US military aviation industry 1940 to 1945

[Geoffrey Sinclair]

First note this is the military aircraft production industry, makers of
purely civil aircraft are excluded, so Piper with 959 personnel first
appears in the employment figures in October 1941, Aeronica with
329 personnel and Taylorcraft with 455 personnel in January 1942.
All three supplied a limited production run Liaison aircraft in 1941,
Piper 48, the other two 24 each, but only Piper is counted in 1941.

It is quite probable the averages being quoted are misleading given the
changes in production, of both types and models of the different items,
while factories were renamed, expanded, shut down and repurposed. In
short the various averages probably disguise a great deal, that said, here
they come.

This is based on the report U.S. Department of Commerce - Civil Aeronautics
Administration Office of Aviation Information - Division of Aviation
Statistics U.S. Military Aircraft Acceptances 1940-1945. (Dated October
1946?), which excludes what it considers to be experimental aircraft. So
only January 1940 to December 1945 but including production in Canada
paid for by the US.

US military aircraft production had received a significant stimulus from
foreign orders in 1938 and 1939, plus the decision to include 12 or more
seat civil transports as military aircraft increased the number of official
military producers as of January 1940 to be the bulk of the industry in
terms of employees and airframe weights, but not as much in terms of
airframes given the light aircraft industry.

Some definitions.

Airframe weight. Defined as the empty weight of the airplane less the
weight of the following ten items. 1. Engine. 2. Propeller hubs, blade,
power control and governor. 3. Wheels, brakes, tires and tubes. 4.
Auxiliary power plant. 5. Turrets and power operated gun mounts. 6. Radio
receivers, transmitters and removable units, but not installation parts and
wiring. 7. Starter. 8. Battery. 9. Generator. 10. Turbo Superchargers.

Employment figures are as of end of month, except mid month for some
June to September and all October to December 1945 figures. In 1943
a more detailed definition had employment as Gross basis, thereby
including out of plant trainees, home office personnel not applicable to
any specific plant and special research projects.

The problems with averages can begin to be seen in the changing mixture of
production, the growth of weight, the reduction in trainers, the 1945 draw
down and the need for spare parts.

1940 Bombers, 19.8% of production, average airframe weight 7,709 pounds,
Fighter 28%, 3,256 pounds, Transport 4.8%, 8,569 pounds, Trainer 45.4%,
2,052 pounds, others 2%, 2,869 pounds.
1941 Bombers, 21.2% of production, average airframe weight 9,935 pounds,
Fighter 22.7%, 3,718 pounds, Transport 2.8%, 7,081 pounds, Trainer 48.2%,
1,932 pounds, others 5.1%, 2,254 pounds.
1942 Bombers, 26.4% of production, average airframe weight 12,869 pounds,
Fighter 22.5%, 4,532 pounds, Transport 4.1%, 9,218 pounds, Trainer 36.9%,
2,228 pounds, others 10.1%, 1,474 pounds.
1943 Bombers, 34.2% of production, average airframe weight 14,408 pounds,
Fighter 27.9%, 5,080 pounds, Transport 8.2%, 7,915 pounds, Trainer 23.2%,
2,360 pounds, others 6.5%, 1,300 pounds.
1944 Bombers, 36.3% of production, average airframe weight 17,415 pounds,
Fighter 40.4%, 5,545 pounds, Transport 10.2%, 11,553 pounds, Trainer 7.9%,
2,515 pounds, others 5.2%, 986 pounds.
1945 Bombers, 34.6% of production, average airframe weight 20,085 pounds,
Fighter 45.5%, 5,797 pounds, Transport 9.7%, 16,284 pounds, Trainer 2.7%,
2,626 pounds, others 7.5%, 1,310 pounds.

The yearly average airframe weight was 3,839 pounds in 1940, 4,190 pounds
in 1941, 5,769 pounds in 1942, 7,621 pounds in 1943, 9,992 pounds in 1944
and 11,329 pounds in 1945, or almost exactly a tripling of average weight.

Airframe weights do not include spare parts, in 1941 10.1% of airframe
plants output by weight were spares, 12.3% in 1942, 13.7% in 1943 and
12.6% in 1944. Spare parts production officially ceased in August 1945
even so 9.9% of total 1945 airframe plant output by weight was spares.

Average glider airframe weight was 2,330 pounds in 1942, 3,800 pounds in
1943, 4,100 pounds in 1944 and 4,000 pounds in 1945.

