|Effective Crew Scheduling Strategies on
John R. Fare
Johnson & Johnson Aviation
Introduction: Current and Future Demands of Our Customers
As many of us know, our customers, e.g. owners, business executives, and charter passengers, are increasingly demanding longer flights and shorter layovers. The needs of our aviation segment are demanding longer range aircraft that fly higher and faster. The days of long layovers and crew changing every flight are over. The newest ultra-long-range aircraft, i.e. Gulfstream, Global, Boeing, and Airbus, are capable of 7,000+ nautical mile range with endurance up to 14 hours. These capabilities are putting conventional city pairs that were previously crew changed at an intermediate tech stop within non-stop reach of one flight crew.
In my experience at J&J, it is not uncommon for our executives to want minimum time on the ground before returning home. We have an increasing amount of flights to Europe that are turning in 12–14 hours with the same crew. Our customer’s time is valuable and we want to be able to meet the demands of their busy schedules while maintaining safety and efficiency.
Alertness in the Aircraft
There are three distinct factors that determine the alertness level of the flight crew. The three factors are:
Our bodies are set on a 24-hour biological clock that we refer to as our circadian rhythm. The physiological purpose of our circadian rhythm is to regulate our body’s functions such as: sleep/wake episodes, core body temperature, digestion, neurological and physical performance, and hormone secretion (Connors, 2005). Synchronization of our circadian clock is done by built-in cues called “zeitgebers” (pronounced “zeitgeebers”) which is German for “time keepers.” Zeitgebers allow for a shortened circadian clock that correlates to the dark light cycle of 24 hours. Without zeitgebers the body’s circadian clock would be 25.3 hours. Zeitgebers that help establish our body’s circadian rhythm are: bright light, temperature, social interactions, work and rest schedules, eating/drinking patterns, and pharmacological manipulation (Caldwell et al., 2009).
A normal sleep/wake pattern is 16 hours awake with 8 hours of sleep. A person’s body is expecting to be at rest is between the hours of 0200–0600 (Flight Safety Foundation, 2005). The circadian low is the time period that our bodies want to be at a restful state, i.e. asleep. The circadian rhythm experiences two time periods of sleepiness on the 24-hour cycle. The first period of sleepiness occurs approximately between the hours of 0200–0500 body adjusted time and the second is 12 hours opposite between the bodies’ adjusted hours of 1500–1700 (Civil Aviation Authority, 2007). Pilots crossing multiple time zones and changing work schedules experience a disruption to the circadian rhythm.
Circadian Adjustment: Phase Advance
Circadian adjustment is more difficult when traveling eastbound due to the shortening of the day, which creates a phase advance of the rhythm. Eastbound flights require a person to shorten their 24 hour cycle to synchronize their rhythm to the new time zone. Flight crews of eastbound flights experience a short sleep episode at the beginning of the layover followed by a long sleep episode towards the end (Gander et al., 1998).
Conversely, westbound flights lengthen a pilot’s day and require a phase delay to extend the circadian rhythm beyond the normal 24 hours for synchronization. Crews of westbound flights experience an initial long sleep episode followed by a nap or shorter sleep episode towards the end of the layover (Gander et al., 1998).
The term “asymmetrical effect” is used to describe the inequality between eastbound and westbound circadian adjustment. Data from several studies was averaged to show that westbound flights crossing eight time zones require 5.1 days to reach 95% of the total re-synchronization of psychomotor performance rhythm, whereas 6.5 days were required for an eastbound flight crossing the same amount of time zones (Billiard & Kent, 2003).
Circadian rhythms re-synchronize towards the new time zone at 92 minutes per day when traveling westbound and 57 minutes per day when traveling eastbound (Billiard & Kent, 2003). New York to London with a 5 hour time difference will require 5 days for total circadian adjustment (57 minutes per day), whereas New York to Los Angles will require 2 days with a 3 hour time difference (92 minutes per day).
Homeostatic Sleep Propensity/Pressure
Homeostasis is the body’s internal mechanism to maintain stability. The normal time of wakefulness for humans is 16 hours +/- one hour (Wyatt et al., 2006). Duty days exceeding the normal time of wakefulness are a factor that contributes to cockpit fatigue. A study (Wyatt et al., 2006, p. R1158) suggests that homeostatic sleep pressure can change with a person’s adjusted schedule. Flight crews that experience disruption to their normal circadian rhythm can have either a shortened or lengthened sleep wake cycle. Adjusted cycles have shown proportional changes in sleep propensity, meaning a sleep wake cycle of 20 hours has an adjusted sleep propensity of 13 hours and a lengthened cycle of 28 hours has the greatest sleep pressure at 20 hours. Keep the shift in sleep propensity in mind when crew rest is less than 24 hours as one’s need for sleep will come at a shorter time interval on the subsequent day. A study (Wyatt et al., 2006, p. 1161) suggests the powerful impact of accumulating sleep pressure even with shortened periods of wakefulness.
