MENASASI Middle East and North Africa Society of Air Safety Investigators


Lamia Flight

By Anthony T. Brickhouse, Embry-Riddle Aeronautical University, U.S.A.; Diego M. Garcia, Embry-Riddle Aeronautical University, U.S.A.; and Julian E. Echeverri, Aviation Accident Investigation Group, Colombian Civil Aviation Authority, Colombia



An Avro 146-RJ85, performing charter flight LAMIA LMI2933 for the Brazilian Chapecoense football team, was destroyed after impacting a wooded hillside south of José María Córdova International Airport in Colombia. The official accident investigation board included more than 20 experts from five states. At the end of the investigative process, the board determined that fuel exhaustion was the cause of the crash. Of the 77 occupants on board, only six survived. Five occupants suffered serious injuries, and one sustained minor injuries. Crashworthiness and survivability analyses were performed to assess the conditions that allowed these occupants to survive the crash.


It is taught in academic circles around the world that for occupants to survive an accident, specific stipulations must be met. Occupants must have occupiable living volume during the dynamic portion of the crash. Decelerative (G-forces) must be within human tolerances. And occupants must survive all postcrash factors until rescue and medical treatment occur. Air safety investigators use the CREEP (container, restraint, environment, energy absorption, and postcrash factors) methodology to assess the different factors that influence survivability in a crash (see Table 1).

Acronym Factor Explanation
C Container

The available living space for occupants resulting during and after the accident dynamics.

R Restraint

Seats, seat belts, and other restraint systems protecting occupants from being injured by other structures and elements, and  preventing them from being projected inside and outside of their living space.

E Energy

The deceleration forces experienced by occupants during the crash; since occupants are not a fixed part of the airframe, energy can be either attenuated or amplified.

E Environment

All the surrounding factors created by the crash that can injury occupants, like fumes, extreme heat or cold, toxic materials, or fast-moving objects within their living space.

P Postcrash Factors All the different situations and elements that can affect occupant survival after the dynamic portion of the accident. Postcrash fire and smoke, wreckage evacuation, and search-and-rescue systems are the most relevant elements under this domain.
Table 1. CREEP survivability methodology

Actual flight data, crash scene analysis, medical and forensic information, and  personnel interviews from this accident were gathered to determine acceleration loads, magnitude and duration, aircraft structural collapse, and energy absorption. Injury causation, search and rescue, and health-care services for the aircraft occupants were also explored. All five CREEP factors were depicted and weighted for each one of the six survivors in order to evaluate what specific conditions contributed to survivability.


CREEP elements played different roles for each of the surviving occupants of the accident aircraft. Energy absorption and restraints were decisive for all six survivors. The container was a protective factor for three of them, while environmental factors during the crash dynamics were also important as a protective element. In contrast, postcrash factors were detrimental for all of the six survivors.


Occupant survival analyses derived from aviation accidents are crucial for crashworthiness design, but also for education, research, and safety enhancements of current aerospace systems. A comprehensive survival analysis, especially when occurrence circumstances diminish the odds of survival for occupants as in this case, becomes paramount. Research can contribute to enhanced aircraft design and restraint systems, improved emergency services, advanced accident investigation techniques, and in general an augmented awareness and understanding of safety promotion and accident/injury prevention for the general public, operators, and regulators.


The aerospace industry nowadays is labeled as an ultrasafe industry given the safety performance indicators assessed by the International Civil Aviation Organization (ICAO) and the International Air Transport Association (IATA). According to recent research, accident rates were in between 12.2 fatalities per billion passengers in 2017 for the former and 1.35 per billion in 2018 for the latter.

Even though these indicators are near their historical best, high-profile accidents involving commercial aviation operators bring high-impact consequences for the general public in terms of trust and willingness to use airline transportation services. Recent studies described a survival percentage of 81% for all aircraft occupants of commercial passenger aircraft accidents, and in 90% of accidents there was at least one survivor.

According to different aspects and elements unique for each one of the occurrences, the outcomes in terms of survivability of an accident are dichotomously categorized as survivable or nonsurvivable. It is not unusual that, in spite of the theoretically exceeded human tolerance capabilities in certain crash events, there are one or more surviving occupants in the aftermath of an accident. This is precisely the case regarding LAMIA Flight 2933 that crashed near Medellin, Colombia, in November 2016 (see Figure 1).