Ignoring the jets, average power per engine was 755 HP in January 1940,
while the 1940 average was 682 HP. By December 1941 average power
was 841 HP, for 1941 as a whole 772.2 HP. In 1942 average engine power
each month stayed in a close band around the yearly average of 1,069 HP.
1943 saw a steady increase from 1,079 to 1,230 HP to push the yearly
average to 1,155 HP, January 1944 average 1,320 HP, December 1,564,
average 1,433 for the year. The 1945 figures peak at 1,840 HP per engine
in October.

Using the employment figures all up 81 airframe factories (counting North
American Dallas as two), 23 glider factories, 26 engine factories (counting
Jacobs Pottstown as two), 9 propeller and 25 modification centres were
involved at one time or another.

Note the Waco Troy and Taylorcraft Alliance factories were considered
airframe producers until October 1943 and August 1944 respectively after
which they were classified as glider producers, so they appear in both the
airframe and glider factory totals. Also Vought-Sikorsky split in 1943 to
become the separate factories of Chance-Vought and Sikorsky, thereby
making 3 different airframe factories in the above count.

Four factories counted in the figures for glider work force did not actually
produce complete gliders, Allied at Dundalk, Bristol at New Haven,
Wichita Engineering at Wichita Falls and York at Kenshoa.

Measuring employment has the problem of when the staff are counted but
also coping with operations supplying more than just the aviation industry.
Apart from the list of obvious aviation production facilities there were the
system for Government Furnished Equipment (GFE) plus the many sub
contractors to the aviation industry.

Employment figures have a tendency to jump as new plants are opened, lots of
people on site before much production has begun. Airframe plants peaked at
around 936,500 personnel in November 1943, engine plants at around 240,000
in February 1944, Propeller plants at around 57,200 in January 1944, Glider
plants at around 16,400 in October 1943, Modification centres at around
43,400
in June 1944, sub contractor peak at an estimated 461,000 in November 1943,
GFE staff at an estimated 258,000 in November 1943. Official peak
employment for the industry was around 2,102,000 in November 1943.

Peak aircraft acceptances were in March 1944, which was also the peak
month in terms of airframe weights, engine shipments peaked in August 1944,
propeller shipments peaked in January 1944. Peak glider production was in
July 1943, peak weight in August. Glider production was cut in mid 1944,
down to around a quarter of the peak (156 versus 678) then expanded back
to around 550 a month in mid 1945 before the end of war shut downs.

In January 1940 the employment estimates are 59,000 airframe, 16,000 engine,
2,500 propeller, 51,000 contractors and 15,000 GFE staff, total 97,000, or
0.097 million. So airframe employment expanded by 15.9 times, engine 21.25
times, propeller 22.8 times, subcontractors 51.6 times, GFE 17.2 times from
January 1940 to their peaks.

Total industry employment had doubled by end October 1940 (9 months),
doubled again by end August 1941 (10 months), doubled again by end May
1942 (9 months), doubled again by end February 1943 (9 months), to around
1,680,000 personnel. Employment passed 1.9 million in July 1943 and stayed
above it until July 1944. By January 1945 employment was down to 1.68
million, down to 1.24 million in July then 0.24 million in December. As
you would expect the period July 1943 to June 1944 when employment was
above 1.9 million is when output levels off before declining.

Efficiency is an interesting concept to try and measure. In January 1940
each airframe worker on average built 0.0043 of an airframe, or 22.6 pounds,
(or the output of 1 airframe required an average of 232.3 workers), peak for
this efficiency measure was in May 1944, each worker built around 0.011
airframes or 121.9 pounds, or each airframe required the output of 94.4
workers. The average airframe worker output stayed around 20 to 25 pounds
until January 1941, increased to 43 pounds by December 1941, 55 pounds by
December 1942, 94 pounds by December 1943 and held in the 110 to 120
pounds band through 1944/45 until the 1945 reductions.

In worker terms the output of engines did not change as much. In January
1940 it took 18.7 workers to produce one engine, the monthly figures vary
from between the best for the entire period of 10.7 workers per engine in
October to 20.1 in February. During the peak output period mid 1943 to
mid 1944 the monthly figures are usually in the band 12 to 14 workers per
engine. So in employee terms there was little efficiency improvement, in
terms of power output the average employee in January 1940 built 40 HP,
by December it was around 60 HP where it stayed for 1941, 1942 saw the
average HP per employee grow from 70 to 80, where it stayed for 1943
before an increase to 110 in August 1944.

