International

The Civil Aviation Authority of Vanuatu (CAAV) investigated a fuel starvation and collision with terrain accident involving Britten-Norman Islander BN2A-20, YJ-AT2, 6 km east-south-east of Port Vila International Airport, Vanuatu, on 15 July 2024. The CAAV requested assistance from the Australian Government represented by the ATSB.

To facilitate this support and to provide the appropriate protections for the information, the ATSB appointed an accredited representative in accordance with paragraph 5.23 of the International Civil Aviation Organization Annex 13 and commenced an investigation under the Australian Transport Safety Investigation Act 2003.

On 14 August 2025, the CAAV released its final report into this accident. This report is available here.

Any enquiries relating to the investigation should be directed to the CAAV.

The Transportation Safety Board (Junta de Seguridad en el Transporte - JST) of Argentina requested technical assistance from the ATSB to support its investigation into the collision between 2 trains on the railway bridge over Figueroa Alcorta Avenue, Palermo, Buenos Aires, that occurred at about 1030 local time on 10 May 2024.

Passenger train 3353 and work train T3021 were both operated by the State Railway Operating Company. Train 3353 was made up of locomotive B951 and 7 passenger cars. The T3021 work train consisted of locomotive 1000 and one car without passengers. The collision resulted in damage to the track infrastructure and rolling stock. It was reported by the JST that there were about 90 injuries among passengers and railway personnel.

To facilitate this support and to provide the appropriate protections for the information, the ATSB commenced an investigation under the Transport Safety Investigation Act 2003. At the request of the JST, the ATSB provided feedback in September 2024 on their preliminary investigation report. No further requests for follow-up assistance have been received. The ATSB's support to the JST has now been completed. 

The JST is responsible for the investigation and release of the final investigation report regarding this accident. Once completed, a copy of the report can be accessed here.

Any enquiries relating to the investigation should be directed to the JST: info@jst.gob.ar

On 8 June 2024, a de Havilland Canada DHC-6 Twin Otter, registered P2-BBM, had a runway excursion at Kikori Airstrip, Papua New Guinea.

The Papua New Guinea Accident Investigation Commission (AIC) is investigating the occurrence. AIC has requested assistance from the ATSB to decompress recovered flight data. 

To facilitate this support and to provide the appropriate protections for the information, the ATSB appointed an accredited representative in accordance with paragraph 5.23 of the International Civil Aviation Organization Annex 13 and commenced an investigation under the Australian Transport Safety Investigation Act 2003. 

The ATSB provided the decompressed file to the AIC on 19 June 2024.

Any enquiries relating to the investigation should be directed to AIC.

On 19 May 2024, a Tecnam P2600T aircraft was being operated on a flight from Salakanagara Tanjung Lesung airfield, Banten, Indonesia to Pondok Cabe Airport, Jakarta, Indonesia, with 3 people on board. At 1344 local time, on approach to Pondok Cable Airport at an altitude of 1,000 ft, the pilot declared a MAYDAY. Shortly after, the aircraft collided with terrain near a main road. All occupants were fatally injured. 

The Indonesian National Transportation Safety Committee (NTSC) is investigating this occurrence. 

NTSC has requested assistance from the ATSB to recover data from a damaged GPS unit. To facilitate this support and to provide the appropriate protections for the information, the ATSB appointed an accredited representative in accordance with paragraph 5.23 of the International Civil Aviation Organization Annex 13 and commenced an investigation under the Australian Transport Safety Investigation Act 2003.

The ATSB has completed its work recovering the GPS data. A copy of the data and a report detailing the work undertaken by the ATSB was provided to the NTSC on 19 July 2024.

Any enquiries relating to the investigation should be directed to NTSC.

The Transport Accident Investigation Commission (TAIC) of New Zealand has commenced an investigation into a runway excursion involving an Airbus A320, registered VH-VFF, at Christchurch Airport, New Zealand, on 30 May 2024.

During descent while conducting a passenger flight from Auckland, New Zealand, to Christchurch, the crew received a warning for one of the hydraulic systems, affecting the nose wheel steering. On landing, the aircraft had a runway excursion and made contact with a runway sign causing damage to the right engine. There were no reported injuries to crew or passengers and minor damage to the aircraft.

The TAIC requested assistance and the appointment of an accredited representative from the ATSB. To facilitate this support and to provide the appropriate protections for the information, the ATSB appointed an accredited representative in accordance with paragraph 5.23 of Annex 13 to the Convention on International Civil Aviation and commenced an investigation under the Australian Transport Safety Investigation Act 2003.

On 19 March 2026, the TAIC released the final investigation report into this accident. Accordingly, the ATSB has concluded its involvement in the investigation. A copy of the report can be obtained from the TAIC at: https://taic.org.nz/inquiry/ao-2024-004.

Any enquiries relating to the investigation should be directed to the Transport Accident Investigation Commission at https://taic.org.nz/.

On 18 February 2024, during a winch launch at approximately 700 feet above ground level, the right wing of the LAK-12 glider failed, resulting in a collision with terrain. The pilot was fatally injured.

The New Zealand Civil Aviation Authority (CAA) is investigating this occurrence. CAA has requested assistance and the appointment of an accredited representative from the ATSB.

To facilitate this support and to provide the appropriate protections for the information, the ATSB appointed an accredited representative in accordance with paragraph 5.23 of the International Civil Aviation Organization Annex 13 and commenced an investigation under the Australian Transport Safety Investigation Act 2003.

The ATSB was requested to examine and assess the possibility of recovering data from a damaged memory card. The ATSB x-rayed the memory card to enable an assessment and determined that the internal damage was such that the data was not recoverable by the ATSB.

With the completion of the examination and assessment and the return of the evidence to the CAA, the ATSB has concluded its involvement in the investigation of this accident.

Any enquiries relating to the investigation should be directed to CAA New Zealand.

The United States National Transportation Safety Board (NTSB) is investigating an accident involving an amateur-built Beck Michael J Sonex, registered N920MB, near Auburn Municipal Airport, Washington, United States, at about 1130 local time on 16 January 2024. 

The pilot reported that while climbing after take-off, a partial engine power loss occurred, and the aircraft began descending. The pilot then planned to conduct a forced landing on a road straight ahead of the aircraft but several seconds later, the engine regained power enabling a climb. About 40 seconds later, the engine lost all power and the pilot turned the aircraft toward the airport. The aircraft then collided with signage structures and terrain and was substantially damaged. The pilot was not injured. 

