18 May 2018 - Jordanian CARC Report issued into 737-400 JY-JAP runway overrun during landing at Tombouctou on 5 May 2017
On the flight from (BKO) Bamako Airport to (TOM) Tombouctou Airport, Mali. After touchdown and before exiting the runway, the aircraft veered to the right of the runway and ended up on the soft area adjacent to the runway and taxi exit.
The investigation showed that while on landing on runway 07 at Tombouctou airport the aircraft touched down just beyond the end of the touchdown zone with manual brakes applied. A tailwind component with an average of 16 knots was recorded when the aircraft was fully configured and aligned with the runway track at 1300 ft AAL and this tailwind effect continued until the aircraft veered right out of the paved area. The aircraft skidded sideways and came to a halt with all landing gears off the runway, shortly before the extended stop way paved surface and left to the taxiway.
The following factors are believed to be the main causal factors of the occurrence:
*** Updated 23 Nov 2020 ***
History of the flight
On the 5th May 2017 the flight, JAV 7843, Operated by Jordan Aviation Boeing B737-400 Registration Mark JY-JAP, was scheduled to depart from Bamako (BKO, GABS) to Tombouctou Airport (TOM, GATB), both airports are located in Mali. The aircraft had last flown two days before the accident, and its crew were adequately rested.
The operating crew reviewed the aircraft technical logs, flight documents and weather forecast for TOM that was indicating normal weather with no significant difficulties. The captain of the flight singed the load sheet of the flight that was prepared by the loadmaster travelling onboard as part of the crew, the load on the aircraft was indicating that the aircraft was loaded so as to operate at all times within its approved Centre of Gravity (CG) envelope.
On the morning of the occurrence the aircraft departed Bamako at 0736 UTC and was flown by the captain on the one-hour flight to Tombouctou Airport (TOM). Before top of descent the crew requested the weather report at TOM aerodrome and the tower reported a surface wind from 250° at 4 kt. Surface visibility was 5 km and local QNH was 1,011 hPa, temperature was 34°C and the runway surface was dry. At 15 NM the crew requested to descent and the tower cleared them to descend to 2500 ft and report final. The crew carried out the checklists and started descending to 2500 ft. at 8 NM from Runway 07 the crew reported final with full configuration. At 7 NM the crew visually captured runway 07 and asked for a clearance to land. The tower cleared them to land on runway 07 and report on ground.
During descent at approximately 5N.M from the airport the first officer stated that he noticed a tail wind component exceeding the operational limitations on the EHSI and FMC displays and he in turns reported that to the captain who acknowledged the information. In his statement to the investigation committee the captain of the flight did not recall this information from the first officer.
The Captain of the flight who was the pilot flying planned to use maximum manual brakes and briefed that maximum reverse thrust would be used during the landing run in order to arrest the aircraft on the short runway of TOM.
The Flight Data Recorder (FDR) analysis indicated that the approach was unstable and that the aircraft crossed the threshold at a speed 15kts higher than the target speed of approximately 141 KIAS and with a tail wind component of 16 knots.
The Aircraft departed BKO with 90 passengers and 11 crew, total on board were 101 persons, 5 tons of cargo, and fuel on board was 9,700 Kg.
The aircraft landed at TOM on runway 07 at 08:35:53 UTC, just beyond the markings of the touchdown zone, approximately 350 m after the runway threshold with a speed of 150 KIAS, flaps 30, landing weight of 55,000 Kg and manually controlled thrust. The speed brakes were set manually to UP position just at touchdown. At touch down the speed reduced to 146 KIAS and engine thrust reversers were deployed manually with the left engine thrust reverser lagging in response with 3 seconds from the right engine resulting in a small drift to the right at 2°. The wheel brakes were applied on the left side only for about 5 seconds in what is believed to be an action from the pilot flying to return the aircraft to the center of the runway. Control column was showing 2.5° pitch up angle which continued for 5 seconds after touchdown. Immediately after the nose wheel touched the ground, the captain applied increased brake pedal displacement and maximum reverse thrust as he considered the rate of deceleration was inadequate. The co-pilot shared the commander’s perception but he did not firmly depress his own brake pedals. The aircraft continued to roll on runway 07 which is 2,170 m long to depart the runway from the taxiway edge located on the right side at the end of runway 25; the aircraft came to rest on the unpaved area southeast to the runway/taxiway intersection. As shown in figure 1.