Figure 1. An Avro 146-RJ85. Source: GRIAA. Final Report. Accident. COL-16-37-GIA Fuel Exhaustion AVRO 146-RJ85, Reg. CP 2933 Nov. 29, 2016, La Unión, Antioquia, Colombia


Figure 1. An Avro 146-RJ85.
Source: GRIAA. Final Report. Accident. COL-16-37-GIA Fuel Exhaustion AVRO 146-RJ85,
Reg. CP 2933 Nov. 29, 2016, La Unión, Antioquia, Colombia

According to the Colombian Aviation Accident Investigation Authority (GRIAA), the accident aircraft, an AVRO 146-RJ85, registered with the tail number CP 2933, was conducting a chartered flight from Viru Viru International Airport in Santa Cruz de la Sierra, Bolivia, to José María Córdoba International Airport (SKRG) in Rionegro, Colombia. The aircraft was carrying the Brazilian football team Chapecoense and some journalist and administrative directives from the team. During a holding pattern waiting to be authorized to intercept the localizer for the approach to Runway 01 of José María Córdova International Airport, the aircraft suffered a sequential flameout of its four engines and impacted the southern slope of a mountain located 10 nautical miles south of the threshold of Runway 01 of SKRG at 02:59 Zulu time at night during rainy weather conditions.

The investigation process identified the following causal factors:

  • Inappropriate planning and execution of the flight by the operator in regard to the amount of fuel required for the safe completion of the intended flight.
  • Sequential flameout of the four engines as a consequence of fuel exhaustion.
  • Inadequate decision-making on the part of the aircraft operator in terms of the implementation of operational safety in its processes.
  • Loss of situational awareness and wrongful decision-making by the flight crew because of the fixation of continuing the intended flight with an extremely limited amount of fuel.

There was no postimpact fire, and the aircraft was destroyed as a result of the crash. Of the 77 occupants, 71 perished and six survived with serious and/or minor injuries. Despite the high-energy impact, the almost complete destruction of the airframe, the rough environmental conditions after the crash, and the limited first responders’ assistance secondary to the geographic conditions of the accident site and the accessibility from there to health-care services, eight survivors were found among the wreckage at the accident site. Unfortunately, one of these survivors was lost at the crash site before the evacuation, and another died from his injuries at the regional hospital shortly after arrival.

After taking into account these different factors, accident investigators can point to one or more of these elements as the potential source(s) of injuries to occupants and the different levels of injury severity generated by their interaction. After an accident investigation, fatalities and both severe and minor injuries are explained in terms of CREEP elements for the specific occurrence and the specific conditions that each accident presented to occupants. This depends on their position in the airframe, the energy amount and dynamics of the crash, the correct and effective use of different types of restraining systems, their opportunity to egress the scene, and the support received after the event, among other factors.

When the investigation process determines that all these elements were against occupant survival, yet one or more occupants survived, other analysis and factors should be considered. This is the case with LAMIA Flight 2933. The objective of the present study is to analyze the different elements of CREEP methodology for the six survivors of the flight.


For the survival factors analysis specific to the crash of LAMIA Flight 2933, the authors of this study referred to the official final report published by the Colombian Civil Aviation Authority (CAA) and GRIAA, where crucial data are depicted and analyzed. Data included weight and balance, speed, distances, accelerations that were mainly retrieved from the flight data recorder (FDR), cockpit voice recorder (CVR), and operational records and forms.

The final report included photos, diagrams, formulas, and calculations for relevant information related to injury causation to occupants. Other publicly available resources such as interviews, expert analysis, documentaries, press releases, and reports were also taken into account.

Following the procedure for CREEP analysis, the authors weighted and modeled all the available information in order to build a general, unified model for the entire airframe and for all occupants. Afterward, a detailed analysis for each one of the surviving occupants was conducted in terms of energy absorption, container preservation, restraint elements, and environmental and postcrash factors. This was done to explain the potential conditions and factors that determined the survivability of six of the 77 occupants of the ill-fated flight.

The first approach for the analysis was energy calculations. This was accomplished using the FDR data, crash scene distances, wreckage distribution, and forensic analysis for specific injuries evidenced in both fatal and nonfatal victims of the crash. This was done to determine the acceleration pulse shape and duration and the onset rate and the magnitude for energy vectors (horizontal and vertical acceleration) using the following equations:

Where the following:

GH = Horizontal G loading
VH1 = Initial impact velocity
VH2 = Secondary impact velocity
g = Acceleration of gravity
SH = Horizontal deceleration distance
Gv = Vertical G loading
SV = Vertical deceleration distance



The second approach for the analysis was the CREEP study for estimating the resulting living space during and after the dynamic portion of the crash, the restraining systems’ characteristics, usability and effectiveness for the surviving occupants, and factoring environmental and postcrash aspects affecting the occupants’ possibilities of receiving timely medical assistance, treatment, and recovery. Finally, a resulting model based on an analog scale comparing the estimated contribution of each one of the CREEP elements was consolidated for each one of the surviving occupants.