As part of the investigation, the NTSB identified an Australian-manufactured Rotec Aerosport TBI Mk 1 throttle body in the aircraft wreckage and considered it an item of interest. To support the investigation, the NTSB requested assistance and the appointment of an accredited representative from the ATSB to facilitate manufacturer examination and functional testing of the throttle body.

To facilitate this support and to provide the appropriate protections for any relevant information, the ATSB appointed an accredited representative in accordance with paragraph 5.23 of the International Civil Aviation Organization Annex 13 and commenced an investigation under the Australian Transport Safety Investigation Act 2003.

To ensure the integrity of the material evidence and to provide independent oversight, ATSB personnel were required to be present for the Australian manufacturer’s examination and testing of the throttle body. However, Rotec Aerosport would not permit ATSB personnel to be present and therefore, the NTSB elected not to proceed with the testing.

As no further involvement has been requested, the ATSB has closed this investigation.

Investigation summary

What happened

On 23 February 2024, an Alliance Airlines Embraer E190-100 aircraft, registered VH-UYI, was operating a scheduled passenger flight, IE700, on behalf of Solomon Airlines, from Honiara, Solomon Islands to Brisbane, Queensland. On board were the captain as pilot monitoring (PM), first officer as pilot flying (PF), 2 cabin crew and 66 passengers. 

Prior to conducting the Before-start checklist, the left seat pilot (captain) was required to set the aircraft’s speed mode selector, with flight management system (FMS) mode recommended. While the crew’s intention was to depart in FMS speed mode, undetected by either flight crew, this step was omitted, and the speed selection remained in manual mode.

As the aircraft was climbing through 1,200 ft, vertical flight level change (VFLCH) mode engaged in the FMS. Once in VFLCH, the flight director (FD) commenced targeting the manual target speed which, at that time, was 125 kt. 

Over the next 20 seconds, the aircraft’s speed gradually reduced and the pitch gradually increased to target the manual speed. Detecting that the aircraft was not accelerating, the PM assessed there was too much drag on the aircraft and retracted one stage of flap. The flap retraction resulted in a visual low airspeed cockpit alert. 

Shortly afterwards, the crew detected the speed mode was incorrectly set, and changed the speed mode to FMS mode, at which time the aircraft commenced accelerating to the correct target airspeed.

What the ATSB found

The ATSB determined that the left seat pilot (captain) unintentionally left the speed selection in manual mode instead of flight management system mode with no manual speed set. The manual speed mode selection was not detected by either flight crew member, resulting in the aircraft decelerating after VFLCH mode engaged.

While the captain was monitoring traffic, weather and making a radio broadcast, the first officer was not effectively monitoring the airspeed and as a result, did not initially detect the aircraft decelerating. Having assessed that the low airspeed was due to excessive drag, the captain retracted one stage of flap while below the minimum flap target speed, resulting in the aircraft entering a low-speed state.

Alliance Airlines’ standard operating procedures manual (SOPM) included a step of initially selecting the speed knob to ‘manual’ in its pre-flight procedures, despite that mode very rarely being used for take-off. This increased the risk of flight crews departing with the manual speed mode unintentionally selected. The ATSB also identified that Alliance’s right seat pilot shutdown flow was undocumented and not in accordance with the manufacturer’s guidance.

Additionally, Embraer's airplane operations manual was inconsistent with its standard operating procedures manual in regard to setting the speed knob to manual in the Before start procedures. 

Likely due to a training deficiency, Alliance Airlines flight crews' conduct of the Before start procedures and Pre-take-off brief review were not being performed effectively to ensure the speed selector knob was correctly set and checked, which increased the risk of a low-speed event after take-off.

A review by the operator of flight data found that in 112 flights operated over a 30-month period, flight crew had also not selected, or detected, the speed selector knob in manual mode prior to commencing the take-off run. In 76 of these events, flight crew were changing the speed selector knob setting during the take‑off run, increasing the risk of distraction during a critical phase of flight.

What has been done as a result

Action was taken by Alliance Airlines to address the safety issues identified in this investigation, including:

• E190 SOPM pre-flight procedures have been amended so the left seat pilot selects FMS mode to mitigate unintentionally departing in manual speed mode.
• The practice of the right-seat pilot setting manual speed and 80 knots during the after‑shutdown flow will be discontinued. Recurrent training and check events will reinforce compliance with the correct procedure.
• Training and procedural guidance to reinforce the correct setting and verification of the speed mode selector knob is being enhanced. Flight crews will receive targeted refresher training through a dedicated simulator training module, emphasising correct procedural discipline for conducting pre-take-off reviews.
• Flight crews will receive targeted instruction emphasising that no adjustments to the speed selector knob are to be made during the take-off roll. This will be reinforced during initial, recurrent and line check assessments. Compliance with this policy will be monitored through the collection and analysis of flight data and will be assessed during recurrent simulator training and annual line proficiency checks.

Proactively, Alliance Airlines has also:

• Issued an article to flight crew highlighting that at the completion of the before start duties, the left seat pilot should apply the technique of selecting ‘TOGA, TARA, SPEED’, followed by confirmation of the relevant modes and settings on the primary flight display, prior to calling for the completion of the Before start checklist.
• Advised that a dedicated training module will be incorporated into recurrent simulator cyclic training exercises commencing 1 January 2025. This module will include a review of this occurrence, its root cause, and a reinforced focus on correct procedures and techniques to prevent reoccurrences.
• Advised that during recurrent line check events, there will be an added emphasis on reviewing and ensuring adherence to the correct procedures and techniques.

Finally, Embraer advised that it proposes to align the contents of the AOM and SOPM in the first half of 2025, likely removing reference to selection of manual mode in the Before start procedure.

Safety message

This incident highlights how important continuous attention to the modes displayed on the primary flight display is to situation awareness. An ICAO safety advisory on Mode awareness and energy state management aspects of flight deck automation stated that loss of mode awareness and mode confusion have been identified as factors in several major accidents around the world. Further, that mode confusion can often result in flight crews' mismanagement of an aircraft's energy state, such as the low-speed state that occurred in this incident. The circular recommended that at any time an aircraft does not follow the desired vertical or lateral flight path, or airspeed, flight crew should adapt the level of automation to the task and/or circumstances, or revert to hand flying or manual thrust/throttle control, if required.