Figure 1 Aircraft ground run
There were no indications of fire and the captain did not order an evacuation. External steps were brought to the right front door of the aircraft. There were no injuries to the passengers or crew.
The aircraft was removed out from the soft area by towing it from the back using cables on both main landing gears; since the airport was not equipped with a tow bar. The aircraft moved to the parking area using its own power from engine No. 2 since engine No. 1 sustained damage on its fan blades due to ingestion of small gravels from the soft area.
Aircraft recorders (DFDR and CVR) were removed for the purpose of investigation.
No injuries to passengers and crew was reported
1.3 DAMAGE TO AIRCRAFT
The aircraft engine No. 1 sustained damage on all fan blades
1.5 PERSONAL INFORMATION
Pilot in Command: Age 63, 9000 hours on type. Aircraft ratings: B737 / B 747 / B727
First Officer: Age 30, 3300 hours on type. Aircraft ratings: B737
1.7 METROLOGICAL INFORMATION
The landing forecast weather information received by the crew from TOM tower was indicating a surface wind from 250° at 4 kts. Surface visibility was 5 km and local QNH was 1,011 hPa, temperature was 34°C and the runway surface was dry. However; the indicated wind on the aircraft instruments was showing a tailwind component of 16 knots from the time the aircraft was aligned with the landing track with full configuration to the point where the aircraft was in landing roll.
1.10 AERODROME INFORMATION
Tombouctou airport is located south to the city of Tombouctou in GAO region, in the central area of MALI; the aerodrome is used for both military and civilian aircraft. The civilian traffic is limited at this airport. It consists of a single runway and two aprons. The apron used for civilian aircraft is the one adjacent to runway 25 end, the runway length is 2170 m and the width is 30 meters.
Because of its runway narrow width, Tombouctou runway 07/25 was assessed for a previous operation and the safety controls and procedures were distributed to the crew flying to TOM airport in pilot briefing file. The occurrence flight crew did not review the controls developed in the assessment. The crew stated that they were not made aware of the assessment.
1.11.1 DIGITAL FLIGHT DATA RECORDER
The data recorded in the DFDR was successfully retrieved and analyzed by CARC investigator in Saudi AIB facilities.
1.11.2 COCKPIT VOICE RECORDER
No transcript is available so far for the investigation committee to make analysis on the human factors part, communication and other relevant matters that can be extracted from voice data recorded as the data was found erased. However; the investigation committee was able to identify some significant impairments that contributed to the incident through interview of the flight crew and collected statements from other parties.
The Investigation into this Incident collected data from various sources for the purpose of determining the causes and contributing factors. This section of the Report explains the contribution of each investigation aspect to the occurrence and to the severity of the consequences. The analysis also contains safety issues that may not be contributory to the Incident but are significant in adversely affecting safety. Nothing in this section is to be understood as asserting blame or liability.
2.2.1 APPROACH ANALYSIS
The FDR Analysis shows that the aircraft approach was unstable as the vertical speed and tail wind component were showing higher than normal readouts and this continued till the point of touchdown.
At 08:34:34 the autopilot was disengaged at an altitude of 2567 ft / 1636 ft AAL, while autothrottles were disengaged at 08:34:47 at an altitude of 2360 ft / 1432 ft AAL, and the descent rate of the aircraft was recording 1800 – 1900 ft/min.
