Statement of the Chairman, Professor Oetarjo Diran,National Transportation Safety Committee on the release of the Final Report of the investigation of the Crash of the SilkAir Flight MI 185
14 December 2000
Synopsis
On December 19, 1997, about 09:13 UTC, 16:13 Local Time, SilkAir flight MI 185, a Boeing 737-300, registration 9VTRF, crashed in the Musi River.
Flight MI185 was operating as a scheduled international flight from Jakarta Soekarno Hatta International Airport, Jakarta, Indonesia, to Singapore Changi Airport.
The flight departed at about 15:37 Local Time, with two pilots, five cabin crew and 97 passengers on board.
The aircraft left its cruise altitude of 35,000 feet, and impacted into the Musi river about 50 kilometers north-east from Palembang, South Sumatra, Indonesia.
Visual meteorological conditions prevailed for the flight, which operated on an instrument flight rules flight plan.
All 104 persons on board perished, and the aircraft was completely destroyed by impact forces.
Investigation Process
The investigation into the crash of the SilkAir Flight MI185 has been the most extensive aviation accident investigation the National Transportation Safety Committee (NTSC) of Indonesia has undertaken. NTSC was assisted by accredited representatives from the National Transportation Safety Board(NTSB) of the USA and Singapore's Ministry of Communications and Information Technology and advisors from the Australian Transport Safety Bureau (ATSB).
This investigation is a technical investigation in accordance with Annex 13 to the Convention on International Civil Aviation.
Its purpose is accident prevention and not to apportion blame or liability.
After 3 years, the NTSC investigation is finally completed. The final report has today been transmitted to the Singapore and US Governments.
Significant Findings
The following were some of the significant findings in the investigation:
Cockpit Voice Recorder (CVR)
Flight Data Recorder (FDR)
Flight control systems examination
Flight tests and simulations
Cockpit crew backgrounds
Navigational aids and telecommunications
The aircraft was certified, equipped, and dispatched in accordance with regulations and approved procedures.
At the time of the accident, there were light winds and scattered clouds in the area, but there were no meteorological conditions that may have been a factor in the accident.
Weather, air traffic control, aircraft maintenance, fire and acts of terrorism were ruled out as factors contributing to the accident.
The investigation therefore focused on the following issues:
Why and how the CVR and the FDR stopped recording Why there was a time difference between the stoppage of the CVR and FDR Why and how the aircraft left the 35,000 feet cruise altitude(initiation of the descent) How to explain the position of the horizontal trim stabilizerjackscrew as found at impact How human factors aspects could have affected the flight performance of the cockpit crew
Analysis
On impact, the aircraft was completely destroyed, fragmented into small and mangled pieces, which made identification difficult. This yielded limited useful data and made the investigation an almost impossible task. It also made the investigation unique.
The safety issues in this accident relate to the issues of the flight recorders' stoppage, the initiation of the sudden descent from cruise flight level 35,000 feet, the position of the horizontal stabilizer trim actuator as found on impact, and the backgrounds of the cockpit crew.
In the beginning the investigation examined possible systems malfunctions, in-flight fire, in-flight breakup, act of terrorism, weather, air traffic control and human factors, but then we narrowed down our investigation to the main issues, for which some limited evidence was available.
High-speed flutter
The basic evidence, the radar data and the distribution of scattered debris strongly indicated that a partial breakup of the aircraft's tail section occurred at relatively low altitude. On the basis of this information the subsequent investigation ruled out the in-flight break up as a factor leading towards the accident. The separation of the horizontal stabilizer and the rudder skin panels was a high speed flutter induced separation.
Flight simulations
As is the usual procedure in aircraft accident investigations, flight simulations were performed. However, unlike typical accidents, there was no FDR data of the descent that could be programmed into the simulator. The only inputs to the simulation were five radar points. This meant that anysimulation performed is, at best, an estimate of the aircraft's descent path.