The circular also identified inadequate training and system knowledge as a key factor contributing to mode confusion. In this case, training in quick identification of mode indications, including speed display colours during pre-take-off checks, would reduce the likelihood of a similar incident occurring.

The investigation

The occurrence

On 23 February 2024, an Alliance Airlines Embraer E190-100 aircraft, registered VH-UYI, was operating a scheduled passenger flight, IE700 on behalf of Solomon Airlines, from Honiara, Solomon Islands to Brisbane, Queensland. On board were the captain as pilot monitoring (PM), first officer as pilot flying (PF),[1] 2 cabin crew and 66 passengers. The flight was scheduled to depart at 1400 local time, but due to several delays (see the section titled Pre-departure delays), departed at 1507. 

As the crew were preparing the aircraft, the captain identified that they were likely to encounter a thunderstorm on the track of the planned standard instrument departure (SID). The captain advised that, in order to avoid it, they suggested to the first officer that they conduct a visual departure. However, the first officer reported being ‘comfortable’ with the weather and advised their preference to fly the SID, which was then agreed. The SID required that the aircraft initially climb to 1,000 ft, then conduct a 180° left turn to intercept the outbound track. 

The operator’s pre-flight procedures required the left seat pilot (LSP) (captain) to set the speed knob to manual mode (see the section titled The operator split the Before start procedures between Pre-flight and Before start procedures). This resulted in a target speed setting of 80 kt. 

Prior to conducting the Before start checklist, the LSP was then required to action the Before start flow[2] which again included setting the speed mode, with the flight management system (FMS) mode recommended. While the crew's intention was to depart in FMS mode, undetected by either flight crew, this step was omitted, and the speed mode selection remained in manual. 

The flight crew then configured the aircraft for a flap 4 take-off (see the section titled Flaps) due to the warm, humid conditions and the 2,200 m long runway. The PF armed take-off (TO) mode (see the section titled Vertical modes) in the FMS. In that mode, the flight director (FD) (see the section titled Flight guidance control system) displayed a pitch attitude for the crew to manually follow during rotation and the initial climb.[3]

The PF then commenced the take‑off, selecting TOGA[4] power. After the aircraft rotated, the aircraft’s controller logic (see the section title Controller logic) automatically changed the target speed from 80 kt to V₁,[5] which was 125 kt. That change in target speed was displayed on the primary flight display (PFD) (in cyan) but was not detected by either flight crewmember.

The PF manually followed the FD guidance for pitch attitude and the aircraft maintained an airspeed of about V₂[6] plus 10 kt (144 kt). In accordance with the operator’s standard operating procedures manual, the PF engaged the autopilot at about 1,000 ft, and subsequently, as the aircraft climbed through 1,200 ft, vertical flight level change (VFLCH) engaged (see the section titled Vertical modes). Once in VFLCH, the FD commenced targeting the target speed, which was 125 kt. 

A few seconds later, the aircraft entered a 25° left turn to comply with the SID. At about that time, the PF reduced the power from TOGA power to climb power, and the PM diverted their attention to visually assessing weather on the departure route and monitoring the traffic collision avoidance system for other aircraft. Over the next 20 seconds, the aircraft’s speed gradually reduced while the PM made a departure broadcast on the common traffic advisory frequency (CTAF).[7]

The PM recalled that at that stage of the flight the aircraft felt ‘draggy’. In response, they looked at the PFD and saw the airspeed indicating around 134 kt and recognised that the aircraft was not accelerating. The PM instructed the PF to ‘roll out’ of the banked turn, to reduce drag. However, the PF did not follow that instruction, likely due to believing the call to roll out was regarding the weather ahead, which the PF was not concerned about.   

The PM recalled further advising the PF ‘you’re not accelerating’, while also being unsure why the aircraft was not accelerating. Knowing that the landing gear was already retracted, the PM retracted flap from 4 to 3 (slats from 25° to 15°) to further reduce the drag. The PF recalled hearing the PM stating that they were selecting flap 3. After the slats were retracted, the speed further reduced to 131 kt and the PF detected the amber pitch limit indicators (see the section titled Pitch limit indicators) on the PFD and alerted the PM accordingly. 

The PF then decided to change the speed mode from FMS to manual, to manually increase the airspeed. However, when they went to change the speed selector knob, they detected that it was already in manual mode and so changed the selection to FMS mode. This occurred 21 seconds after the flap handle was moved to the flap 3 detent. Once the speed mode was in FMS mode, the aircraft accelerated to the correct target speed of 190 kt. 

Context

Flight crew information

The captain and first officer both held an air transport pilot licence (aeroplane) and a class 1 aviation medical certificate. The captain had accumulated 17,000 flight hours, and the first officer had about 13,000 flight hours of experience.

Both crewmembers reported feeling well rested prior to the flight, however, the captain reported feeling ‘mentally tired’ prior to take-off, following ground-handling and dispatch irregularities and delays. 

Airspace information 

Honiara Airport operated a flight information service[8], with the surrounding airspace being non‑controlled. That is, the airspace had no active supervision by air traffic control and pilots were responsible for their own separation from other traffic.

Aircraft information 

The aircraft was an ERJ 190-100 IGW, manufactured in Brazil in 2006 and issued serial number 19000053. It was registered in Australia as VH-UYI on 24 January 2022. The aircraft was fitted with 2 General Electric Company CF34-10E5 turbofan engines. It had an integrated avionics system with either ‘load 25’ or ‘load 27’ software installed. VH-UYI had load 27 software. 

Flight management system and manual speed modes

Flight guidance control system

According to the E190 maintenance manual, the flight guidance control system (FGCS) has 2 relevant functions:

• flight director (FD) guidance: The FGCS calculates the FD guidance commands that show on the primary flight display (PFD). The FD is selected by pressing a button on the guidance panel (GP).
• autopilot: this sends automatic pitch and roll guidance to the elevator and aileron servos. Its control authority is limited to keep the aircraft within a safe operating envelope. It is engaged by the AP button on the GP.

The aircraft also had an autothrottle, which the operating policy stated should be used during the entire flight, engaged just prior to take-off and disengaged after touchdown or at the pilot flying (PF) discretion.

The avionics pilot guide stated: 

When engaged, the A/T system automatically positions the thrust levers to control the aircraft thrust throughout the flight regime. The A/T system keeps the aircraft within the thrust and speed envelopes and controls the engine thrust modes in synchronization with the active FGCS modes.