Identified complications from fatigue are: slowed reaction time, degraded cognitive functions/decision making, decreased alertness, lack of concentration, mood changes to include irritability and complacency, increase in errors including missed radio calls, checklist items and sloppiness, and a decrease in stick and rudder abilities (Civil Aviation Authority, 2007).
Figure 1. Performance Decrements after Continuous Periods of Wakefulness. From p. 16 of CAA Paper 2005/04, “Aircrew Fatigue: A Review of Research Undertaken on Behalf of the UK Civil Aviation Authority.”
Sleep inertia is the grogginess that one feels after waking up from a deep sleep. Sleep inertia’s effects are most prominent when waking up extremely early for a duty day and when reporting back to duty on the flight deck after crew bunk rest. We’ve all experienced the 0330 wake-up and are still looking at bloodshot eyes in the mirror even after taking a long hot shower. Sleep inertia is more prevalent when a nap is preceded by a prolonged period of wakefulness and or accumulated sleep debt (Gander, Rosekind & Gregory, 1998). Longer in-flight sleep is encouraged for three-pilot crews but amble time (at least 40 minutes) needs to be afforded to offset sleep inertia’s effects prior to resuming flight deck duties (George, 2011). Short naps of 40 minutes or less are effective at combating sleep inertia and increasing alertness on extended duty days and work well when crew bunk rest is not long enough to afford sleep inertia recovery.
Crew Types and Logistics: Two-Pilot Crew
The most basic standard for crewing business aircraft is the two-pilot crew. Consideration should be given to time of day, i.e. normal vs. circadian low time periods. The Flight Safety Foundation recommends that a two-pilot crew be limited to a 14 hour duty period during normal working hours and 12 hours during circadian low operations.
Augmented or Three-Pilot Crew
This scheduling technique adds a third pilot to a trip and is most effective when a crew bunk or separated area with the ability to rest in the supine is available. The Federal Aviation Administration (FAA) has defined the types and “sleep opportunity” of crew bunks aboard aircraft.
Most modern widebody aircraft have the ability to be fitted with Class 1 crew rest facilities. Class 1 facilities are completely separated from the cockpit and passenger cabin areas and are designed with full, flat bunks in rooms with doors that are well insulated, quiet and temperature controlled (George, 2011). Class 2 rest facilities are not completely isolated from passenger compartments but they must have first-class or business class style seats that fold flat to at least 80 degrees from vertical. The rest facilities must have curtains that attenuate light and noise and must be “reasonably free” from disturbances by other crewmembers or passengers (George, 2011). Class 3 rest facilities have chairs that recline 40 degrees with leg and foot supports and cannot be located in economy class.
Corporate jets lack the space of their larger airline counterparts and aircraft such as the Gulfstream 5/550 and Global series can only afford Class 2 crew rest facilities. The FAA has designated Class 1, 2, and 3 facilities as having a sleep opportunity credit of 75%, 56%, and 25% respectively (George, 2011). The sleep opportunity credit is a reasonable estimation of the percentage of sleep a crewmember can expect to get while resting in the applicable facility.
When a trip has a suitable tech stop available and if the duty day is going to exceed either 18 or 20 hours, then a crew change will be the most beneficial to act as a fatigue countermeasure. Consideration should be given to the logistics of “prepositioning” a crew and weather. Winter tech stops in Alaska or Scandinavian countries may be logistically more challenging and should be thoroughly investigated. Again, sleep propensity needs to be considered. Ensure prepositioned crews arrive at least 24 hours prior to departure and up to 48 hours prior if circadian adjustment toward tech stop time zone is advantageous, e.g. Honolulu for Sydney flights.
Fatigue Study: Overview
I decided to start a study to determine the alertness levels of our pilots on all flights that involved either two- or three-pilot crews that had extended operations during circadian low periods. Between 2009 and 2010, I collected a total of 86 forms that represented flights between the U.S. and international destinations in Europe, South America, and Asia. There were 15 pilots and 4 Flight Maintenance Engineers that took part in the study. At J&J, we always fly a three-person crew with a mechanic acting as flight engineer in the cockpit for takeoff and landing and in the cabin as a flight attendant while en route.
Crewing Technique of the Respondents
Crewing Technique Frequency Percentage___________________
2-Pilot Crew 32 37.21%
3-Pilot Crew 54 62.79%
Total 86 100.00%______________________
The hypothesis of my study was to prove that a three-pilot crew was more alert during the last two hours of a flight to include the top of descent, approach, landing, and post-flight than a two-pilot crew.
The alertness scale that I used was developed by Dr. William C. Dement of the Stanford University’s Sleep Well Clinic.