Glider production efficiency calculations run into data problems. Of the
1,609 gliders accepted in 1942 no less than 1,417 were built by four
companies (Aeronica, Cessna, Piper and Taylorcraft) whose entire work force
is probably counted under airframes, and while production began in January,
employment data begins in June. Then in 1944 (and probably 1945) four
companies (Ford, G & A, Gibson and Pratt-Read) count the entire workforce,
not just those making gliders, adding about 700 personnel in January 1944,
for a total glider work force of 16,200. Using the 1943/45 figures at its
most efficient glider production required around a fifth the number of
workers
per (lighter) airframe built while in weight terms producing about 40 to 50%
more per worker than the powered aircraft workers. It should be noted of
the
15,821 gliders produced 1942 to 1945, 13,195 were CG-4A, the type was 5,832
out of 6,298 in 1943 and 4,289 out of 4,434 in 1944 and there were far fewer
design modifications. Against this is the way there were no less than 16
companies building CG-4A, 5 of which built under 200.

In January 1940 it took around 3.85 workers to produce 1 propeller and while
the monthly figures do move around this sort of efficiency continued until
mid 1941 when it dropped to around 2.6 workers per propeller. It moved back
up to around 3 in the second half of 1942 and early 1943 before declining to
around 2.5 to 2.6 for July 1943 to June 1944, then up to over 3 in 1945.

Overall as expected airframe production showed the biggest improvements as a
system geared to small production runs was changed to mass production, the
gains in engine and propeller production efficiencies were largely offset by
an increase in size and complexity of the products being built.

One unknown factor is whether subcontractors ended up producing more
complete or more complex etc. sub assemblies, thereby affecting the above
efficiency figures.

The expected lifetimes and maintenance requirements of the piston engines
can be inferred from the fact new US aircraft production 1940 to 1945
required around 492,000 engines, versus the 813,000 engines (Including 1,300
jets) built. Certainly the US was exporting engines to the Commonwealth,
Britain reports 31,100 imports January 1941 to June 1945, but the totals are
small compared with the 320,000 difference between output and domestic
requirements, ignoring exports for the moment the US built 1.6 engines per
one fitted to new production, and maybe 1.5 after removing exports.

Modification centers first appear in the employment figures in June 1942.
Of the 25 centers 16 were run by the airframe companies and 6 by the
airlines. There were 20 centers operating in February 1943 but
Transcontinental and Consolidated-Vultee Fort Worth shut down mid
year, the peak of 23 centres operating was reached in January 1944. In
mid 1944 Bell Marietta, Consolidated-Vultee Tuscon, Curtiss Buffalo and
Douglas Oklahoma City were reclassified as airframe plants, along with
shut downs meant there were 14 centers operating in December, then
down to 12 in September 1945.

Given the purpose of the Modification Centers was to make changes to new
aircraft it seems correct to consider them as part of the airframe builders.
In December 1942 center employment was 2.7% of airframe, that steadily
increased to peak at 5.4% of airframe in June 1944, reducing to around 4.1%
by December. Certainly from mid 1944 onwards the US in general could afford
more interruptions to production lines to make changes versus the usually
significantly higher cost of changes done after production.

The following costs are for the factory, much of which could be recouped
from post war sales.

The Wright Lockland/Cincinnati engine plant coat $141,000,000 to build
62,413 engines, or around $2,260 per engine. The Allison plant received
a $116,000,000 upgrade and produced 69,998 engines 1940 to 1945.
Consolidated-Vultee Fort Worth cost $59,000,000 and produced 3,148
aircraft. North American Kansas City cost $30,400,000 and built 6,608
B-25. Republic Evansville cost $14,500,000 and built 6,242 P-47.

Geoffrey Sinclair
Remove the nb for email.

[Geoffrey Sinclair]

"Geoffrey Sinclair" <gsinc...@froggy.com.au> wrote in message
news:1b-dncf1TbFAjynF...@westnet.com.au...

> Employment figures have a tendency to jump as new plants are opened, lots
> of
> people on site before much production has begun. Airframe plants peaked
> at
> around 936,500 personnel in November 1943, engine plants at around 240,000
> in February 1944, Propeller plants at around 57,200 in January 1944,
> Glider
> plants at around 16,400 in October 1943, Modification centres at around
> 43,400 > in June 1944, sub contractor peak at an estimated 461,000 in
> November 1943, GFE staff at an estimated 258,000 in November 1943.
> Official peak employment for the industry was around 2,102,000 in November
> 1943.

A USAAF report on aircraft production gives a snapshot of industry
employment by gender for January and December 1944.