Flight management system

According to the avionics pilot guide, the flight management system (FMS) is: 

an integrated system providing data for the cockpit displays and flight control system (FCS). The FMS serves as an aid to performance, flight planning, navigation, database, and redundancy management.

The FMS is used for complete flight planning activities… Once programmed, the FMS gives control outputs to the autopilot system to fly the aircraft along the planned route, both laterally and vertically.

Speed modes

The flight crew had the option of selecting either of 2 speed management modes on the GP – FMS or manual mode (Figure 1).

Figure 1: E190 speed selector knob on guidance panel

Source: Alliance interim safety report

In FMS mode, the airspeed selection was managed by the FMS. The airspeed was displayed on the PFD in magenta (Figure 2). 

In manual mode, flight crew retained responsibility for airspeed control, by selecting the airspeed using the SPEED selector knob, which then displayed the selected airspeed on the PFD (Figure 2). This target speed was used by both the autothrottle and the autopilot. A manual selected airspeed was displayed on the PFD in cyan. 

Figure 2: FMS and manual speed displays

Example of speed tape showing the different speed modes. FMS mode on the left and manual mode on the right. Source: Honeywell, annotated by the ATSB

Prior to take-off, the pilot entered the required V speeds[9] from the multifunction control display unit (MCDU). These were then displayed on the lower portion of the speed tape. 

The operator specified that if manual speed mode was selected for take-off:

• when conducting a take-off in flap 1 to 3 configuration, the airspeed was to be set to final segment speed (V_FS), which was the speed to be achieved during the final take-off segment, with landing gear up and flaps retracted
• when conducting take-off in a flap 4 configuration, the airspeed should be set to 175 kt
• when using noise abatement procedures, the airspeed should be set to V₂ +10 kt.  

The selection of speed mode was referenced in pre-flight, before-start and shutdown procedures (see the section titled Mode selection procedures). 

Vertical modes

The autoflight control system (AFCS) has 11 modes to control the aircraft’s flight path. Two modes were relevant to this incident – take-off (TO) and vertical navigation flight level change (VFLCH). 

In TO mode, the flight crew fly with reference to the flight director (FD). The flight crew set take‑of/go around (TOGA) power and then manually rotate the aircraft to the displayed pitch attitude on the FD to achieve the required airspeed.

When the mode was changed to a mode other than TO, the AFCS will start using the target airspeed as a reference. In VFLCH, the autothrottle holds the thrust lever at the set thrust value and the FD changes the pitch angle to maintain the airspeed at the target airspeed selected on the PFD. According to the operator’s standard operating procedures manual (SOPM), VFLCH is the preferred climb mode. 

Flaps

The E190, has both flaps and slats. With the flap lever in position 4, the flaps extend to 20° and the slats to 25º. Moving the flap lever to position 3 moves the slats to 15º, with the flap position remaining unchanged.

With slats extended, the critical angle of attack (AOA)[10] is increased, enabling the aircraft to operate at a greater AOA. Conversely, once slats are retracted, the critical AOA decreases. 

Flap retraction – F-Bug

The PFD shows the ideal flap selection speed using a symbol (green dot) on the speed tape. The flap retraction speed is shown using a magenta bug (F-bug). The flap manoeuvring speed is calculated based on the airplane weight and slat/flap setting and does not change with bank angle or turbulence.

The operator’s E190 aircraft operations manual (AOM) stated that:

During flap retraction, the next flap setting should be selected when the F-Bug is reached.

The F-Bug calculation is designed to meet minimum safe margins to VFE[11] and shaker speed. A minimum margin of 20% above the stall speed[12] is set for the next flap.

F-bug speeds calculated for the flight, provided by the manufacturer:

Stall warning protection system

Overview 

The E190 stall warning protection system (SWPS) is a 2-stage system that warns and protects the aircraft from aerodynamic stall conditions. The first stage warns the pilot of the impending stall by:

• showing a low-speed awareness indication on the airspeed tape
• showing a pitch limit indication on the attitude direction indicator on the primary flight display
• activating the stick shaker motor on each control column results in each control column to shake (simulating the aircraft buffeting).

The second level is an AOA limiter protection system that limits the maximum AOA to a safe value below the predicted aerodynamic stall (preventing a stall).

Low-speed awareness system

A low-speed awareness (LSA) indicator is displayed along the lower-right side of the airspeed tape (Figure 3Figure 3). LSA varies with the load factor[13] of the aircraft and the bar position is based on airspeed, aircraft configuration and AOA. The bar rises from the bottom of the tape to show both the calculated stall and stick shaker[14] speed (Vshaker) in 2 coloured ranges:

• the amber range displays from Vshaker to 1.13 x stall speed
• a stall occurs at the top of the red range, when the airspeed drops below Vshaker. 

If the speed enters the red range, the stick shaker will activate, and an audible alarm will sound. The speed displayed on the PFD will also change colour to red reverse video.

Figure 3: Low speed awareness indicator

Source: Honeywell, annotated by the ATSB

Controller logic

While in TO mode with the speed control in manual, if the speed set is lower than V_1, the controller logic automatically increases the speed to V₁, after the ground‑to‑air transition. 

Further, the aircraft manufacturer and avionics manufacturer advised that the controller logic ensures that the LSA indicator does not allow the selected speed to stay in the red range. If the manual speed selected on the guidance panel was in the amber band, the LSA automatically increased the speed to target the top of the amber band. 

In this instance, the manual speed target initially increased to 125 kt after rotation and then, due to load effects during the climbing turn, the top of the amber band (and the target airspeed) further increased to 131 kt.

The aircraft manufacturer advised that the aircraft’s controller logic would always attempt to avoid unsafe conditions within a given envelope. The protective measures would not stop working unless the limit of conditions were met – in which case they would be announced to the crew by way of speed displayed in amber on the LSA indicator, the activation of pitch limit indicators (PLI) (see the section titled Pitch limit indicators) and eventually a stick shaker. 

Autothrottle low speed protection

When the autothrottle was providing speed control, it would ensure both high and low speed envelope protection. The manual target speed was limited to the minimum maneuvering speed, flap/gear placard speed or the low speed awareness speed. The autothrottle lower speed limit was the greater of the manual speed target or 1.2 the stall speed. When flaps were extended, the autothrottle would maintain 1.2 x stall speed.

However, when the system was in VFLCH mode, airspeed was controlled by the elevator, with the autothrottle targeting a fixed thrust setting and therefore not providing low speed protection. The FD pitch controller providing speed control did not contain low speed protection and targeted the selected target speed on the guidance panel.