Stanford Sleepiness Scale
Degree of Sleepiness
Feeling active, vital, alert, or wide awake
Functioning at high levels, but not at peak; able to concentrate
Awake, but relaxed; responsive but not fully alert
Somewhat foggy, let down
Foggy; losing interest in remaining awake; slowed down
Sleepy, woozy, fighting sleep; prefer to lie down
No longer fighting sleep, sleep onset soon; having dream-like thoughts
Figure 2. Stanford Sleepiness Scale. Obtained at http://www.stanford.edu/~dement/sss.html
All crews were operating through and/or during their circadian low. It was assumed that a two-pilot crew consists of two pilots and one Flight Maintenance Engineer (FE). A two-pilot crew cannot be scheduled over 12 hours of duty during circadian low so all two-pilot crew data that exceeds 12 hours was from FEs that were part of a three-pilot crew. FEs that were part of a three-pilot crew were unable to obtain supine rest due to space limitations. FE alertness levels between duty hour 12 and 20 are assumed to be equal to a two-pilot crew operating during these extended duty times. A three-pilot crew (augmented) consists of three pilots and one FE. Three-pilot crews were used on flights that exceeded 12 hours of duty. All three pilots were afforded separated supine rest but the FE was not. Flights that exceed 20 hours of duty were scheduled with an entire relief crew at a predetermined tech stop and were counted as two separate two-pilot crews for the study. The last two hours of duty included the top-of-descent, approach, landing, and securing of the aircraft. Pilots and FEs began their duty day two hours prior to takeoff and concluded ½ hour after landing at the final destination.
The data obtained by the research was subjective and it was impossible for me to determine if the input from the volunteers was accurate or biased. Non-reportable human factors such as health, emotional stability, family life, quality of sleep, alcohol/substance abuse cannot be measured to see how they affected the fatigue of my research participants. Ambient conditions such as night, daylight, weather, and turbulence were not taken into account in the “Fatigue Study” to see how they too affected the crew’s sleepiness and rest.
Flight crews received a “Fatigue Study” form whenever a trip would meet the previously discussed parameters. Duty hour 1 began with the show time at the airport and the duty day would terminate 30 minutes after landing. Participants would subjectively enter their perceived alertness level on an hourly basis. Nightly sleep beginning with the night prior to the trip was also recorded.
Figure 3. Johnson & Johnson Aviation Fatigue Study Form
Treatment of Data
The start of duty was changed to reflect body adjusted time according to the previously discussed correction of 57 minutes per day Eastbound and 90 minutes per day Westbound. The mean of the Stanford Sleepiness Scale (SSS) values were separated by the following categories:
Start time of duty day
Lastly, all recorded crew rest and sleep during rest periods were disseminated to show the relationship between duty hour and amount of sleep attained during crew rest.
Here are the cumulative results of the Fatigue Study:
SSS MEAN FOR THE LAST TWO HOURS OF DUTY
Crewing Technique Hour 1 Hour 2____________
2-Pilot Crew 2.23 2.43
3-Pilot Crew 2.20 2.27 ______
“Hour 1” and “Hour 2” represent the second to last and last hour of duty of all submitted forms for respective crewing technique.
Figure 4. SSS Mean During Last 2 Hours of Duty
The research data supported the hypothesis that a three-pilot crew is less tired than a two-pilot crew during the last two hours of duty during extended circadian low operations. SSS data was consistent and supports the three-pilot crewing technique for extended operations that conventionally required complete crew swaps at tech stops. The data obtained during the research process should give managers, schedulers, and pilots increased confidence in using three-pilot crews to mitigate the risks posed by in-flight fatigue.
2-PILOT VERSUS 3-PILOT SSS PER DUTY HOUR
Figure 5. Crewing Technique vs. Hourly Mean SSS
Two-pilot and three-pilot crews exhibit similar fatigue levels until duty hour 11(flight hour 9) whereas three-pilot crew SSS levels remain significantly lower than two-pilot levels. This result led my initiative to lower our two-pilot circadian low duty day to 9 hours of flight time.
SLEEP PROPENSITY’S EFFECT ON SSS
SSS averages per duty hour were separated by start time of duty. All start times were adjusted by the previously discussed factor if away from the home base of Trenton, NJ. The results were:
SSS vs. Adjusted Start Time of Duty Day
Time Frequency SSS ( mean)__
0100 19 2.06
0400 14 1.83
0600 55 1.51
0700 84 2.28
0800 71 1.48
0900 31 1.72
1000 88 1.78
1100 84 1.83
1200 209 1.85
1400 118 1.65
1500 25 1.50
1600 85 1.78
1700 148 1.79
1800 21 2.58
1900 56 1.96
2000 23 2.62
Total: 1131 1.92________
Figure 6. SSS Mean for Entire Flight vs. Start Time of Duty Day
Sleep Propensity does have an effect on the success of the three-pilot crews. There is a significant increase in SSS of start times between 1800 and 0700. The data shows a definite limit of 1700 when cumulative SSS levels begin to increase.
