resilience to systemic threats and can detect issues such as habitual non-compliance with SOPs. [ASC-ASR-16-01-019]

• that the Civil Aeronautics Administration should strengthen surveillance on TransAsia Airways to assess crew’s discipline and compliance with standard operating procedures (SOPs). [ASC-ASR-16-01-020]

• that the Civil Aeronautics Administration should implement a more robust process to identify safety-related shortcomings in operators’ operations, within an appropriate timescale, to ensure that the operators meet and maintain the required standards. [ASC-ASR-16-01-021]

• that the Civil Aeronautics Administration should provide inspectors with detailed guidance on how to evaluate the effectiveness of an operator’s safety management system (SMS), including:
  • Risk assessment and management practices;
  • Change management practices;
  • Flight operations quality assurance (FOQA) system and associated data analytics; and
  • Safety performance monitoring. [ASC-ASR-16-01-022]

• that the Civil Aeronautics Administration should provide inspectors with comprehensive training and development to ensure that they can conduct risk-based surveillance and operational oversight activities effectively. [ASC-ASR-16-01-023]

• that the Civil Aeronautics Administration should enhance inspector supervision and performance evaluation to ensure all inspectors conduct surveillance activities effectively and are able to identify and communicate critical safety issues to their supervisors. [ASC-ASR-16-01-024]

• that the Civil Aeronautics Administration should enhance the oversight of operators transitioning from traditional safety management to safety management systems. [ASC-ASR-16-01-025]

• that the Civil Aeronautics Administration should develop a systematic process for determining the relative risk levels of airline operators. [ASC-ASR-16-01-026]

• that the Civil Aeronautics Administration should review the current regulatory oversight surveillance program with a view to implementing a more targeted risk-based approach for operator safety evaluations. [ASC-ASR-16-01-027]

• that the Civil Aeronautics Administration should ensure (that) all safety recommendations issued by the occurrence investigation agency are implemented by the operators. [ASC-ASR-16-01-028]

• that the Civil Aeronautics Administration should develop detailed guidance for operators to implement effective fatigue risk management processes and training. [ASC-ASR-16-01-029]

• that the Civil Aeronautics Administration should review runway approach lighting systems in accordance with their existing radio navigation and landing aids to ensure that adequate guidance is available for pilots to identify the visual references to the runway environment, particularly in poor visibility condition or at night. [ASC-ASR-16-01-030]

• that the Civil Aeronautics Administration should review the design procedures for determining the location of missed approach point with the intention of increasing the likelihood of pilots to locate the runway without compromising the required obstacle clearance altitude. [ASC-ASR-16-01-031]

• that the Civil Aeronautics Administration should request TWR controllers to advise the flight crews of aircraft on final approach of the updated information in accordance with the provisions of the air traffic management procedures (ATMP). [ASC-ASR-16-01-032]

• that the Civil Aeronautics Administration should coordinate with Air Force Command Headquarters to review and improve the weather information exchange and runway availability coordination between civil air traffic control and military personnel at Magong Airport. [ASC-ASR-16-01-033]

• that ATR-GIE Avions de Transport Régional should evaluate the feasibility of a modification to allow the (latest version) of the Enhanced Ground Proximity Warning system (EGPWS) to be fitted on all ATR72-500 aircraft. [ASC-ASR-16-01-034]

• that ATR-GIE Avions de Transport Régional should review the flight data recorder (FDR) readout document for any erroneous information and provide timely revisions of the manual to assist airline operators and aviation occurrence investigation agencies. [ASC-ASR-16-01-035]

• that Air Force Command Headquarters, Ministry of National Defence should coordinate with the Civil Aeronautics Administration to ensure the reliability and validity of automated weather observation system (AWOS) runway visual range (RVR) sensors and their data. [ASC-ASR-16-01-036]

• that Air Force Command Headquarters, Ministry of National Defence should conduct the runway visual range (RVR) reporting operations and requirements in accordance with the provisions of the Air Force Meteorological Observation Manual. [ASC-ASR-16-01-037]

• that Air Force Command Headquarters, Ministry of National Defence should coordinate with the Civil Aeronautics Administration to review and improve the weather information exchange and runway availability coordination between civil air traffic control and military personnel at Magong Airport. [ASC-ASR-16-01-038]

The Final Report was published on 29 January 2016. It noted that a number of Safety Actions had been reported as taken in response to the investigated accident by both the CAA and TransAsia Airways but at the time of publication, these had not been "verified by the Aviation Safety Council".

Further Reading

• CFIT Precursors and Defences
• Controlled Flight Into Terrain (CFIT)
• Cross-checking Process
• Missed Approach
• Energy Management during Approach
• Loss of Control
• Non-Precision Approach
• Safety Management System
• Approach and Landing Accidents (ALA)
• Approach and Landing Risks