How LogAloft plans your rest
Leer en español · Lire en français · Ler em português
In this article
A LogAloft rest plan is two models working together. One is a body clock, which says where your internal day sits and how light and melatonin move it after you cross time zones. The other is a model of alertness that weighs each way of using a layover and keeps the one that leaves you most alert on the next flight. A few plain checks sit on top, so no plan asks for something a tired person can't do. For the basics, read why rest is not sleep and what sleep science says about jet lag east and west.
The body clock
The plan's reference point is your body temperature minimum, called Tmin, the low point of your body's night. LogAloft places it 3 hours before your usual wake-up time, the rule of thumb from Eastman and Burgess's jet lag work. If you usually wake at 07:00 at home in Boston, your Tmin is about 04:00 EDT. Tell the app you're a morning type and it moves 30 minutes earlier; an evening type, 15 minutes later.
To follow the clock through a trip, LogAloft runs the revised oscillator model of Jewett, Forger and Kronauer (Journal of Biological Rhythms, 1999), with the parameter set published for it (an intrinsic period of 24.2 hours, μ = 0.13, and light constants G = 19.875, α0 = 0.16, β = 0.013 and p = 0.6) and the critical phase 0.8 hours after the minimum of the model's x variable, as the paper places it. The model is fed the light the plan prescribes, in LogAloft's working values:
| Light | Lux at the eye |
|---|---|
| Asleep | 0 |
| Avoiding light (dark glasses, a dim room) | 10 |
| Indoors | 150 |
| A bright room | 500 |
| Daylight outdoors | 5,000 |
When the model's estimate of your Tmin differs from the rule's by more than 2 hours, the plan keeps the rule's times and warns that the clock may take a day or two longer, because the light you actually get is a guess.
How the clock shifts across time zones
The CDC Yellow Book gives "an average rate of adaptation of 1.5 hrs/day for westward travel and 1 hr/day for eastward travel." LogAloft uses those rates. Under 3 hours of time difference it makes no day-by-day plan. On a stay of 2 days or less it keeps you on home hours, as the CDC advises for short trips. Up to 8 hours east it moves the clock earlier; from 8½ hours east it goes the long way around, later at 1.5 hours a day, as the clock often does on its own. Before you leave, it can shift you an hour a day for up to 3 days, and credits only 0.6 hours a day when you're moving earlier without melatonin.
The light windows follow your Tmin as it shifts:
| To move the clock | Seek bright light | Avoid bright light |
|---|---|---|
| Earlier (after flying east) | Tmin + 1 h to Tmin + 5 h | Tmin − 4 h to Tmin + 1 h |
| Later (after flying west) | Tmin − 4 h to Tmin | Tmin to Tmin + 4 h |
The advance window opens an hour after Tmin. In the 1-hour light pulse study by St Hilaire and colleagues, the response turned from delays to advances about 2 hours after Tmin, so light right at Tmin can push the clock the wrong way. Next to the bright window, the plan asks for 2 hours of ordinary room light rather than darkness.
Say a Boston crew member who sleeps 23:00 to 07:00 EDT flies to Madrid, 6 hours east, for a week. On arrival day the plan avoids bright light from waking until 11:00 CEST and seeks it from 11:00 to 15:00 CEST. Each day both windows move an hour earlier, and by the fifth day the light window opens at 07:00 CEST.
Melatonin, as implemented
Melatonin is always optional, and its times are on Pro only. It appears only when the clock has to move earlier. The dose is 0.5 mg, timed about 10 hours before the middle of your body clock's sleep, which is about 6 hours before its bedtime for an 8-hour sleeper. That is where Burgess and colleagues found the 0.5 mg curve peaks. It is never later than 3 hours before your body's bedtime, nor later than local bedtime. In the Madrid example that is 23:00 CEST on arrival day, 6 hours before the body's bedtime (still 05:00 CEST), then an hour earlier each day. Plans that move the clock later have none.
On a layover, melatonin can also appear as a sleep aid, at 0.3 to 0.5 mg of ordinary melatonin (never extended-release) taken 30 minutes before a bedtime outside your body's night. It is offered only between 7½ hours and 30 minutes before your body's usual bedtime, at least 8 hours before the report, and never before flying back west while your body clock is more than 2 hours behind local time, since an evening dose would pull it the wrong way.
The CDC says "0.5–1 mg is often sufficient" and advises against more than 5 mg, and US supplements aren't regulated, so the dose can differ from the label. For pilots, the FAA's guidance on sleep aids permits only "occasional or limited use" and a wait of "5-times the maximum elimination half life." Oral melatonin's half-life is about 54 minutes (Andersen et al., 2016), so that rule gives under 5 hours. LogAloft's own margin is stricter: 8 hours, or 5 half-lives if that's longer, which for extended-release melatonin would be about 20 hours. Ask your doctor, and pilots your AME, before taking it.