The simulation was based on the aircraft descending at speeds exceeding its design envelope. Any simulation of the aircraft's descent as it exceeds the limits of its design envelope may not be accurate.
The team adopted an approach of simulating various combinations of flight control inputs into the simulator. However, due to the lack of FDR data, the team cannot be certain that any of the combinations used were indeedpresent at the initiation of descent.
The team found one combination of control inputs which could match time history of descent from 35,000 feet to 19,500 feet. However, this was only one possible match. Even if this combination was present at the initiation of descent, there was no evidence to indicate whether this combination of inputs were caused by mechanical failure or inputs from the flight crew. There could be other permutations of inputs that could have sent the aircraft through the five radar points.
Structural failure, explosion or decompression
We have found no evidence that a structural failure and decompression initiated the sudden descent from 35,000 feet, and no traces of explosives were found on the wreckage parts and pieces. The aircraft was a relatively new aircraft, and the examination of the wreckage did not reveal any evidence of fatigue, corrosion or any other structural fault that couldhave been a factor in the sudden descent and the break-up.
Control systems failures
We have considered the possibility of control systems failures, which could explain a sudden initiation of the descent. Only a small proportion of the control systems was recovered, of which an even smaller percentage was of a condition suitable for analysis. Although no evidence was found of controlsystems failures, mechanical failure cannot be totally ruled out as only 73% of the highly fragmented wreckage was recovered.
CVR and FDR stoppages
About 7 minutes before the descent from 35,000 feet, the CVR stopped recording. Approximately half a minute before the descent the FDR stopped recording. Analyses could not explain why and how the flight recorders stopped, and at different times. From the available CVR recording, therewere no indications of unusual behaviour of the cockpit crew. From the available FDR recording, there were no unusual aircraft performance parameters.
The team performed tests to determine the cause of the CVR stoppage. The team ruled out short circuit and overload. The team then examined whether there was a break in the wire or a manual pull. From the recording, it was not possible to determine whether either one of these possibilities occurred.
The last ATC transmission, which was performed by the First Officer five minutes after the CVR stoppage, was normal.
Horizontal stabilizer trim
Examination of the wreckage indicated that the horizontal stabilizer trim was at a position different from the last recorded position at 35,000 feet. The position of the trim as found at impact corresponds with an aircraft nose down attitude (orientation). However, further in depth investigation did not yield evidence as to how and why the trim was at the position it was found.
Human performance
The human factors team did a thorough investigation on the possible human factors that could have contributed to the accident. Questions were raised within the first week after the accident happened, about whether cockpit crew actions could have been a factor. Again no technical evidence or proofcould be found that may indicate whether the cockpit crew's action or actions could have initiated or could have been a factor in the occurrence.
Investigation on the financial and work related background of the cockpit crew did not yield any evidence of an impact on the cockpit crew's flight performance.
Both pilots were properly trained, licensed and qualified and had sufficient rest prior to the flight.
There was no evidence found to indicate that the performance of either pilot was adversely affected by any medical or physiological conditions.
Conclusions
After nearly three years and a monumental effort by the NTSC, all possible directions of the investigation into how and why the accident happened had been exhaustively examined.
The investigation into the crash of the SilkAir Flight MI 185 has been the most extensive aviation accident investigation the National Transportation Safety Committee of Indonesia has undertaken.
Due to the highly fragmented wreckage, and the nearly total lack of useful data, information and evidence, the NTSC has to conclude that the technical investigation has yielded no evidence to explain the cause of the accident.
Recommendations
The aim of our investigation, which according to the principles of the Annex 13 to the Convention on International Civil Aviation is not to apportion blame or liability. The aim is to find out what happened, how it happened, and why. The lessons learned from the investigation will be used to try to prevent similar accidents in the future.
The report proposed several recommendations
It is recommended that the ICAO Flight Recorders Panel undertake a comprehensive review and analysis of flight data recorders and cockpit voice recorders systems design philosophy be undertaken by aircraft and equipment manufacturers. The purpose of the review and analysis would be to identify and rectify latent factors associated with stoppage of the recorders in flight, and if needed, to propose improvements to ensure recording until time of occurrence.
