IL76 / B741, en-route, west of Delhi India, 1996

IL76 / B741, en-route, west of Delhi India, 1996


On 12 November 1996, an Ilyushin IL76TD and an opposite direction Boeing 747-100 collided head on at the same level in controlled airspace resulting in the destruction of both aircraft and the loss of 349 lives. The Investigation concluded that the IL76 had descended one thousand feet below its cleared level after its crew had interpreted ATC advice of opposite direction traffic one thousand feet below as the reason to remain at FL150 as re-clearance to descend to this lower level. Fifteen Safety Recommendations relating to English language proficiency, crew resource management, collision avoidance systems and ATC procedures were made.

Event Details
Event Type
Flight Conditions
Flight Details
Type of Flight
Public Transport (Passenger)
Flight Origin
Take-off Commenced
Flight Airborne
Flight Completed
Phase of Flight
Flight Details
Type of Flight
Public Transport (Passenger)
Intended Destination
Take-off Commenced
Flight Airborne
Flight Completed
Phase of Flight
Aircraft-aircraft collision, Copilot less than 500 hours on Type, Deficient Pilot Knowledge
Language Clarity, Read Back Clearance not followed
Inappropriate crew response - skills deficiency, Procedural non compliance
Accepted ATC Clearance not followed
Accepted ATC Clearance not followed, Level Bust, Mid-Air Collision
Damage or injury
Aircraft damage
Hull loss
Non-aircraft damage
Non-occupant Casualties
Occupant Fatalities
Most or all occupants
Number of Occupant Fatalities
Off Airport Landing
Causal Factor Group(s)
Aircraft Operation
Air Traffic Management
Safety Recommendation(s)
Aircraft Operation
Aircraft Airworthiness
Air Traffic Management
Investigation Type


On 12 November 1996, an Ilyushin IL76TD (UN-76435) being operated by Shymkent Avia, a subsidiary of Kazakhstan Airlines, on a non scheduled international passenger flight from Shymkent to Delhi as KZ1907 collided with a Boeing 747-100 (HZ-AIH) being operated by Saudia Airlines on a scheduled passenger fight from Delhi to Dhahran and level at FL140 in day IMC. Both aircraft were destroyed by impact forces and fire and all 349 occupants of the two aircraft were killed. 


On learning of the accident, the Indian DGCA immediately appointed a senior member of their Flight Standards team as an Inspector of Accidents to lead an Investigation. Shortly after this the Indian Prime Minister and the Haryana State Chief Minister visited the accident site at Charkhi-Dadri and thereafter, a sitting High Court Judge was appointed to lead a formal investigation into the Accident in the form of a Court of Inquiry established under extant legislation. The Court then appointed the Air India Director of Safety and a retired Air Force Director of Operations (ATC) as Court Assessors and the DGCA Controller of Airworthiness as Secretary.

The FDR and CVR were recovered undamaged from the wreckage fields of both aircraft. The data from the IL76 recorders was downloaded under supervision at the Moscow-based Interstate Aviation Committee (IAC) and that from the 747 recorders was downloaded at the UK AAIB.

The flight crew involved were as follows:

  • The IL76 Captain (born 1952) who was PF had a total of 9,229 hours flying experience. This total included 1,488 hours on type out of which 1,287 hours were in command on type. His overall experience was recorded as including 708 hours “on international routes”.
  • The IL76 Co Pilot (born 1959) had a total of 6,822 hours flying experience which included 409 hours on type. His overall experience was recorded as including 207 hours “on international routes”.
  • The IL76 Navigator (born 1945) had a total of 12,789 hours flying experience which was recorded as including 581 hours “on international routes”.
  • The IL76 Flight Engineer (born 1946) had a total of 9,201 hours flying experience which included 1,248 hours on type. His overall experience was recorded as including 1,178 hours “on international routes”.
  • The IL76 Radio Operator (born 1955) had a total of 1,583 hours experience as a radio operator on type and his experience was recorded as including 645 hours “on international routes”.
  • The 45 year-old 747 Captain had a total of 9,837 hours flying experience which included 4,313 hours in command of which 104 hours were in command on type. 
  • The 37 year-old First Officer had a total of 7,779 hours flying experience which included 1,952 hours on type.

What Happened

It was established that the IL76 had first contacted Delhi Radio on HF 35 minutes after takeoff for what was expected to be a flight taking around three hours and reported maintaining FL330 with an ETA for Delhi of 1323Z. About 17 minutes before the collision occurred, the southbound IL76 flight checked in with Delhi ACC West and reported waypoint ‘LUNKA’ on airway G452 at FL330, a position 177 miles west of Delhi. Descent clearance was initially given to FL 250 and subsequently to FL 180 to report passing FL 200. Then, whilst passing FL 240 the flight was transferred to Delhi APP, subsequently checking in there as it passed FL230 when 74 miles from the Delhi VOR. Delhi APP then re-cleared the flight to continue descent to FL 150 and call reaching that level. 

