Several studies have identified the use of erroneous takeoff parameters as a safety issue. All pilots are vulnerable because the problem is not specific to any one aircraft type, region, or operator. Using erroneous takeoff parameters can result in early rotation causing a tail strike, collision with obstacles in the departure path, loss of control after takeoff, or a runway overrun.
Several recent incidents and accidents have involved intersection takeoffs, where pilots either take off from the wrong intersection or make errors while calculating takeoff performance parameters.
Wrong Intersection
On February 22, a Marathon Airlines Embraer ERJ-195LR operating for Air Serbia was substantially damaged following a runway overrun at Belgrade Nikola Tesla Airport (LYBE) in Serbia. There were no injuries.
According to the preliminary report, the flight crew planned an intersection takeoff from Runway 30L at Taxiway D6. Takeoff runway available (TORA) from this intersection is 7,706 feet. The crew based the takeoff performance calculations on this distance.
ATC provided taxi instructions to Runway 30L at Taxiway D6. Nearing the runway, the crew informed ATC that “Air Serbia 86C, approaching D6, ready.” The controller responded, “Hold short,” followed by a clearance to “line up and wait via D6.”
The crew lined up on Runway 30L at Taxiway D5 instead of D6 as instructed. According to the transcript, ATC then urgently inquired if the crew realized that they were at intersection D5 instead of D6. ATC also informed the pilots that the TORA from the D5 intersection was only 4,135 feet, suggesting that it was insufficient for a safe takeoff.
At this point, the crew requested a “minute to perform checks.” According to the crew’s statements, they performed the flight parameter calculations on the first officer’s tablet, while ATC offered the option to backtrack to D6 if needed. Thirty seconds later, the pilots confirmed that they were “ready for takeoff” and ATC issued a takeoff clearance for Runway 30L at the D5 intersection.
During the takeoff roll at 100 knots, the crew noticed that there was insufficient runway remaining to take off. Rather than rejecting the takeoff, they felt it was safer to continue and added maximum thrust. As the aircraft departed the paved runway surface, the tower controller noted that there was a lot of dust and the aircraft climbed out very slowly. The pilots reported that the aircraft began to shake as they left the runway and subsequently hit an object.
Investigators found that the aircraft became airborne about 1,650 feet past the runway end and that the aircraft struck several objects, including the approach lights and an ILS antenna. The aircraft initially climbed at a very slow rate and attained a height of 50 feet nearly 6,700 feet beyond the runway end.
Once airborne, the crew received multiple alerts related to issues with onboard aircraft systems, including the flaps and bleed air system. After working through several abnormal checklists, the crew declared “Mayday” and returned to Belgrade for a landing using a higher approach speed due to flap issues.
After landing, emergency crews informed the flight crew of a fuel leak from the left-hand wing. Later, investigators would find several objects embedded in the left wing, including parts of the airport boundary fence.
The preliminary report identified that “one of the likely causes of this accident is the inadequate assessment of parameters for takeoff during preflight preparation by the flight crew of the aircraft, following the decision to take off from a shorter length of the runway in relation to the initially planned one.”
Wrong Data
On Sept. 12, 2021, a KLM Cityhopper Embraer 195E2 was involved in a serious incident and became airborne 1,450 feet before the end of Runway 25R in Berlin, Germany. The incident occurred due to the flight crew performing an intersection takeoff and both pilots inadvertently selecting the wrong intersection on the takeoff performance computer during preflight planning. Investigators determined that the takeoff and initial climb were unaffected, but a high-speed rejected takeoff or an engine failure immediately after takeoff would have caused an accident.
Investigators found that Cityhopper pilots use an iPad as an EFB. Pilots of the Embraer fleet use the manufacturer’s “ePerf” application for takeoff and landing calculations. The flight crew of the incident flight had planned a takeoff from Runway 25R at the L5 intersection.
However, each pilot independently calculated takeoff performance data from Runway 25R at K5, not L5. The K5 intersection is almost the full length of Runway 25R; L5 is 4,330 feet further down the runway.
This error was significant. Takeoff data for Runway 25R/K5 uses Flaps 1 and an assumed temperature (derated or “flex” takeoff) of 57 deg C. Takeoff data for Runway 25R/L5 requires Flaps 3 and an assumed temperature of 35 deg C.
A greater flap selection and higher thrust setting would shorten the takeoff roll. Although not mentioned in the report, the V-speeds would have been significantly different as well.
Investigators identified several shortcomings with the ePerf application. It was observed that pilots’ data entry into the application was done very quickly since it is considered a “routine operation.” Likewise, it was noted that access points for each available runway were listed via a drop-down menu in alphabetical order. As an example, Berlin’s Runway 25R would present the K5 intersection before L5 (with no pictorial display for visual feedback). Investigators felt it was “relatively easy” for both pilots to make the same mistake.
Another issue identified relates to touchscreens, commonly recognized to be vulnerable to selection errors due to two factors—a lack of any system feedback and the “fat-finger” problem. A fat-finger problem relates to a large, relatively crude device pointing at small targets.
Of interest, during the flight crew interviews, the pilots pointed to some automation and performance differences between the Embraer fleet. The E2, as noted, is more automated than earlier variants and is heavier. Investigators felt that frequently switching between type variants may have been a contributing factor considering the incident pilots flew three different E-Jet variants—the E175, E190, and 195E2—during a four-day pairing.
A Global Problem
In 2007, the French BEA and Civil Aviation Authority commissioned a study entitled, “Use of Erroneous Parameters at Takeoff.” The study followed two serious tail-strike incidents at Paris Charles de Gaulle Airport and examined several other accidents and incidents around the world.
These events, as explained, “generally involved new-generation aircraft and had causal factors that included undetected crew errors, of varying degrees of significance, when they entered takeoff parameters.”
The study does list many common errors such as weight and balance entry errors, incorrect V-speeds, incorrect flaps/slat and thrust settings, and a departure from a runway intersection, wrong intersection, or even a wrong runway.
These errors related to takeoff data were identified as “frequent” but are generally detected by the application of SOPs or by personal methods such as mental calculations. The study found that nearly half of the pilots surveyed had experienced errors in parameters or configuration at takeoff; those pilots had a global dispersion from many operators flying different types of aircraft.
The opinions expressed in this column are those of the author and are not necessarily endorsed by AIN Media Group.