Results of kinematics

For energy calculations, estimated weights, distance, and speeds were derived from dispatch records and FDR-related elements. This was done in order to replace available terms of the energy magnitude and duration equations. Based on the data and scene and impact dynamics reconstruction, it was determined that after the initial impact at the top of the hill, a descending energy dissipation trajectory (approximated 55-degree slope) was generated along a magnetic course of 296 degrees, continuing for around 140 meters (462 feet) downhill on the northern slope of the ridge.

This is where the majority of the aircraft wreckage came to rest almost completely destroyed. The only recognizable sections of the airframe were the tail and empennage section (which was preserved and found at the top of the hill slightly behind the initial impact site) and the right wing with a small fuselage section attached directly below it (see Figure 2).

Figure 2. Postimpact path and main wreckage location. Source: GRIAA. Final Report. Accident. COL-16-37-GIA Fuel Exhaustion AVRO 146-RJ85, Reg. CP 2933 Nov. 29, 2016, La Unión, Antioquia, Colombia

Figure 2. Postimpact path and main wreckage location.

Source: GRIAA. Final Report. Accident. COL-16-37-GIA Fuel Exhaustion AVRO 146-RJ85, Reg. CP
2933 Nov. 29, 2016, La Unión, Antioquia, Colombia.

Final distribution and destruction level of the debris also suggested that the main wreckage dissipated the remaining postimpact energy in a snowball-like pattern, with its center in the front portion of the fuselage, which was also the most badly destroyed. This final distribution pattern also explains the final location of most of the deceased passengers, especially those who probably were not using any restraint system at the moment of the initial impact.

The resulting energy calculations for a triangular pulse showed a peak around 70 Gs in the vertical axis at approximately 0.6 seconds, with an initial rapid deaccelerating force until around 0.8 seconds and then a more steady deceleration until around 4 seconds after the initial impact (see Figure 3).

Figure 3. G loads and onset time modeling.

Figure 3. G loads and onset time modeling.

Moving forward into the model, the approximate location of the six surviving occupants in the aircraft at the moment of the initial impact was assessed to determine the container’s integrity, the restraint conditions, and environmental aspects such as potential blunt and penetrating trauma produced by fast-moving and high-energy elements surrounding them during the dynamic portion of the crash. To perform this assessment, the approximate occupant distribution within the aircraft cabin and type of injuries were taken into account (see Figure 4).

Figure 4. Approximate distribution and injuries of surviving occupants

Figure 4. Approximate distribution and injuries of surviving occupants.
Adapted from GRIAA. Final Report. Accident. COL-16-37-GIA Fuel Exhaustion AVRO 146-RJ85, Reg.
CP 2933 Nov. 29, 2016, La Unión, Antioquia, Colombia.

For occupants 1 and 2, the initial impact was estimated to have occurred right below and behind their seats. The energy affectation and dynamics experienced by these two occupants was different from the other four occupants since their seats were rear facing and supported by a structural wall dividing the galley from the rest of the passenger cabin.

The container element for these two occupants was protective at least for the initial sequence of events, after which they were likely ejected from the airframe and were recovered near the empennage section according to rescuer statements and their own narratives.

Occupant 3 was recovered outside of the main wreckage on higher ground compared to the rest of occupants and main wreckage. For this occupant, there was not enough evidence to determine if the container played a protective role, but the probability is low given the injuries received by the close and immediate passengers around him and the condition of the container at the section where he was estimated to be seated.

For occupants 4, 5, and 6, it is highly probable that the container aspect was the most protective element, since evidence from the final wreckage revealed that the upper section of the fuselage, below the attachment to the right wing, was the only fuselage section
that was almost intact after the crash impact.


The restraint element, along with energy absorption, was factored into the type and relatively low severity of the injuries of occupants 1 and 2. As previously mentioned, these two occupants were rear facing, wearing four-point restraint systems. These restraints offered extra protection and prevented further injuries from decelerating forces and during the dynamic part of the accident.

For occupant 3, rescue personnel stated that he was attached by his two-point seat belt to the middle seat of a row of three, where the occupants to his left and right side were found fatally injured and also attached to their respective seats. These findings indicate that the restraint systems might have played a crucial role in the survival of this occupant.

Occupants 4, 5, and 6 were using two-point restraint systems like the rest of the passengers. According to their statements, they were using the seat belts at the time of the initial impact. This most definitely contributed to their final survival. For all the seats on the ill-fated aircraft, the G loads encountered by the airframe and the attaching structures were well beyond the threshold that they can support by design (usually 16 Gs for passenger seats).