Judging each sleep opportunity
The second model is the Three-Process Model of alertness, as Ingre and colleagues tested it on 136 airline crews in 2014. Sleep pressure (S) falls the longer you're awake and recovers while you sleep, and the body clock (C) rises and falls on a 24-hour cosine. Together they predict sleepiness on the 1 to 9 Karolinska scale, where 7 is sleepy.
In LogAloft's code, awake, S falls toward 2.4 at a rate of 0.0353 per hour. Asleep, it recovers in a straight line up to 12.2 and then curves toward 14.3. C has an amplitude of 2.5 and peaks at 15.28 hours, adjusted for your usual wake time. Sleepiness is 9.56 − 0.43 × (S + C). In the model, sleep comes once you're in bed and S + C is under 8.5, or under 9.5 after half an hour in the dark, and a sleep ends by itself when S + C passes 13. In the hours before your body's usual bedtime it hardly comes at all unless you're exhausted. Your body clock moves toward local time at 1 hour a day after flying east and 1.5 after flying west, hardly at all in the first 12 hours.
For each layover the planner tries thousands of schedules, combining a sleep on arrival (none, 30 minutes, one or two 90-minute cycles, or a main sleep when you land in your body's night), a main sleep starting at any hour and lasting 6 to 9 hours, and a nap before the report. Each is scored on the next flight's average sleepiness, worst half hour and landing, weighted 0.7, 0.6 and 0.4. Smaller costs keep it livable (hours in bed in the local day, hours fighting sleep, an alarm at your body's low point), and any time in bed where you'd lie awake more than 40% of it is thrown out. On augmented flights of 8 hours or more, a third of the cruise counts as bunk rest at 70% efficiency, the figure from Signal and colleagues' in-flight sleep study.
How much sleep the plan expects
How much of your time in bed you sleep depends on where it falls in your body's day. In a controlled study by Wyatt and colleagues (Sleep, 2006), young adults sleeping in their biological day slept 77% of their time in bed, against 83 to 84% with melatonin, while melatonin made no difference in their biological night. LogAloft counts:
| When the time in bed falls | Asleep, without melatonin | With melatonin |
|---|---|---|
| Your body's night | 88% | 88% |
| Your body's day | 77% | 83% |
| The 3 hours before your body's usual bedtime | 60% | 67% |
The last row is LogAloft's own conservative estimate, not a measurement. Wyatt's subjects were healthy 18 to 30-year-olds in a lab, and those hours, which Strogatz and colleagues found to be "a zone of minimal sleep tendency approximately 1-3 h before habitual bedtime", are when sleep comes hardest. The figure for the body's night is LogAloft's working value too.
Sleep before an operating flight comes first
The plan aims for 8 hours asleep before every duty when the layover allows it, or up to 9 if your answers to "How did you sleep?" show you need more. It never puts you in bed for more than 9 hours, or your usual night if that's longer.
Before a flight you operate, that sleep comes first. When the report would cut the last sleep short, the plan opens it early, about an hour after you reach the room and once you've been up about 12 hours, if that gives at least 45 minutes more sleep. Meals fit around it, with something light on arrival and a quick breakfast while you get ready.
Before a deadhead, the target is lower, because you can sleep on board. The plan protects 8 hours less what the deadhead allows, counted as 60% of its block time after the first 90 minutes, and never less than 6 hours.
The safety net
Every plan is checked against a plain one built from the sleep you can expect, and replaced when it fails one of three checks:
- no sleep at all, or a sleep that runs past the time to get up;
- under 6 hours of expected sleep before an operating flight, when an hour more is there to take;
- the first sleep coming more than 22 hours after you got up, when it could come sooner (after 18 hours awake, a sleep within an hour of reaching the room is expected).
Before a night flight, the plain plan includes an afternoon sleep of 90 minutes to 4 hours. On a deadhead, only the first and last checks apply. A quick turn, with less than 2½ hours between reaching the room and getting up, gets no rest plan at all.
The trip to the room counts too, with 15 minutes to deplane, 15 for passport control abroad, 40 for the drive (or the hotel's known drive time), 10 for check-in and 20 to settle. You're up at least an hour before the pickup, or, with no pickup or hotel data, 90 minutes before the report.
Meals
Meals follow the hours you're awake, three a day by default and at least 4 hours apart. Dinner comes about 3 hours before bed and nothing big in the last 90 minutes. Between midnight and 06:00 by your body clock, the plan offers a light snack rather than a meal.