Pitch limit indicators

The aircraft’s SWPS computed a pitch limit indicator (PLI). This was a ‘pitch-based indication of the margin (in degrees) between the stick shaker speed and the current airspeed’ (Figure 4). The PLI margin was calculated continuously and displayed on the PFD when the airspeed was less than 1.2 the stall speed.

Figure 4: Pitch limit indicators

Source: Honeywell, annotated by the ATSB

Operational information 

Mode selection procedures

The aircraft manufacturer provided operators with an airplane operations manual (AOM) and standard operating procedures manual (SOPM). 

Regarding the 2 documents, the aircraft manufacturer stated:

The intention of the AOM is to gather all the information related to the operation of the aircraft, while the Embraer SOP is generated to provide operational guidance (it is ‘our way’ of operating the aircraft).

Both can be used by the operator, who must produce their own procedures (usually, dealing with mixed fleets) … Thus, operators can create their SOP using our SOP as a starting point and can add or change some points using information from the AOM, or even mix some internal operational information, but the final set must be approved.

This is a flexibility that operators can take advantage of. However, in all cases, following the instructions of the Embraer AOM/SOP entirely, as well as the operator’s SOP, should not lead to undesirable conditions.

As long as the Embraer AOM or SOP is followed in full, no undesirable results will occur.

Manufacturer’s airplane operations manual 

The manufacturer’s AOM Before start procedures called for the speed selector knob to be set to manual:

SPEED Knob............................................. MAN

Subsequently, the Shortly before startup procedures (Figure 5), called for the speed selector knob to be set at pilot discretion, with FMS recommended for Load 27 aircraft (including VH-UYI).

Figure 5: Shortly before startup procedures

Source: Embraer E190 airplane operations manual – normal procedures

Manufacturer’s standard operating procedures manual (SOPM)

Unlike the AOM, the SOPM did not have a requirement to set the speed selector knob to manual in the Before start procedures. The first reference to speed selector knob was in the Shortly before start procedures (Figure 6), where the left seat pilot (LSP) was required to set it at pilot’s discretion (with FMS speed recommended). 

Figure 6: Shortly before start procedures

Source: Embraer E190 standard operating procedures  manual – normal procedures

The manufacturer advised that:

There is no need to set the speed to manual in the pre-flight, because once the necessary data is filled in the FMS (via MCDU), the correct speed target will be automatically set, provided that the selector is in FMS SPEEDS.

The step of bringing it [speed] to manual, mentioned in the AOM, is not included in our SOP, and is not taught in our [original equipment manufacturer] OEM training, as the SOP procedures are followed instead.

Operator’s standard operating procedures manual (SOPM)

The operator advised that: 

As is industry practice, Alliance Airlines adapted the manufactures E190 manual suite, including the E190 AOM and SOPM to develop SOPs that are suitable for our operation. Like other airlines, Alliance Airlines continuously monitors manufacturers recommendations and conducts ongoing reviews of our SOPs to ensure they are fit for purpose, compliant with regulations and follow industry best practice.

The operator split the Before start procedures between Pre-flight and Before start procedures 

Pre-flight procedures

Due to the alignment with the manufacturer’s AOM, the operator’s Pre-flight procedures required the LSP to set the speed selector knob to manual mode.

The operator understood this step was required to clear any previously set FMS speeds. However, the avionics manufacturer advised that: 

From the FMS side (Load 27), there is not a mandatory requirement to set it [speed knob] to manual and it does not have logic to reset the previous autospeed target by setting the speed selector to manual. Rather, the previous flight information is cleared 2 minutes after landing.

Although not documented in their SOPM, the operator advised that it was customary for the right seat pilot (RSP) to switch the speed selector knob to manual during a Shutdown flow, which was intended as an additional protection to ensure that the speed selector knob was selected to manual.

Before start procedures

As part of the Before start flow, the operator’s procedures required the LSP to set the speed selector knob to either manual or FMS, and for aircraft running load 27, FMS mode was the recommended mode. This was not required to be stated out loud, nor challenged by the right seat pilot. 

The Before start flow was termed ‘TOGA TARA SPEED’ as follows:

The LSP should perform the following actions

• Press the TO/GA [take-off/go-around] button to arm the Flight Director for take-off mode.

• Select TA/RA on the transponder.

• At the pilot’s discretion set the SPEED Knob to FMS or MANUAL. If MANUAL, set VFS on the speed window for Flap 1–3 take-off or 175kts for Flap 4 take-off.

• If required, select NAV for a LNAV [lateral navigation mode] departure. 

The crew then performed the Before start procedures, which again called for the speed selector knob to be set, with FMS mode recommended.

The engines were then started, and the relevant checklists completed.

The operator’s procedures subsequently included:

As late as practical approaching the take-off point, the PF for the sector should complete a take-off brief review in accordance with Alliance Airlines [Operations Policy and Procedures Manual] OPPM. This should also include items such as flap setting, runway and intersection, RNP status, assigned altitude setting as applicable and changed weather conditions.

However, the OPPM referred to did not include any information regarding speed settings. 

The manufacturer stated that: 

while the Embraer [standard operating procedures] SOP did not explicitly mention speeds, it was expected that operators' SOPs cover speeds and cross-check the parameters, which are: V1, VR, V2, VFS values and bugs on the speed tape and V2 magenta or VFS blue on the speed target. In this way, the take-off briefing is a second opportunity to verify the correct value set.

Take-off procedures 

The operator’s Take-off procedures stated: 

FMA modes are not called until 400 ft to allow the PM to call any abnormal conditions as well as to reduce distraction during the take-off roll.

The manufacturer advised that as the take-off is a critical phase of flight: 

Embraer’s SOPs recommend that selections and other heads-down tasks be avoided during this phase. The PF is flying the aircraft even while on the ground, and should be concentrating on maintaining directional control, cross-checking the instruments and alerting for any abnormal conditions. While the PM should be monitoring flight parameters (speed), engine parameters, the engine indicating and crew alerting system (EICAS) and monitoring the actions of the PF. Furthermore, the pilot in command is also monitoring for any abnormalities to make the Go/No-Go decision until V₁ is reached. 