Most of the recovered bodies from the crash site were found restrained to their respective seats, but a significant number of bodies were recovered outside the main wreckage, along the path of postimpact energy dissipation. This is a good indicator that those occupants might not have been wearing their respective restraint systems at the moment of the accident.

Energy absorption

The energy absorption and dynamics for the six surviving occupants represent special difficulty for a general modeled assessment. It is highly probable that all six of them received high decelerating G forces at the moment of the first impact (around 70 Gs on the vertical axis). Because of the direction of the higher G peak, the seat design, and according to Eiband curves, this energy load might have been survivable for most of the occupants. Unfortunately, energy dissipation and injury prevention that would have allowed them to survive the dynamics of the crash were decidedly influenced by the other four elements of the CREEP model.

Figure 5. Surface graphs for CREEP element analysis for each surviving occupant.

Figure 5. Surface graphs for CREEP element analysis for each surviving occupant.


Environmental aspects during the dynamic portion of the crash behaved differently for the six surviving occupants. For occupants 1 and 2, due to their facing toward the rear of the aircraft, the container and restraint factors, and ejection outside of the main wreckage, the other elements moving with very high energy did not represent any injury risk for them. This was also true for occupant

A different scenario existed for the other three surviving passengers, since they were seated in the middle section of the fuselage, where most of the occupants received fatal blunt injuries from the heavy and fast-moving elements inside their occupiable space. These elements were seats, luggage, interior cabin structures, and other passenger bodies.

For occupants 4, 5, and 6, this element was most likely attenuated by the fact that during the crash dynamics, the fuselage broke into three main sections: one containing the front part of the fuselage, including the flight deck, which was destroyed and mostly disintegrated; the middle section, which included their location and which was fairly preserved around the right wing, precisely where they were seated; and a third section right behind the seats of passengers 4, 5, and 6, which was also completely destroyed.

Postcrash factors

Postcrash factors were determined to be detrimental for all six surviving occupants. Eight occupants survived the dynamic portion of the crash, but unfortunately two perished either during the evacuation or shortly after arriving at an appropriate medical facility. This was the result of persistent rain and cold temperatures, the high altitude, and the long response times for rescue teams because of the relative remoteness of the crash site. The rescue efforts were complicated by the nonexistent access roads to the site and the
unavailability of air rescue services.

Despite those adverse factors, rescue teams evacuated all survivors on stretchers by foot for at least one kilometer (0.6 miles) to a narrow unpaved road where ambulances and rescue vehicles could pick them up for an approximately hour-long drive to the nearest health-care facility. Also noteworthy was that occupant 4 was rescued from the wreckage with serious injuries approximately four hours after the crash.

Survivability and future safety

Future safety is the aim of any aircraft accident investigation. For survival factors, investigators must ask: If the same accident occurs again, will the outcome for occupant survivability be improved? If a similar accident to LAMIA Flight 2933 occurs again, will more than six passengers survive? Hopefully the answer to each of these questions will be a resounding yes, but this will not happen automatically. Occupant survival analysis derived from aviation accidents is crucial for crashworthiness design, but also for education, research, and safety enhancements to current aerospace systems—not to mention search-and-rescue teams and first responders for events of such magnitude.

The safety performance of the aerospace industry is currently at its best, but when accidents occur, survivability of occupants is still a
topic with significant opportunities for improvement. A comprehensive survival analysis, especially when occurrence circumstances diminish the odds of survival for occupants like those in this particular case, can contribute to the enhancement of aircraft design and restraint systems, the improvement of emergency services, and the advancement of accident investigation techniques. In general, an augmented awareness and understanding of safety promotion and accident/injury prevention for the general public, operators, and regulators can also occur.

CREEP analysis presents a comprehensive inventory of factors to take into account when evaluating the different circumstances that contribute to occupant injury causation and severity in an aircraft accident. Yet more research is needed regarding possible redesign or reevaluation of the CREEP model to assess new and developing factors. Some of the factors that could influence the future of the
current CREEP model include newer aircraft seats and pitches, composite materials, and supersonic air transport.

A redesigned CREEP model should also account for personnel variability and individual conditions such as age, fitness, physical condition, and gender.

(Adapted with permission from the authors’ technical paper LAMIA Flight 2933: Who Lived, Who Died, and Why presented during ISASI 2019, September 3–5, in The Hague, the Netherlands. The theme for ISASI 2019 was “Future Safety: Has the Past Become Irrelevant?” The full presentation can be found on the ISASI website at in the Library tab under Technical Presentations.—Editor)