Three studies back this. Morris and colleagues (PNAS, 2015) found glucose after the same meal 17% higher in the biological evening than in the morning. In simulated night work, Chellappa and colleagues (Science Advances, 2021) found that eating at night impaired glucose tolerance and "restricting meals to daytime prevented it." And in simulated shift work, Gupta and colleagues (Nutrients, 2019) found people less sleepy after a night snack of a tenth of the day's food than after a full meal or nothing at all.
So on a flight through the night where you land, or through your body's night, the plan suggests one small snack in that stretch instead of a meal. On the ground it adds a light snack when the first meal would be over 5 hours after block in, or after 7 hours awake without food.
What the plan does and doesn't promise
The plan gives you times and the reason for each. It can't promise sleep. People differ, the models describe averages, and noise or worry are outside them. After each planned sleep the app asks how you slept, and it slowly adjusts your sleep need and your clock's rates from the answers, within set limits.
It is not medical advice and doesn't replace your judgment or your airline's fatigue policy.
How it relates to the rules
The rules guarantee an opportunity to sleep, not sleep. Under 14 CFR 117.25, a pilot's rest before a flight duty period is at least 10 consecutive hours and "must provide the flightcrew member with a minimum of 8 uninterrupted hours of sleep opportunity." In Europe, ORO.FTL.235 sets rest away from base at "at least as long as the preceding duty period or 10 hours whichever is greater," including "an 8-hour sleep opportunity."
LogAloft's target of 8 hours asleep goes beyond both, because an 8-hour opportunity rarely gives 8 hours of sleep. The plan doesn't judge whether a rest period is legal and is not legal advice; when a rest won't give the opportunity the rules require, follow the rules and your company's procedures.
Sources
- Revised limit cycle oscillator model of human circadian pacemaker (Jewett, Forger and Kronauer), Journal of Biological Rhythms 14(6), 1999
- Jet Lag Disorder, CDC Yellow Book 2026, CDC
- Human phase response curve to a 1 h pulse of bright white light (St Hilaire, Gooley, Khalsa, Kronauer, Czeisler and Lockley), Journal of Physiology 590, 2012
- Human phase response curves to three days of daily melatonin: 0.5 mg versus 3.0 mg (Burgess, Revell, Molina and Eastman), Journal of Clinical Endocrinology and Metabolism 95(7), 2010
- Guide for Aviation Medical Examiners: sleep aids, FAA
- Pharmacokinetics of oral and intravenous melatonin in healthy volunteers (Andersen et al.), BMC Pharmacology and Toxicology 17, 2016
- Validating and extending the Three Process Model of alertness in airline operations (Ingre et al.), PLOS ONE 9(10), 2014
- Sleep-facilitating effect of exogenous melatonin in healthy young men and women is circadian-phase dependent (Wyatt, Dijk, Ritz-De Cecco, Ronda and Czeisler), Sleep 29(5), 2006
- Circadian pacemaker interferes with sleep onset at specific times each day: role in insomnia (Strogatz, Kronauer and Czeisler), American Journal of Physiology 253, 1987
- Endogenous circadian system and circadian misalignment impact glucose tolerance via separate mechanisms in humans (Morris et al.), PNAS 112(17), 2015
- Daytime eating prevents internal circadian misalignment and glucose intolerance in night work (Chellappa et al.), Science Advances 7(49), 2021
- Subjective hunger, gastric upset, and sleepiness in response to altered meal timing during simulated shiftwork (Gupta et al.), Nutrients 11(6), 2019
- 14 CFR 117.25, Rest period, Cornell Legal Information Institute
- ORO.FTL.235 Rest periods, Regulation (EU) No 965/2012, UK CAA regulatory library
This is information, not medical advice. For questions about your own health or a pregnancy, talk to your doctor or aviation medical examiner.
A sleep plan for your next trip. Add your flights to LogAloft and open Coming up: every layover before your next duty gets a Rest plan from landing to the report, and trips across three or more time zones get a day-by-day plan.
Open the logbookMore from the blog
Jet lag east and west: what sleep science says for crews and travelers
Why flying east is harder than flying west, when light helps and when it hurts, and how crews on short layovers should sleep after a night flight. With the studies behind each point.
Rest is not sleep: what a crew rest period really gives you
Ten hours of rest can mean six hours of sleep. How the rules define rest, why sleep at the wrong time of your body's day is shorter, what a break in flight is worth, and how to plan a layover so the sleep actually happens. With the rules and studies behind each point.
Red-eyes and your body clock: a practical guide for crews and passengers
A red-eye puts work or travel on top of your body's night. When alertness is lowest, how to nap before a night duty, when caffeine helps and when it costs you sleep, what light does on the way home, the cautions on melatonin, and how to recover afterward. With the rules and studies behind each point.