Speed mode training

The operator advised that, although it was not documented in their training manual: 

Thorough simulator type rating training is conducted to consolidate understanding…

the general practice of confirming a selection using the PFD/ND [navigation display], rather than the position of a physical switch/lever, is best practice and a key principle of operating advanced aircraft. This philosophy is integrated into the Alliance E190 Type Rating program, with a strong emphasis on correct technique starting during the ground school phase.  

during line operations (and training), the before start procedures (scans), cover the requirement for confirmation that either cyan or magenta, are as anticipated, and that the calculated speeds are accurate.

The captain reported that manual speed would have been displayed in cyan, which should have stood out during the scans. They advised they had not used manual speed mode for take-off other than in the simulator during type rating training. 

The FO also reported not having used manual mode previously and had not specifically been told to check for FMS/manual speed displaying in cyan or magenta in the speed window. 

The manufacturer did not provide a specific training syllabus, rather expected the aircraft operator to develop the training manual. Therefore, type rating training for the aircraft was operator-specific. 

Review of other E190 operator’s procedures

The ATSB conducted a review of another E190 operator’s procedures and found that they did not have a requirement for the speed mode to be set to manual during pre-flight, before start procedures or shut down procedures. That is, the speed mode remained in FMS speed throughout all phases of flight. 

In addition, the operator had introduced into their SOPM, a verbal check of speed settings (FMS or Manual) as part of their FMS performance review prior to the flight.

Flight crew roles

The manufacturer established areas in the cockpit that were placed under the responsibility of a specific flight crewmember. Ground operations are divided between the left seat pilot (LSP) and the right seat pilot (RSP) while in-flight operations are divided between pilot flying (PF) and pilot monitoring (PM).

Regarding roles and responsibilities, the operator’s manual stated that:

The PF is responsible for controlling the vertical flight path and horizontal flight path and for energy management by either: 

• supervising the auto pilot vertical and lateral modes through awareness of modes being armed or engaged, mode changes and of selected mode targets; or

• hand flying the aircraft, with or without flight director guidance.

The PNF [pilot not flying – same concept as PM] is responsible for:

• systems related monitoring

• monitoring tasks

• performing the actions requested by the PF.

The manufacturer manuals stated that when the aircraft does not perform as expected, the autopilot must be disconnected and manual flight promptly established. The manufacturer further advised it is ‘primary airmanship to monitor airspeed during every phase of flight, especially during take-off and initial climb’. 

Pre-departure delays

Prior to departure, there were multiple irregularities involving ground handling, dispatch and loading sheets, passenger boarding and catering that required the captain’s attention. These included:

• final loading sheet with a 2-tonne discrepancy from initial loading sheet
• flight plan issued for Extended-range Twin-engine Operational Performance Standards (ETOPS) when not an ETOPS flight
• final paperwork approximately 50 pages instead of 3–4 pages of relevant information
• HF radio failure – requiring extra fuel to hold under 24,500 ft until ATC (Brisbane Centre) could be contacted
• requirement to leave the bay due to incoming company aircraft.

Resolution of these issues resulted in the aircraft pushing back 30 minutes later than scheduled and the captain reportedly feeling mentally drained by the time they entered the cockpit. Additional delays due to other traffic, a closed taxiway and a displaced runway threshold resulted in the aircraft departing a further 30 minutes after pushback. 

Recorded data

Quick access recorder data

Flight data from the aircraft’s quick access recorder (QAR) for the incident flight and the previous flight was analysed by the ATSB and the aircraft manufacturer. 

The data showed that the speed selector knob had been selected to manual after touchdown on the previous flight.

QAR data confirmed that TOGA power was set and LNAV was armed, which were both items on the Before-start ‘TOGA TARA SPEED’ flow, indicating that part of the captain’s flow was conducted. TA/RA, another required item on the flow, was not a recorded parameter. 

QAR data also confirmed that the captain was using the push to talk button as the aircraft’s speed decreased. 

Stall speed, the low speed awareness (LSA) indicator and the pitch limit indicators (PLI’s) were not recorded by the QAR, however, the manufacturer provided the ATSB with calculations based off the QAR data.

They advised that prior to the flap lever retraction from 4 to 3, the top of the amber band on the LSA was 124.5 kt. When the slats retracted, 20 kts prior to the target speed (F-bug), the upper value of the amber LSA indicator[15] increased to 135.1 kt.

Figure 7: QAR data during initial climb

 Source: ATSB 

The data showed that once VFLCH was engaged, the airspeed (red line) reduced towards the FMS target speed (green line) at 125 kt. When the flap lever was retracted (purple line), the low speed awareness indicator increased to approximately 135.1 kt (dotted orange line), which was above the current airspeed of 134 kt. The PLIs would have been displayed on the PFD at this stage. The target speed then increased when the speed mode was selected to FMS. 

Similar occurrences 

As part of the operator’s internal safety investigation into this incident, a review of QAR data was conducted on flights operated in manual speed. From January 2022–July 2024, 254 flights were identified where manual speed setting was used. In 112 of those (including this incident), the flight crew had manual speed selected with no associated target speed set (80 kt). 

In 76 of these 112 events, the QAR data showed that flight crew adjusted the speed mode selector to FMS during the take-off run and departed in FMS mode. 

The operator advised that a review of the last 50 events where speed had not been set correctly, 38 individual captains were involved, operating from a range of different bases throughout the operator’s network. 

Safety analysis

Introduction

During the pre-flight checks, undetected by either flight crew member, the captain unintentionally omitted to select the speed selector knob to flight management system (FMS), and the aircraft departed in manual mode without a target speed set. After take-off, with the airspeed at about 144 kt, vertical flight level change (VFLCH) mode engaged in the FMS, resulting in the flight director targeting the airspeed displayed on the primary flight display, of 125 kt. The aircraft then decelerated while in a banked turn with flap extended. The captain responded by retracting one stage of flap, resulting in a low-speed state. The flight crew then identified and corrected the speed mode selection. The stick shaker did not activate.

This analysis will discuss the flight crew’s actions in setting and monitoring modes and speeds, and flap retraction. The contribution of training and differences between the manufacturer’s and operator’s documentation will also be examined. Finally, other similar occurrences by Alliance Airlines flight crews and their responses will be examined. 

Speed mode selection

Recorded flight data showed that the speed selector knob had been set to manual, as per the operator’s procedure for the copilot’s flow at shutdown, following the previous flight and the Pre‑flight procedures. It also showed the selector was not switched to FMS mode, as per the captain’s intention in the Before start procedures. Neither flight crewmember detected the speed selector knob in manual or that the speed colour on the primary flight display (PFD) was cyan (compared to the expected magenta) in the before start scans or during the take-off brief review. 