At 08:35:03 the aircraft was at 1010 ft AAL, with landing gears extended and flaps set at 30 deg, engine thrust was at flight idle and recording 35% N1. The vertical acceleration was still above the normal recommended limit as it was recording 1790 ft/min with tail wind component of 16 kts
At 08:35:19 at a height of 507 ft AAL, the vertical speed was indicating 1540 ft/min with an indicated speed of 165 kts which is higher than the approach speed with 24 kts. The stabilized approach criteria give the pilot an acceptable deviation of 10 kts higher than the approach speed if the airspeed is trending towards the selected approach speed. The indicated tail wind component was showing a recorded value of 16 kts, which is 6 kts higher than the maximum operational limitation for this aircraft.
At 08:35:29 at a height of 250 ft AAL, the vertical speed of the aircraft went down, recording a value of 943 ft/min with consistent tailwind component of 16 kts, the indicated airspeed was 156 kts (15kts higher than selected approach speed) and the control column was pushed backwards resulting in 2.5 deg in pitch attitude.
At 08:35:40 the aircraft was at 104 ft AAL, the vertical speed was 727 ft/min and indicated airspeed maintained at 156 kts, with the wind still coming from the tail side with a magnitude of 16 kts. The engine thrust was increased to 72% N1and the control column was pushed forward to reduce the pitch to 1.4 deg.
Jordan aviation has a policy on unstabilized approaches that is based on internationally accepted standards and on the recommendations issued by Boeing, the manufacturer of the aircraft. The flight safety office periodically informs crews on unstabilized approaches detected via reports and analysis generated as part of the FDM program. This is intended to promote awareness among crews and encourage them to discontinue landings (execute go around maneuvers) in such cases.
In understanding why the flight crew continued the approach and did not consider a go-around, it is important to know that the captain of the flight had experienced some delays in Tombouctou airport in previous flights which according to his statement comes as a result of services non availability in that airport, so; his plan was to continue the approach without affecting the fuel quantity. According to his statement; asking for extra fuel would result in an extra ground time as they have to request that fuel from another air base near to TOM airport and this delay may extend to 2 hours. This would in turn affect the schedule, flight duty time and customer satisfaction.
Flight crews can be subject to a plan continuation bias. Without salient triggers, they will continue with their original plan (that is, to carry out the landing). In this occurrence and according to his understanding there was nothing particularly significant to cause the captain to re-evaluate the original plan of action. Therefore, the flight crew’s recognition that the aircraft was unstable at that point during the approach would have been affected because of plan continuation bias and reliance on the captain skills to land the aircraft uneventfully.
According to his statement, the First officer who was the pilot monitoring on the occurrence flight said that he did not show an assertive action when he realized that the approach was unstabilized because he had previous experience in which the captain of the occurrence flight continued an unstable approaches to landing and he added that in some cases when he was flying the aircraft in some of these flights the captain was taking the controls over whenever the approach was found unstable.
Some previous flights for the same pair of operating crew were examined and analyzed using the FDM system and revealed that the unstabilized approach policy of conducting a go-around when the approach became unstable was not sufficiently ingrained by them, however these unstabilized approaches continued uneventfully.
2.2.2 FLARE, TOUCHDOWN AND LANDING ROLL
At 08:35:45 the aircraft passed runway 07 threshold at 48 ft AAL, the speed was still higher than the recommended Vapp+10, the indicated airspeed was recording 157 kts and tailwind component was 16 kts with a ground speed of 182 kts. The pitch was recording 2.8 deg while the vertical speed was showing a value of 608 ft/min.
At 08:35:47 the aircraft was at 25 ft AAL and the pilot flying started the flare by pushing the control column to produce a 4.9 deg pitch angle, this slowed the rate of descent to 466 ft/min. the engine thrust was decreased by retarding the thrust levers to 35% N1. The flare continued for 7 seconds at which the vertical speed was bled off to 233 ft/min and the aircraft touched the ground at a distance of approximately 350 m from the runway threshold. However; the indicated airspeed was higher than the recommended approach speed at touchdown and with tailwind recording 16 kts.