A few minutes later, in response to a query from APP, the fight reported reaching  FL150 on the Delhi VOR radial 270° and was identified on primary radar and instructed to maintain FL 150. The controller added traffic information on a reciprocal direction Saudi Boeing 747 "12 o’clock at 10 miles likely to cross in another 5 miles". Both the instruction and traffic information were acknowledged but the IL76 crew asked for the distance of the traffic again to which APP replied with "traffic is at 8 miles now FL 140". This call was acknowledged with “traffic at 8 miles” and that they were looking for it. This was the last transmission from the aircraft on the Delhi APP frequency. All communications between the flight and ATC were conducted in English by the radio operator.

Meanwhile, the Saudi Airlines 747 aircraft had taken off from runway 28 at Delhi and followed a PARVI-1 SID before establishing on airway G452 northbound. Identified on radar after takeoff by Delhi APP on the same frequency as the IL76, the flight was initially cleared to continue climbing to FL 100 and “to standby for higher”. On subsequently reporting that it was approaching FL100 four minutes after takeoff, the flight was re-cleared to FL140 which it then called approaching two minutes later, requesting a higher level. APP responded by instructing the flight to maintain FL 140 and standby for higher which was read back correctly.

One minute later, the two aircraft, which had briefly entered cloud, collided head on, each crew getting only a very brief sighting of the other aircraft at the same level (based on FDR data analysis) before impact with avoiding action impossible. The APP controller saw that the radar returns from both aircraft had disappeared and soon afterwards, a U.S. Air Force transport aircraft en route from Islamabad to Delhi and descending in VMC from FL 200 reported seeing “a cloud lit up with an orange glow in a 2 o'clock relative position from his aircraft”. The crew of this aircraft added that the glow had persisted and as their descent was continued, they saw “two fireballs diverging away from each other below the cloud” which were then seen to hit the ground. 

An examination of the IL76 CVR data showed that there appeared to have been a degree of uncertainty on the flight deck about the continued descent to - or almost to - FL 140 (this level or close to it was identified as the collision altitude by reference to FDR data). After correctly acknowledging the descent clearance to FL 150, the radio operator “promptly converts it to 4570 metres” (the IL76 was operating with the standard Russian/CIS metric altitude instrumentation) and re-confirmed it with the Navigator (all intra flight crew communication being in the Russian language). It was considered that since all crew were listening out, “it can safely be assumed” that each crew member would have understood that the descent clearance limit was FL 150.

The IL76 radio operator could also have reasonably have been expected to have heard the call from Saudia 763 "approaching FL 100" and being cleared by to climb to FL 140. Also, it was considered that the IL76 radio operator could likewise have reasonably been expected to have heard the “approaching FL140, requesting higher” call from Saudia 763 and the reply instructing the flight to level off at FL 140. 

The Investigation made a series of Findings which it was considered had a direct bearing on the accident which, in summary, were as follows:

  • The entire communication from the IL-76 aircraft to ATC was by the Radio Operator and it was in English.
  • ATC instructions given to both the aircraft were clear and correct and in accordance with the corresponding established procedures.
  • Neither the CVR nor the DFDR of either aircraft contain any indication that any evasive or avoiding action was taken prior to the collision.
  • Throughout the time when Delhi ATC was in contact with the two aircraft, no transmission was made from either of them with reference to any observed abnormality or any real or anticipated emergency.
  • Both aircraft were airworthy prior to the collision and all relevant ATC equipment was serviceable. 
  • The collision was not caused directly or indirectly by sabotage, internal explosion or by any cause external to the flight crews or the aircraft.
  • The collision occurred at FL 140 with the two aircraft on reciprocal tracks at that level. The Saudia Boeing 747 was in compliance with its clearance but the Kazak IL76 was not, since it had been instructed to maintain FL150 in order to ensure safe separation until it had passed the 747.

The Cause of the collision formally recorded as the unauthorised descent of the Kazak aircraft to FL140 after failing to maintain its cleared level FL150.

A number of Contributory Factors were also identified:

  • The unauthorised descent of Kazak aircraft from its assigned FL 150 resulted from: 
    • an inadequate knowledge of the English language on the part of the Kazak Captain which led to wrong interpretation of ATC instructions. 
    • poor airmanship and lack of proper CRM skills on the part of the Kazak Captain compounded by his lack of leadership quality.
    • the casual attitude of the whole Kazak flight crew and their failure to adequately co-ordinate the performance of their respective duties. 
    • the absence of standard call-outs from any members of the Kazak flight crew.
  • Nearly 30 seconds before the collision, both aircraft had entered a cloud layer which resulted in reduced visibility. Although the entry into cloud was accompanied by light to moderate turbulence, this was nowhere near enough to lead to an abrupt loss of 1000 feet of altitude by the Kazak aircraft.
  • The absence of direct pilot-controller communications between the Kazak aircraft and Delhi ATC.
  • The absence of SSR in Delhi airspace. However, installation of current generation radar (both primary and secondary) along with other ATC automated systems was already in progress. 