After the aircraft rotated, the controller logic automatically increased the target speed to V₁, which was 125 kt. However, at that time, as the aircraft was in take-off (TO) mode, the target speed was not affected, as the pilot flying was following flight director TO mode pitch guidance, and the system was targeting the desired V₂ plus 10 kt (about 144 kt) for the take-off, and climb was maintained as expected. 

Passing 1,200 ft, vertical flight level change (VFLCH) mode engaged. At this time, the flight director (FD) began to target the selected airspeed (125 kt) by increasing the pitch of the aircraft, and the airspeed gradually began to decrease.

Not monitoring speed

During the initial climb in non‑controlled airspace, the captain was making the required departure call, and monitoring traffic and weather.

As the aircraft entered a 25° bank turn to comply with the standard instrument departure (SID), the airspeed further decreased. The speed reduction and increased pitch was not detected by the PF, which may have been due to anticipating a slower climb in flap 4 configuration and a banked turn.  

Flap retraction

Although the aircraft was slower and pitched higher than usual at that time, it was operating at a safe speed. Manufacturer calculations showed that the top of the LSA amber band was at approximately 125 kt during this time and the airspeed was decelerating through 134 kt. 

However, when the PM detected that the aircraft was not accelerating, they retracted flap from 4 to 3 (slat retraction) to reduce drag. The manufacturer’s calculations showed that this resulted in the upper value of the amber range on the LSA indicator increasing to about 135 kt. When slats were retracted, the airspeed was around 134 kt, therefore within the amber band, before further reducing to 131 kt. During that time, the pitch limit indicators (PLI’s) would have activated and been displayed on the primary flight display. 

The manufacturer required that if the aircraft was not performing as expected, the flight crew were required to disconnect the autopilot and manually fly the aircraft. 

Manufacturer’s airplane operations manual 

The manufacturer’s airplane operations manual (AOM) was inconsistent with their standard operating procedures manual (SOPM). Specifically, it required the speed selector knob to initially be set to ‘manual’ in the Before start procedures, which was not an SOPM requirement.

While the use of manual speed was valid, it was unnecessary for the AOM to require initial setting of manual speed in the Before start procedures, as the selection of speed mode was later made in the Shortly before start procedures. It also increased the risk of flight crews not switching the speed selector knob back to the commonly-used FMS speed prior to departure.  

Operator’s procedures 

While the manufacturer’s AOM was inconsistent with their SOPM, the operator advised that it had reviewed both documents when developing their own SOPM. 

Despite manual speed mode rarely being used, the operator included a requirement for it to be initially set to manual, and then selected to FMS prior to take-off, in line with the AOM. The inclusion of this requirement in its SOPM may have also been influenced by an incorrect belief that it was required to clear previous speeds entered in the flight management system (FMS). In that context, while well intentioned, it increased the risk of the aircraft departing with the incorrect speed mode selected.

In addition, the operator also included an undocumented right seat pilot flow at shut down to switch the speed selector knob to manual. This step was not in accordance with the aircraft manufacturer’s AOM or SOPM and similarly increased the risk of the incorrect speed being selected for take-off.

Training

While the operator stated that the training and line checks ensured that crews were checking that the speeds were displayed in cyan or magenta as expected, these requirements were not documented in the training manual. Additionally, the first officer (FO) reported never having been told to specifically look at the colour (magenta/cyan) of the speed display as part of their scans. 

The captain did not ensure the target speed was depicted in the expected colour. They had only conducted a take‑off with the speed selector knob in manual mode once, as part of their initial training in the simulator. 

Unlike standard operating procedures, type rating training was not stipulated by the manufacturer, instead was developed by each aircraft operator. The manufacturer was unaware of any other operator having had similar events. Given there were 112 occurrences where the operator’s flight crew had not correctly set the speed selector knob, nor detected the incorrect speed selector knob selection and the speed colour in cyan in the display prior to the take-off run, it is likely that the operator’s training and line checks were not adequate in ensuring flight crew were completing their scans as per the manufacturer’s and operator’s requirements. 

Similar incidents

A review of past occurrences identified 112 incidents (including this incident) in the preceding 30 months where flight crews had unintentionally not set the speed selector knob to FMS. In 76 of these events it was identified that the flight crew had adjusted the speed selector knob during the take-off run, a critical stage of flight. The manufacturer’s SOPM recommended that mode selections and other heads-down tasks be avoided during critical phases of flight. To mitigate that risk, the manufacturer’s and operator’s SOPM required that correction to the speed mode selection should be made when passing 400 ft on climb.

Findings

From the evidence available, the following findings are made with respect to the incorrect configuration involving Embraer E190, VH-UYI at Honiara Airport, Solomon Islands on 23 February 2024:

Contributing factors

• During the Before start procedure, the captain unintentionally left the speed selector knob in manual mode instead of flight management system mode, with no manual speed set. The manual speed mode selection was not detected by either flight crewmember, resulting in the aircraft decelerating after vertical flight level change mode was engaged.
• While the captain was monitoring traffic, weather and making a radio broadcast, the first officer was not effectively monitoring the airspeed and, as a result, did not initially detect the aircraft decelerating.
• Having assessed that the low airspeed was due to excessive drag, the captain retracted one stage of flap while below the minimum flap target speed, resulting in the aircraft entering a low‑speed state.
• Embraer's airplane operations manual was inconsistent with its standard operating procedures manual in relation to speed mode selection. This increased the risk of flight crews departing with the manual speed mode unintentionally selected. (Safety issue)
• Consistent with Embraer’s airplane operations manual, the Alliance Airline's pre-flight procedure required flight crew to unnecessarily initially set the speed knob to ‘manual’. This increased the risk of the aircraft departing with the incorrect speed mode selected. (Safety issue)
• Likely due to a training deficiency, Alliance Airlines flight crews' conduct of the Before start procedures and Pre-take-off brief review were not being performed effectively to ensure the speed selector knob was correctly set and checked, which increased the risk of a low-speed event after take-off. (Safety issue)

Other factors that increased risk

• Alliance Airlines’ right seat pilot shutdown flow was undocumented and not in accordance with the manufacturer’s guidance.
• Alliance Airlines flight crews were regularly changing the speed selector knob setting during the take‑off run. This was contrary to Embraer's guidance, and Alliance Airline’s own standard operating procedures manual. This increased the risk of distraction during a critical phase of flight. (Safety issue)

Safety issues and actions

Manufacturer’s manual inconsistencies 

Safety issue number: AO-2024-007-SI-07

Safety issue description: Embraer's airplane operations manual was inconsistent with its standard operating procedures manual in relation to speed mode selection. This increased the risk of flight crews departing with the manual speed mode unintentionally selected.

