The multitude of considerations affecting structural design, the complexity of the load distribution through
a redundant structure, and the large number of intricate systems required in an airplane, make weight estimation
a difficult and precarious career. When the detail design drawings are complete, the weight engineer can calculate
the weight of each and every part--thousands of them--and add them all up...and indeed this is eventually done.
But in the advanced design phase, this cannot be done because there are no drawings of details. In the beginning,
the advanced design engineer creates only a 3-view and some approximate specifications. The rest of the design
remains undefined.

One may start the design process with only very simple estimates of the overall empty weight of the aircraft based
purely on statistical results. Some of these correlations are not bad, such as the observation that the ratio of
empty weight to gross weight of most airplanes is about 50%. Of course, this is a very rough estimate and does
not apply at all to aircraft such as the Voyager or other special purpose designs.

One of the interesting aspects of this data is that it does not seem to follow the expected "square-cube"
law. We might expect that the stress in similar structures increases with the linear dimensions if the imposed
load is proportional to the structural weight because the latter grows as the cube of the linear dimension while
the material cross-section carrying the load grows as the square. There are several reasons that the relationship
is not so simple:

1. Some aircraft components are not affected very much by the square-cube law.

2. New and better materials and techniques have helped empty weight.

3. Higher wing loadings are used for larger aircraft.

4. Some portions of airplanes have material size fixed by minimum "handling" thickness.

The figures below show some of this effect. They are from a classic paper by F.A. Cleveland entitled, "Size
Effects in Conventional Aircraft Design" (J. of Aircraft, Nov. 1970).

" As might be expected there is a considerable diversity of scaling among components. This is particularly
apparent between the airframe components where the square-cube law has a strong influence, as on the lifting surfaces,
and those where it has little effect, as on the fuselage. The landing gear, powerplant, and air-conditioning system,
tend to increases gross weight, but the electrical system, electronics, instruments ice-protection and furnishings
are affected more by mission requirements than by aircraft size. On balance, the overall factor of about 2.1 reflects
the tendency of the square/cube law to project a modestly increasing structural weight fraction with size."

The next step in weight estimation involves a component build-up, in much the same fashion as we considered aircraft
drag. This is the approach described here. It involves a combination of structural analysis and statistical comparisons,
with the complexity of the analysis dependent on the available information and computational resources.

If the analysis is too simple or the statistical parameters are not chosen properly, these correlations have dubious
validity. In some cases such correlations can be expected to hold for a very restricted class of aircraft, or to
hold with accuracy sufficient for presentation only on log-log plots. It is very important that the method be based
on the fundamental physics of the design rather than on a ad-hoc correlation parameter. One must also be cautious
of the self-fulfilling nature of such correlations. If one expects, based on historical precedent that a wing should
weigh 20,000 lbs, one may work hard to reduce the weight if the original design weighs 25,000 lbs. When the design
is finally brought down to the initial estimate the project leader may be satisfied, and the new design appears
as a point on the next edition of the plot.

The following sections provide methods for estimating the component weights for advanced design purposes. Some
of the sections (e.g. wing weight estimation) provide a more in-depth discussion of the derivation of the method
and comparisons with several aircraft. The correlations vary from fair to very good, and provide a reasonable basis
for estimating weights. They are based on a variety of sources, from published methods of aircraft manufacturers
to methods developed by NASA and some developed originally here. We do not use Boeing's method or Douglas' method
because these methods constitute some of the most proprietary parts of the preliminary design systems in use at
these companies.

In the following sections, aircraft weights are divided into the following components. Each company divides
the weight into different categories, so it is sometime difficult to compare various components from different
manufacturers. Here we divide the system into the following categories:

Wing

Horizontal Tail

Vertical Tail

Fuselage

Landing Gear

Surface Controls

Propulsion System

APU

Instruments and Navigation

Hydraulics and Pneumatics

Electrical System

Electronics

Furnishings

Air Conditioning and Anti-Ice

Crew

Flight Attendants

Operating Items

Payload

Fuel

Companies typically present a summary of these items in an airplane weight statement. Some examples are available from this link.

The component weights are grouped together to form a number of total weights that are routinely used in aircraft design. This section lists some of the typical weights and their definitions.