It is recommended that, to facilitate the recovery of flight recorders after impact into water, a review of the flight recorders design philosophy be undertaken by the equipment manufacturers to ensure that the underwater locator beacons (ULB) are fitted to the flight recorders in such a manner that the ULB would not be separated from the recorders in an accident.
It is recommended that the ICAO Flight Recorders Panel recommend the aircraft and equipment manufacturers to include recording of actual displays as observed by pilots, in particular for CRT(Cathode Ray Tube) types of display panels a review of the flight recorders design philosophy be undertaken by aircraft and equipment manufacturers to include recording of actual displays as observed by pilots in particular for CRT type of display panels.
It is recommended that a review of the flight crew training syllabi be undertaken by aircraft manufacturers to include recovery from high speed flight upsets beyond the normal flight envelope. The purpose of developing the additional training is to enhance pilot awareness on the possibility of unexpected hazardous flight situations.
It is recommended that a review of aircraft auto-flight systems be undertaken by aircraft and equipment manufacturers to provide all passenger aircraft with auto flight systems that could prevent an aircraft from flying beyond the high speed limit of its flight envelope. It is also recommended that such auto flight systems limit the rate of descent of theaircraft to a certain value that operationally safe.
It is recommended that regional investigation framework for co-operation in aircraft accident investigations be established to enable fast mobilization of resources and coordination of activities to support those states that do not have the resources and facilities to do investigations on their own.
Final remarks
After nearly three years of effort, all possible directions of theinvestigation into how and why the accident happened had been exhaustively examined.
The NTSC has tried to do the utmost to the best of its knowledge to leave no stone unturned.
The NTSC would like to reiterate that the almost total lack of useful data made the investigation an almost impossible task. It also made the investigation unique.
The human factors team did a very thorough investigation on the possible human factors that could have contributed to the accident.
Over 30,000 man-hours were spent in this investigation. A total of 24 staff members of the NTSC participated spending numbers of man-hours during the investigation. A large number of the international investigation leading experts and professionals one time or the other contributed their aviationindustry expertise and experience.
The NTSC acknowledges that the SilkAir MI 185 investigation is technically very complex, and that experts and knowledgeable people can disagree over some or more of the substantive matters, approaches and methods applied..The US has said that the accident can be explained by intentional pilot action. This was a possibility that the investigation team explored extensively. However due to the lack of concrete evidence, this could not be proven conclusively. The US' position, in my opinion, is based oninsufficient evidence. While the US has put forth a plausible hypothesis, the objective of this investigation has to go beyond hypothesising to establish conclusions based on concrete evidence and proof. There is insufficient evidence to prove this theory. I have given detailed clarifications to the US' views in Annex N of my report.
However, all comments, corrections and critique were and are very much appreciated, and if and where possible were incorporated in our analysis and conclusions.
In its very short history of existence, the NTSC has to earn itsreputation, for objectivity, independence, honesty, and diligence.
The NTSC has adhered to those values during our MI 185 investigation.
However, due to the highly fragmented wreckage, and the nearly total lack of useful data, information, findings and evidence, the investigation cannot explain how and why the accident happened.
Nevertheless, it is hoped that this investigation and results will have its contribution to the practices, methods and approaches of investigation, and, however small, contribute to the efforts to prevent similar accidents from happening again.
Finally the NTSC would like to thank all those excellent experts and professionals for their efforts in making this investigation possible.
The NTSC thanks also the governments of Indonesia, Singapore, the United States and Australia for supporting the NTSC in this difficult investigation.
For press queries, please contact the NTSC at:
National Transportation Safety Committee (NTSC)
Gedung Karsa Lt. 2, Departemen Perhubungan
Jl. Medan Merdeka Barat No. 8
Jakarta 10110
INDONESIA
Phone/Fax: +62-21-3517606