DUTY HOUR’S EFFECT ON SLEEP DURING CREW REST
All rest periods were analyzed on the submitted forms and percentage of sleep was broken out to demonstrate the ratio of sleep compared to the duty hour when a crew rest period was taken. The results:
Crew Rest Sleep Percentages vs. Duty Hour
Duty Hour Frequency Sleeping Awake % Sleep % Awake
1 1 0 1 0 100
2 1 1 0 0 100
3 7 2 5 29 71
4 13 6 7 46 54
5 14 4 10 29 71
6 16 5 11 31 69
7 14 9 5 64 36
8 14 5 9 36 64
9 13 9 4 69 31
10 12 10 2 83 17
11 9 5 4 56 44
12 9 5 4 56 44
13 11 6 5 55 45
14 13 7 6 54 46
15 10 6 4 60 40
16 9 6 3 67 33
17 4 4 0 100 0
18 3 3 0 100 0
19 1 0 1 0 100
20 0 0 0 0 0
Analyzed crew rest showed an increase in sleep in rest periods taken after the 9th duty hour. Crew rest taken earlier in flight had less sleep because there was no physiological need due to a lack of sleep propensity. Crews coming off of rest with little or no sleep did not see as much SSS reduction as crews with sleep. Crew rest obtained between Duty Hour 10 and 17 was most beneficial with the highest percentages of sleep which corresponded to lower SSS levels compared to two-pilot crews with data from FEs that had no provision for sleep. Lower SSS levels beginning at Duty Hour 11 for three-pilot crews correspond to increased percentages of sleep obtained while on crew rest.
Three-pilot crews are less tired than two-pilot crews on extended circadian low flights
Sleep Propensity must be considered when augmenting
Consider crew change if flight takes off prior to 0700 or after 1700
Make a strategy for crew bunk rest
Rostering prior to the flight
How much time is available? Long nap or short nap? When will there be the greatest physiological need
Give priority to the Pilot Flying
Employ In-flight Fatigue Countermeasures
Example 1: 2 Pilots
Depart KTEB @ 1800 Local
Arrive LFPB @ 0630 Local
12 hour rest period + 2 hours for travel and “unwinding”
Depart LFPB @ 2030 Local
Arrive KTEB @ 2330 Local
Example 2: 3 Pilots
Depart KTEB @ 0800 Local
Arrive RJTT @ 1230 Local the next day
Figure 7. Chart obtained at National Highway Traffic Safety Administration website at: www.nhtsa.gov/people/injury/drowsy_driving1/human/drows_driving/wbroch/wbroch_lg/wbroch_lg.html
Billiard, M, & Kent, A. (2003). Sleep: physiology, investigations, and medicine. New York,
NY: Kluwer Academic/Plenum
Caldwell, John A., & Caldwell, J. Lynn (2003). Fatigue in Aviation: A Guide to
Staying Awake at the Stick. Burlington, VT: Ashgate Publishing Limited
CEriksen, C.A., Torbjorn, E., & Nilsson, J.P. (2006). Fatigue in trans-atlantic airline operations: Diaries and actigraphy for two- vs. three-pilot crews. Aviation, Space, and
Environmental Medicine, 77(6), 605-612.
Gander, P.H., Gregory, B.S., Miller, D.L., Graebner, R.C., Connell, L.J., & Rosekind, R.
(1998). Flight crew fatigue V: Long-haul air transport operations. Aviation, Space,
and Environmental Medicine, 69(9), B37-B48
Gander, P.H., Rosekind, M.R., & Gregory, K.B. (1998). Flight crew fatigue VI: A synthesis. Aviation, Space, and Environmental Medicine, 69(9), B49-B60.
George, F. (2011, February). Fatigue risk management. Business & Commercial Aviation, 32-37.
Miller, J. C. (2005, May). Operational Risk Management of Fatigue Effects (AFRL-HE- BR-TR-2005-0073). : United State Air Force Research Lab.
Neri, D., Oyung, R., Colletti, L., Mallis, M., Tam, D., & Dinges, D. (2002), Controlled Breaks as a Fatigue Countermeasure on the Flight Deck. Aviation, Space, and Environmental Medicine, 73(7)
United Kingdom Civil Aviation Authority (CAA), Safety Regulation Group. (2007). Aircrew fatigue: A review of research undertaken on behalf of the UK Civil Aviation Authority (CAA PAPER 2005/04). Retrieved from http://www.caa.co.uk
Business Aviation Safety Seminar ● FSF and NBAA ● Montreal, Quebec, Canada ● April 2013