>From the looks of things the amount of non industry work undertaken
by the different factories and subcontracting used varied significantly
from month to month.

For the airframe industry the workforce in January was 39% female,
ranging from 51% at both Douglas Long Beach and Oklahoma City,
to 5% at Nash-Kelvinator, but since that actually means a whole 12
women out of 228 staff, perhaps Culver at Wichita at 21% is a more
realistic minimum. Incidentally Nash-Kelvinator had a big expansion
in 1944, staff up 11 fold with women making up 29% of the December
workforce.

Culver at Wichita holds the minimum in December as well, 21%,
maximum was Douglas Tulsa at 50%.

The airframe industry was slightly expanding in terms of floor area,
up 8% January to December, despite the workforce dropping by
over 100,000 or 18%

7% of January and 11% of December production was for other
prime contractors or non aeronautical customers. In January the
individual factories ranged from over half at 0% to Ryan San Diego
at 84%. In December only a fifth of factories were pure aircraft
production, with Douglas El Segundo 91% and Nash-Kelvinator
Detroit 94% non aviation production.

Sub contracting made up 29% of January (0% Curtiss Louisville to
68% Martin Omaha) and 38% of December production, (0% Fisher
Body Cleveland to 84% at Curtiss Louisville).

The report has what it calls equivalent time of full plant operation,
which I take to mean the amount out of 168 hours the factory is
in full production, in January this ranges from 47.5 hours at
Consolidated Vultee New Orleans to 111.2 hours at Douglas
Oklahoma City, while in December it ranged from 53.1 at
Sikorsky Stratford to 113.4 at Martin Omaha) Overall it was
89.7 hours in January and 76.6 hours in December.

Average output in terms of pounds of airframe per employee
does not give a time period, implying it is for 1944, ranges from
2 at Sikorsky Stratford to, as expected, 198 at Ford Willow Run,
about a quarter of the factories averaged over 100 pounds per
employee.

Data for only 4 glider factories is given, females 17% in January
and 19% in December, almost no change in total employment
but a 12% decrease in floor space. Production for other prime
contractors or non aeronautical customers was 2% in January
and 8% in December. Sub contracting 48% and 40%. Hours
of production 73.9 and 66.2. Output per employee was 80
pounds, but Ford Iron Mountain was doing 152, next best was 72.

Engine makers, 30% female staff in January (17% Jacobs
Pottstown #1, 43% Chevrolet Tonawanda), and 28% in
December (13% Jacobs Pottstown #1, 48% Pratt and Whitney
Kansas City) Employment dropped 20% over the year but
floor area went up by 3%. Work for other prime contractors or
non aeronautical customers was 2% in January and 1% in
December, Allison Indianapolis was doing 21% such work
in January and 0% in December.

Overall subcontracting was 30% in January (0% Dodge Chicago
and Pratt and Whitney Kansas City to 60% Aircooled Syracuse)
and 28% in December (22% Dodge Chicago and Aircooled
Syracuse, to 56% Ranger Farmingdale) with the note 3 out of
the 18 factories do not report a December figure.

Factory hours were 104.1 in January (72.4 Ranger Farmingdale
to 125.7 Continental Aviation Muskegon) and 83.3 in December
(55.4 Lycoming Williamsport to 129.5 Wright Cincinnati).

Employee output is measured in HP per worker, ranging from
24 at Ranger Farmingdale to 132 at Ford Dearborn.

The 6 propeller factories had 24% female staff in January (8%
Nash-Kelvinator Lansing to 35 35% Hamilton Standard East
Hartford) in December the same two factories were least
and most at 10% and 36% respectively. Over the year staff
numbers went down 6%, floor area went up 15%.

No information on sub contracting, production for others or
non aviation customers or efficiency.

Factory hours in January averaged 108.7 (77.9 Aeroproducts
Dayton to 135.3 Hamilton Standard East Hartford) in December
83.1 (62.3 Frigidaire Dayton to 110.7 Curtiss-Wright Caldwell).

Overall Ford was remarkably productive.

As originally noted the sub contracting and production for other
prime contractors or non aviation customers seems to vary widely
from month to month and individual factories could be anything from
ceasing production to undergoing a major expansion.

How much you can read into female employment is unclear, the
airframe industry managed 40%, the glider industry under 20%,
presumably thanks to more woodwork skills needed versus
assembly line work. The engine makers had around 30% female
staff the propeller makers similar by December.

In 2011 the US workforce was around 47% female.