Operator’s standard operating procedures

Safety issue number: AO-2024-007-SI-05

Safety issue description: Consistent with Embraer’s airplane operations manual, the Alliance Airline's pre-flight procedure required flight crew to unnecessarily initially set the speed knob to ‘manual’. This increased the risk of the aircraft departing with the incorrect speed mode selected.

Multiple occurrences of the speed selector knob not being set correctly and flight crews not detecting the incorrect setting

Safety issue number: AO-2024-007-SI-04 

Safety issue description: Likely due to a training deficiency, Alliance Airlines flight crews' conduct of the Before start procedures and Pre-take-off brief review were not being performed effectively to ensure the speed selector knob was correctly set and checked, which increased the risk of a low-speed event after take-off. 

Multiple occurrences of speed selector knob being reset during a critical phase of flight

Safety issue number: AO-2024-007-SI-03

Safety issue description: Alliance Airlines flight crews were regularly changing the speed selector knob setting during the take‑off run. This was contrary to Embraer's guidance, and Alliance Airline’s own standard operating procedures manual. This increased the risk of distraction during a critical phase of flight. 

Safety action not associated with an identified safety issue

Additional safety action by Alliance Airlines

In addition to the safety action taken to address the identified safety issues, Alliance Airlines: 

• Issued an article to flight crew highlighting that at the completion of the before start duties, the left seat pilot should apply the technique of selecting ‘TOGA, TARA, SPEED’, followed by confirmation of the relevant modes and settings on the primary flight display, prior to calling for the completion of the Before start checklist.
• Advised that a dedicated training module will be incorporated into recurrent simulator cyclic training exercises commencing 1 January 2025. This module will include a review of this occurrence, its root cause, and a reinforced focus on correct procedures and techniques to prevent reoccurrences.
• Advised that during recurrent line check events, there will be an added emphasis on reviewing and ensuring adherence to the correct procedures and techniques.

Glossary

Sources and submissions

Sources of information

The sources of information during the investigation included:

• flight crew of the incident flight
• Alliance Airlines
• recorded data from the quick access recorder
• Embraer
• Honeywell

References

• Alliance Airlines, ‘E190 standard operating procedures manual’, Issue 1.2, August 2023.
• Embraer S.A., ‘170/175/190/195 Standard Operating Procedures Manual’, Revision 29, June 2024.
• Embraer S.A., ‘Embraer E190 Airplane Operations Manual, Volume 1, AOM-1502-047, November 27 2020, Revision 5 – January 31, 2024
• Honeywell, Primus Epic pilot’s guide – integrated avionics system for E190 Load 27, October 2023.
• Regional Aviation Safety Group-Pan America ‘Mode Awareness and Energy State Management Aspects of Flight Deck Automation’, November 2022

Submissions

Under section 26 of the Transport Safety Investigation Act 2003, the ATSB may provide a draft report, on a confidential basis, to any person whom the ATSB considers appropriate. That section allows a person receiving a draft report to make submissions to the ATSB about the draft report. 

A draft of this report was provided to the following directly involved parties:

• flight crew of the incident flight
• Alliance Airlines
• Civil Aviation Safety Authority
• Honeywell
• Embraer
• Brazilian Aeronautical Accidents Investigation and Prevention Center
• United States National Transportation Safety Board

Submissions were received from:

• the first officer of the incident flight
• Alliance Airlines
• Embraer

The submissions were reviewed and, where considered appropriate, the text of the report was amended accordingly.

Purpose of safety investigations The objective of a safety investigation is to enhance transport safety. This is done through:  identifying safety issues and facilitating safety action to address those issues providing information about occurrences and their associated safety factors to facilitate learning within the transport industry. It is not a function of the ATSB to apportion blame or provide a means for determining liability. At the same time, an investigation report must include factual material of sufficient weight to support the analysis and findings. At all times the ATSB endeavours to balance the use of material that could imply adverse comment with the need to properly explain what happened, and why, in a fair and unbiased manner. The ATSB does not investigate for the purpose of taking administrative, regulatory or criminal action. Terminology An explanation of terminology used in ATSB investigation reports is available here. This includes terms such as occurrence, contributing factor, other factor that increased risk, and safety issue. Publishing information Released in accordance with section 25 of the Transport Safety Investigation Act 2003 Published by: Australian Transport Safety Bureau © Commonwealth of Australia 2025 Ownership of intellectual property rights in this publication Unless otherwise noted, copyright (and any other intellectual property rights, if any) in this report publication is owned by the Commonwealth of Australia. Creative Commons licence With the exception of the Commonwealth Coat of Arms, ATSB logo, and photos and graphics in which a third party holds copyright, this report is licensed under a Creative Commons Attribution 4.0 International licence. The CC BY 4.0 licence enables you to distribute, remix, adapt, and build upon our material in any medium or format, so long as attribution is given to the Australian Transport Safety Bureau.  Copyright in material obtained from other agencies, private individuals or organisations, belongs to those agencies, individuals or organisations. Where you wish to use their material, you will need to contact them directly.

At about 0650 local time on 7 February 2024 at Timaru Aerodrome, New Zealand, a Q300 passenger aircraft, registration ZK-NEF, aborted an attempted take-off at speed and stopped at the end of the runway. There were no injuries and there was no reported damage to the plane.

The New Zealand Transport Accident Investigation Commission (TAIC) investigated this occurrence. TAIC requested assistance from the ATSB to download the aircraft’s cockpit voice recorder (CVR) and flight data recorder (FDR) to assist its investigation.

To facilitate this support and to provide the appropriate protections for the information, the ATSB appointed an accredited representative in accordance with paragraph 5.23 of the International Civil Aviation Organization Annex 13 and commenced an investigation under the Australian Transport Safety Investigation Act 2003.

The CVR and FDR were successfully downloaded at the ATSB’s technical facilities in Canberra, Australian Capital Territory, on 14 April 2024. The audio files and flight data were provided to TAIC for use in its investigation.

TAIC published an investigation report on this event on 14 May 2026. The final report is available here.