At 08:35:53 the aircraft touched down the ground with an indicated airspeed of 148 kts and a ground speed of 173 kts this increase in ground speed is due to the tailwind factor that affected the aircraft during its approach and landing phases. The vertical speed at touchdown was bled off to 233 ft/min which is below the limit of hard landing (360 ft/min) and the gravitational vertical acceleration recorded a maximum value of 1.28 g, which indicates that the landing was not hard. The thrust reversers were deployed immediately at touchdown. The FDR data shows that the left engine thrust reverser lagged for 4 seconds from the right engine, resulting in a drift angle of 2 degrees to the right. At 08:35:56 the nose landing gear touched the ground and after 2 seconds at 08:53:58 the thrust reversers were set to maximum thrust with 95 – 97% N1untill the aircraft came to rest on the unpaved area at which they were stowed to close position.
The pilot flying applied a left brake pedal pressure in an attempt to maintain the aircraft in the center line and to compensate this 2 deg of drift. This left brake application lasted to 6 seconds after touchdown.
The brake pressure on the right side was evidenced 5 seconds after touchdown with a value reaching to 3400 PSI.
At 08:36:00 both brake pedals were released momentarily, the brake pedal release lasted to 2 seconds for left and 4 seconds for right brakes. The right brake pressure was not sufficient to provide proper braking performance in which it reached 2000 PSI.
At 08:36:10 both pedal were depressed to the maximum pressure and were showing a brake pressure of 3200 – 3300 PSI.
According to FCTM, the standard procedure to stop the aircraft using wheel brakes comes as following: Use an appropriate autobrake setting or manually apply wheel brakes smoothly with steadily increasing pedal pressure as required for runway condition and runway length available. Maintain deceleration rate with constant or increasing brake pressure as required until stopped or desired taxi speed is reached. The crew did not follow the SOP during brake pedal depression.
The FCTM also provide precautionary notes concerning brake pumping and modulating, as following: • do not attempt to modulate, pump or improve the braking by any other special techniques • do not release the brake pedal pressure until the airplane speed has been reduced to a safe taxi speed If the pilot modulates the brake pedals, the antiskid system is forced to readjust the brake pressure to establish optimum braking. During this readjustment time, braking efficiency is lost.
At 08:36:19 the brake pedals were released again and the ground speed was showing 60 kts, the pilot flying turned the nose wheel steering toward the taxiway entrance which is perpendicular to the runway. The total distance travelled on the runway from the point of touchdown to the point at which the pilot flying used the hand wheel steering to turn the aircraft was 4860 ft (the remaining runway length is ((7118-1148) – (4860) = 1110 ft)).
According to the above performance calculation, it is believed that if the crew kept a continuous maximum brake pressure on the brake pedals, the aircraft would stop on the available remaining length of the runway. Baring in mind that the runway is provided with an extra 60 meters as a stopway at both ends of the runway.
The aircraft started to deviate from runway heading towards the taxiway 25 seconds after the landing gear contact with the runway surface.
Experience and previous analysis of flight data shows that low braking action produce longitudinal acceleration values of up to 0.20 g, this brake performance is equivalent to autobrake 1 setting. Higher values between 0.20 g to 0.30 g indicate a medium braking performance equivalent to autobrake 2 or 3 setting. Values between 0.30 g to 0.35 g indicate a higher braking performance and this is equivalent to autobrake MAX setting. Values between 0.35 g to 0.40 g indicate a high braking force which is similar to maximum manual brake performance. Values higher than 0.4 g indicate a hard braking action that highly energetic and uncomfortable to passengers.
During the first 10 seconds of the landing run, while the aircraft forward speed was higher than normal due to the high rate of descent and tailwind component effect, the braking efficiency varied between low to medium. It then increased during the following seconds of the landing run reaching the highest value at the end of the run just before heading deviation. The high brake energy released from the braking action resulted in tires prints on the runway just prior to aircraft turn towards the taxiway track.