It was concluded that although the bi-directional ATS route involved was not in itself a contributory factor for the accident, uni-directional routes do enhance ATC traffic handling capacity, which is in the national interest.

It was also found that the outcomes of investigations by the DGCA and the Airports Authority of India (which is responsible for ATS) were not being disseminated to controllers as part of their training. 

A number Findings in relation to Incidental Issues were documented:

  1. Altitude parameter accuracy limits in respect of FDR installed in IL-76 were not in accordance with those laid down in ICAO Annex 6 Part I.
  2. Neither of the aircraft involved was equipped with an ACAS.
  3. The IL-76 was not equipped with either an Altitude Alerting System or an Altitude Acquisition System. 
  4. The organisational set-up of the Indian DGCA did not include any oversight of the ATS provided throughout the country by the Airports Authority of India (AAI). 
  5. Within the AAI, the highest post which an ATC professional can fill was that of Executive Director (Air Traffic Management), an arrangement which is not adequate.  
  6. The prevailing system of civil/military ATC coordination in India suffers from serious shortcomings which adversely affect air safety in the country. 
  7. In India, the ATC profession, which has become highly specialised due to the present day complex flying environment, does not enjoy the recognition and status it deserves. 
  8. Working conditions at the Delhi Airport ACC are not adequate, especially in respect of Area/Approach Control, ATC Simulator provision and IAF capability. In view of the anticipated increase in air traffic, the present number of work stations is also inadequate.
  9. India has no system of licensing air traffic controllers and the proficiency standards being followed in civil and military ATC are not uniform. 
  10. An accident/incident prevention unit in the DGCA staffed by just one individual is inadequate.

A total of 15 Safety Recommendations were made as a result of the Inquiry as follows:

  1. The requirement of proficiency in English, which is the language accepted by ICAO for radio communications on international flights, should be strictly ensured by contracting States. ICAO should devise ways and means to ensure such compliance by contracting States so as to avoid lapses on their part. 
  2. A meaningful Crew Resource Management Programme should be made an integral part of the flight crew training curriculum with special emphasis on the importance of standard call-outs and its efficacy should be evaluated during periodic licence renewal checks.
  3. Before a pilot is appointed as "pilot-in-command", his having acquired effective CRM skill and qualities of leadership should be meticulously ensured.
  4. Air-ground communications with ATC should be governed as follows:
    1. In general, the emphasis should be on direct pilot-controller communications irrespective of crew composition. 
    2. In the terminal control areas, the requirement should be of direct pilot - controller communication invariably so as to avoid time lag in compliance of ATC instructions. 
    3. In the enroute phase, a crew other than pilots may handle radio communications with ATC subject to basic flying instruments being in their view. 
  5. The Airports Authority of India should expedite commissioning of automated ATC systems. 
  6. The Airports Authority of India should bifurcate ATS Route G-452 (which is a high density traffic route) into unidirectional arrival/departure corridors within the limits of Delhi TMA to coincide with the commissioning of ATS automated systems. Other bi-directional routes may also be restructured wherever warranted.
  7. The use of DFDRs/FDRs which do not record parameters to the accuracy limits in ICAO Annex 6 Part I should not be permitted on public transport aircraft by Contracting States. This can be ensured by the regulatory agency of the country of manufacture including the DFDR/FDR in the scope of the Type Certificate and by ICAO taking steps to emphasise the need of implementation of its recommendation by Contracting States.
  8.  Public Transport Aircraft should be equipped with:  
    1. an Airborne Collision Avoidance System (ACAS) 
    2. an Altitude Alerting System 
    3. an Altitude Acquisition System.
  9. The Government of India should create a suitable ATC section at a senior level in the DGCA to properly oversee all aspects of ATC provision.
  10. The Airports Authority of India should have a Member (ATC) on its Board to look after ATC matters. Regional/Field ATC units should be placed under unified command of a discrete ATC group.
  11. The Government of India should integrate civil and military ATCs preferably on the pattern of NATS in the UK.
  12. The Government of India should recognise the importance of the ATC profession and accord special status to it, preferably by examining the feasibility of de-linking ATS from the wider organisational set-up and creating an independent ATC profession to be governed by separate provisions. 
  13. The Airports Authority of India should introduce sectorisation to approach control and re-organise working space in the Delhi airport ATC (new complex) so as to match functional requirements of Area/Approach Control, the ATC Simulator and the Indian Air Force element. The adequacy of the planned number of work stations in the new ATC space should also be reviewed in the light of anticipated increase in air traffic. 
  14. The Government of India should introduce a scheme of licensing for controllers and make it also applicable to the military so as to achieve uniform standards in controlling.
  15. The Government of India should establish an adequately staffed Accident/ Incident Prevention Directorate in the DGCA so as to enhance the level of safety in civil aviation in India.

The Report of the Court of Inquiry was received on 15 July 1997, accepted by the Government as presented and subsequently released for unlimited access.

Related Articles

SKYbrary Partners:

Safety knowledge contributed by: