On a long-haul flight, the flight plan carries a Mach number well above the one that, in the manual’s tables, gives the best specific range — and nobody suggests slowing down. The point at which the aircraft covers the most miles on a kilogram of fuel is already not the point at which it glides best, and the cheapest point for the operator is neither of the two.
One mile per kilogram, not one kilogram per hour
Specific range is a distance-to-fuel ratio: the ratio of true airspeed to total fuel flow, most often expressed in nautical miles per tonne of fuel, the unit varying with the documentation. The definition fits on a single line and already contains the whole subject, because it relates two quantities that do not vary in the same direction when the thrust levers are advanced.
It is a strictly instantaneous quantity: it depends on the current mass, the pressure altitude, the Mach number, the temperature deviation from the standard atmosphere, the centre of gravity position and engine deterioration. The range of a complete flight is not one of these: it is the sum of the instantaneous specific ranges accumulated over the whole mass of fuel burned, and since that mass decreases continuously, the optimum moves during the flight. Comparing two Mach numbers therefore only makes sense at fixed mass, altitude and temperature.
The instinct on the flight deck is to read an hourly fuel flow, since that is what the panel displays. But an hourly fuel flow answers a question of endurance: it says how long the aircraft can fly, never how many miles it can cover.
Why the maximum lift-to-drag ratio is not the right answer
In stabilised cruise, the thrust required equals the drag, that is the mass divided by the lift-to-drag ratio; the fuel flow is, to a first approximation, that thrust multiplied by the specific fuel consumption of the engines. Specific range is therefore written as speed multiplied by lift-to-drag ratio, divided by mass and by specific fuel consumption. It is not the lift-to-drag ratio that must be maximised, it is the product speed × L/D.

At the maximum lift-to-drag ratio the drag is at its minimum, and so therefore is the fuel flow: this is the maximum endurance point of a jet aircraft, the one converged upon when holding. But as soon as it is multiplied by speed, the optimum moves to the right. Accelerating costs drag, and as long as that drag grows more slowly than the speed does, the nautical mile is cheaper.
The graphical reading speaks more clearly than the algebra. On the curve of fuel flow against true airspeed, the bottom of the bucket marks the minimum fuel flow; maximum range, for its part, corresponds to the minimum of the ratio of flow to speed, that is to the point where a straight line drawn from the origin is tangent to the curve. That tangent always touches the curve downstream of the bottom of the bucket. On the parabolic drag polar model taught in performance, the ratio of the two speeds is the fourth root of three, about 1.3: an order of magnitude, valid for an aircraft with no compressibility effect, which wave drag then brings back down.
| Quantity | Turbojet | Turboprop or propeller |
|---|---|---|
| Fuel flow follows… | the thrust required | the power required |
| Maximum endurance | at the maximum lift-to-drag ratio, minimum drag | at minimum power, more slowly than the maximum lift-to-drag ratio |
| Maximum range | at the maximum of the product speed × L/D | at the maximum lift-to-drag ratio |
| What bounds the fast side | wave drag divergence, then the maximum operating Mach number | the power available and the propeller speed, compressibility remaining distant |
| Effect of a headwind | optimum shifted towards a higher Mach number, the shift firmly curbed by the drag rise | optimum speed raised as well, but with no nearby compressibility limit to curb it |
| What an error of cruise regime costs | asymmetric: slowing down is paid for moderately, accelerating is paid for quickly | more even on either side, in the absence of a nearby wave drag rise |
Compressibility bounds the shift
If drag depended only on angle of attack, the product speed × L/D would grow for a long time and the optimum would lie very high in Mach number. Yet the local flow over the upper surface reaches the speed of sound well before the aircraft does: beyond the critical Mach number a shock wave sets in, wave drag appears, then it diverges. A few thousandths of Mach more are then paid for in drag counts.
The product therefore stops growing and falls back. It is this rise that fixes the position of the maximum range Mach number, a little below the divergence Mach number, and that makes the curve asymmetric: soft on the slow side, steep on the fast side. That asymmetry alone explains why every operating regime is placed on the fast side.
Nor is the specific fuel consumption of the engines a fixed coefficient, and its variation with Mach number works against aerodynamic intuition. Slowing down at a given altitude reduces the thrust demanded, but moves the engines away from the regime at which their efficiency is best: part of the hoped-for drag gain is taken back by the powerplant. This is one of the reasons why an optimum computed on the drag polar alone does not coincide with that of the manufacturer’s tables, which incorporate the real behaviour of the installed engines.
The envelope also narrows on both sides — high angle of attack buffet on the slow side, maximum operating Mach number on the fast side — and an aerodynamic optimum only makes sense within the usable domain. On a light aircraft at high level, the buffet margin may rule out the slow side well before range actually degrades there: the constraint that decides is then not an economic one.
The best fuel efficiency isn’t at the slowest speed. Flying faster costs more drag, but if speed increases more than drag does, you get more miles per gallon. However, as you go faster, air resistance suddenly jumps up (wave drag), so the sweet spot for range is just before that happens.
An Airbus A330 flying at Mach 0.82 (optimal range speed) burns 6,000 kg/hour. At Mach 0.85 (long-range cruise), it burns 6,500 kg/hour but covers 30% more distance per hour, meaning better overall fuel efficiency despite higher fuel flow.
The peak is flat: where long-range cruise comes from
In the neighbourhood of a maximum, a smooth function is flat to first order. This is true of specific range as a function of Mach number: around the optimum, the speed can be changed appreciably while losing almost nothing in distance per kilogram.
Long-range cruise is the direct application: manufacturers conventionally accept a loss of one per cent on the maximum specific range, and a speed gain several times larger in proportion is collected in exchange. The regime retained is the one on the fast side, never the slow side: for an equal penalty, the fast side gives time back where the slow side costs it.
The choice has another virtue. On the slow side, at high altitude, a speed reduction moves the aircraft closer to high angle of attack buffet and makes holding the level more demanding in thrust; the fast side, for its part, costs nothing but fuel.
The trap: slowing down towards the maximum range Mach number in the face of a strong headwind often degrades the distance actually covered. That Mach number optimises an air distance; what operations pay for is a ground distance. By slowing down, the time of exposure to the opposing wind is lengthened and the ground speed falls more than the fuel flow does.
Wind shifts the origin of the tangent
What counts then is the ratio of ground speed to fuel flow, that is true airspeed increased by the wind component. This amounts to drawing the tangent no longer from the origin, but from a point shifted horizontally by the value of the wind: it is the second line of the first figure. A headwind shifts the origin to the right, the tangent touches the curve further along, the optimum rises in Mach number; a tailwind does the opposite.
The magnitude of the shift remains modest, of the order of a few thousandths to a few hundredths of Mach even in a strong wind, because the steepness of the drag rise curbs the movement. Hence an instruction that appears paradoxical but is systematic: accelerate in a headwind, slow down in a tailwind. The computer applies it without comment, from the wind field held in its memory; a badly entered field therefore produces a false econ speed, with no warning message.
The cost index: fuel is no longer what is optimised
The maximum specific range minimises a mass of fuel. An operator, for its part, minimises a total cost of which fuel is only one term, the other being proportional to flight time: crew pay, maintenance charged by the hour, immobilisation of the aircraft, missed connections. The cost index is the ratio of these two costs, expressed in the unit proper to each manufacturer: neither a comfort setting nor an environmental indicator, but a company economics parameter.
The two limits are read immediately. A zero index means that time costs nothing: econ speed then coincides with the maximum range Mach number. A maximum index means the opposite and sends the aircraft towards the fastest schedule the envelope allows. Between the two, the computer resolves the compromise for the current mass, altitude and wind, continuously: the econ Mach number therefore drifts en route without anyone having touched anything.
| Regime | Quantity optimised | Position on the curve | Typical use |
|---|---|---|---|
| Maximum endurance | flight time per kilogram | bottom of the fuel flow bucket | holding, managing a situation |
| Maximum range | air distance per kilogram | tangent from the origin | sector limited by fuel |
| Long-range cruise | distance, to within one per cent | fast side of the optimum | simple schedule, without a computer |
| Econ speed | total cost of the sector | according to the cost index, the wind, the mass | routine operations |
| Fast schedule | flight time | bounded by the maximum operating Mach number | recovering delay, high index |
The right reflex: before discussing Mach number, place the aircraft relative to its optimum altitude. The hierarchy of gains is almost always the same, level first and Mach number second: from two levels of deviation onwards, the altitude penalty generally weighs at least as much as moving from a long-range regime to maximum range, and it is corrected by a single request to ATC. Then check that the level is really sustainable at the current temperature, and not merely optimum on paper.
Optimum altitude climbs while mass falls
At fixed Mach number, specific range also passes through a maximum in altitude. Climbing reduces the air density, hence increases the lift coefficient required, hence brings the lift-to-drag ratio closer to its maximum; but climbing reduces the thrust available and ends up imposing an angle of attack incompatible with the Mach number retained. The optimum altitude is the one at which the cruise lift coefficient takes the value that maximises the product speed × L/D at the current mass.
Since mass decreases continuously, this altitude climbs continuously, and does so smoothly. Airspace, for its part, is discretised: only levels are available, assigned by ATC. Hence step cruise, in which the aircraft slowly falls below the optimum, then makes up the gap by climbing when the gain justifies the manoeuvre.
Here again, the flatness of the peak saves the operation: flying a little below the optimum altitude costs of the order of one per cent in distance per kilogram, flying well below it several times more. This is what makes step cruise acceptable, and what makes a low level imposed over a long portion of the route very expensive. A positive temperature deviation moreover degrades the thrust available and lowers the level that can really be sustained: it is that level which decides, not the theoretical optimum.
What to go and check in your own documentation
Three points deserve verification, and they vary from one operator to another. First the set of assumptions of the cruise tables: reference mass, temperature deviation, bleed air and power offtakes — a table read without its header produces discrepancies that do not exist. Then the cost index policy: a single value or one per sector, and who has the authority to change it in flight. Finally the step policy: the minimum gain justifying a climb request, and what is done when the optimum level is not available.
A grey area remains, economic as much as technical. Contingency fuel is established from the trip fuel, itself computed at a planned speed: flying faster consumes part of it before the contingency has even arisen. In-flight monitoring does not always say whether the deviation comes from the wind, from the mass or from a regime faster than planned, and knowing which of the three is at work changes the decision.
For anyone preparing a type rating or a performance examination, the requirement is not to memorise values but to be able to redraw the three constructions: the tangent from the origin, its shift by the wind, the maximum in altitude at fixed Mach number. The FL330 training resources cover this block, tables in hand.
In the event of any discrepancy, the applicable regulation, the aircraft flight manual and the operator’s approved documentation take precedence over this article, which sets out a mechanism only.
Questions fréquentes
The econ Mach displayed does not look like the one on the flight plan: what should be checked?
Three inputs, before suspecting the computer: the mass actually entered, the wind and temperature field en route, and the active cost index. An erroneous mass or wind shifts the optimum without producing the slightest message, since the machine correctly solves a badly posed problem. The comparison must then be made on equal assumptions: a plan computed for a given level and temperature is not comparable with a computation made at the level actually flown. If the three inputs agree, the discrepancy almost always comes from the flight level or from the actual wind profile.
What is the difference between long-range cruise and econ speed?
Long-range cruise is a purely fuel-based convention: a loss of one per cent on the maximum specific range is accepted, on the fast side, in order to gain appreciably more in speed. Econ speed minimises a total cost that includes time, by way of the cost index. The two can cross numerically, but they do not answer the same question and do not depend on the same parameters.
Should one slow down towards maximum range when fuel becomes tight?
That is not the first reflex to have. The most effective lever is the flight level: a deviation of two levels or more from the optimum altitude generally weighs at least as much as moving from a long-range regime to maximum range. Next comes the wind: in a marked headwind, the maximum range Mach number computed in still air is no longer the right one, and slowing down to it degrades ground distance instead of improving it. Finally, a fuel decision is taken on the estimated fuel remaining at destination, not on a cruise regime considered in isolation.
Should one accelerate or slow down in the face of a strong headwind?
One should accelerate slightly, and not slow down. What counts in the end is ground distance per kilogram, hence ground speed divided by fuel flow: an opposing wind amounts to shifting the origin of the tangent to the right, which moves the optimum towards a higher Mach number. The shift remains of the order of a few hundredths of Mach, because the drag rise curbs it rapidly.
What does a zero cost index mean?
It means that flight time is considered free compared with fuel. The computer then seeks only to minimise the mass consumed, and econ speed coincides with the maximum range Mach number corrected for wind. It is the regime found when a sector is limited by the fuel available rather than by the schedule.
Why does the optimum altitude change during the flight?
Because it depends on mass. At fixed Mach number, the optimum altitude is the one at which the cruise lift coefficient takes the value that maximises the product speed × L/D; when mass decreases, that value can only be obtained by climbing. The optimum therefore climbs continuously throughout the flight, whereas airspace offers only discrete levels — hence step cruise.

Leave a Reply
You must be logged in to post a comment.