2.3 CRM AND HUMAN FACTORS
2.3.1 THE LEADERSHIP OF THE CAPTAIN AND TEAM WORK
The interview held with the operating crew of the occurrence flight revealed that the captain who was pilot flying made every decision and at no time did he ask the F/O his opinion. He decided to continue the approach in spite it was unstable. He eventually decided to land the aircraft using the manual brakes not according to the SOPs and without an assessment to the landing runway. Finally he decided to vacate the runway at higher than allowable speed to the taxiway in spite of the F/O calls not to do that action.
2.3.2 THE F/O LACK OF ASSERTIVENESS
When the captain decided to continue the approach, The F/O did not exhibit the necessary assertiveness to check with the tower for an update to the weather report after he was aware of the deviation in tailwind values that were exceeding the limitations. The F/O lack of assertiveness was also evident in when he did not tell the captain that the rate of descent was going higher than the known stabilized approach limits. The F/O lack of assertiveness to tell the captain that he did not agree with what he was doing is undoubtedly a contributing factor to the development of the incident.
The investigation committee has concluded that this occurrence could have been avoided if it took place in a pressure free environment, in which the crew had several options available to correct the chain of events that resulted in this occurrence.
Available options were summarized in the following points:
OPT 1) The runway in use at the time of the incident was 07, landing on runway was an option available to the crew when they became aware that the tailwind along the approach track was higher than the operational limitation described in the aircraft manual. It was evident that there were no obstacles around the aerodrome that limit their decision to choose the other end of the runway.
OPT 2) Once lined up on the approach until the aircraft have reached to the height of 500 ft AAL, another option was available to carry out a go-around, in light of the worsening situation. The company procedures clearly state that the pilot monitoring must require a go-around if the approach is not stabilized. The approach was obviously unstable as evidenced by the flight data analysis and collected crew statements.
OPT 3) Another option was also available for the crew to avoid this final result if efficient braking techniques were used according to the standard operating procedures provided by the aircraft manufacturer by maintaining a continuous stand on pressure and by avoiding brake modulation. According to his understanding; the captain of the flight used to use manual braking action to reduce brakes and tires wear and deterioration.
OPT 4) It is also believed that if CRM aspects learnt in classes were exercised in the cockpit during the flight, the decision to go for any of the above mentioned options would be more likely to occur and achievable.
During the course of investigation it was found that:
3.2 CAUSES AND/OR CONTRIBUTING FACTORS
The cause of the occurrence was a high energy unstabilized approach followed by a landing with an excessive speed with higher than operational limit tailwind component, in addition to non-efficient usage of the wheel brakes. A contributing factor to the incident was a combination of deficiencies involving aspects of crew resource management and human factors in which involved the captain leadership, team work and F/O assertiveness.
REC 1) It is recommended that Jordan Aviation’s Training and Flight Operations departments review and enhance the CRM training of the crew to raise their awareness of the importance of CRM skills.
REC 2) It is recommended that Jordan Aviation’s flight operations department increase its monitoring and evaluation of crews abilities in flight through conducting more frequent enroute inspections and incorporating the human based behavior, CRM and Human Factors findings to the inspection report then forwarding this information to training department.
REC 3) It is recommended that Jordan Aviation’s flight operations department stress the applicability of unstabilized approach policy, and in particular, the requirement to go around when the approach does not meet the stability criteria.
REC 4) It is recommended that Jordan Aviation stress on the importance of preserving flight data after occurrences and to set a procedure for consulting the management of the company whenever an occurrence take place to confirm the requirement for this data. This procedure is recommended to include administrative guidance to technical staff who may be requested to follow the data preservation and protection procedures to register their actions in the technical log and to sign against these actions.
REC 5) It is recommended that Jordan Aviation and all Jordanian operators emphasize on stabilized approach criterion and to monitor the available flight data to ensure crew compliance to such procedures.
REC 6) It is recommended that Jordan Aviation and all Jordanian operators conduct safety risk assessments of their operations specially in airports that have frequent unstabilized trends that may be extracted from flight data monitoring systems.