Yearly Production Serial Numbers

Thanks again Sean!

I think this was the same PDF that you had sent to me last year. The Ohio DMV accepted it for changing the year on my title back then.

Brachus

I believe it noteworthy to include this too.  A letter sent to all W/O dealers regarding serial numbers and model year distinction of 48's and 49's.

Edited by Scott R - 08 June 2009 at 8:44am

Based on the service bulletin information posted here When was my CJ-2A made? and the Willys Production Serial Numbers that you posted in this thread, I'd say it shows pretty clearly that Harold's data on the History page should be noted with a caveat that it likely describes model year information and not calendar year production information.  Also, I think that the Willys Production Serial Numbers should be added to Harold's data to make clearer the apparent distinction between model year and calendar year of production.

Are there any opposing or alternative views in how we should interpret these somewhat disparate yearly production data sources?

sean wrote: . . . . The "history" figures were culled from yearly figures such as posted in the manuals.  I believe these are "calendar" year production numbers, not "Model" year sales numbers. The usual "yearly" numbers correlate well to numbers w/dates in the service bulletins, with the exception of 48/49 transition. . . .

Sean, I think you are overlooking the information that can be derived from your own service bulletins.

I made a few computations of the average number of vehicles produced each year based on the first and last dates for changes having a serial number listed in your list of changes by serial number and production date.  I came up with 195 per day in 1946 and 1947 and 204 per day in 1948.  

( ----------- As an aside, I think the production figures for 1947 were higher than I computed because I think the serial numbers or dates for the the first two entries for 1947 don't fit the other information very well.
     SN          Date       Bulletin
97484      2/14/1947    829
97740        3/3/1947    829

This data suggests that only 256 vehicles were produced from mid February to the 3rd of March when we know that is a bit over 1 day's typical production.   Perhaps one of the numbers were copied incorrectly or entered that way originally. ----------)

Even using lower than overall average daily production figures, look at the conclusions one can draw from your bulletins.

The last serial number/ production date combination listed for 1947 is #146344 on 10/22/1947.  If you extrapolate production for the remaining 70 days in 1947 at 195 per day, you go well beyond 148458, supposedly the SN of the last 2A produced that year.  That does not pass the giggle test.  Using a number of 210 vehicles a day (the estimated rate of production between September and October of 1947 and consistent with the early 1948 production rate BEFORE production tapered off at the end of 1948, you get a last serial number of 161044 that makes much more sense.   The same thing happens if you examine the last serial number/date combination in 1948.  This would be from bulletin 48-60 showing # 215648 was produced on 9/20/1948.  At ~204 per day, you could easily produce #224764 before the end of the calendar year.   

Here is the production data from the first edition of the Willys Master Parts List, 1950:


Year          Start SN    End SN     Quantity
1945         10001          11824        1824
1946         11825          83380      71556
1947         83380        161651      78272
1948       161652        224764      63113
1949           -             -          -

It does have an obvious error in the ending SN for 1947 that should be one less based on all the other information we have.  Also, I added the computed "Quantity" column (also off by one for 1947).  I find this data much more consistent with the service bulletin information as well as with what we know about the average production rate over the life of the 2A.  I have seen you state the belief that there were no 2As produced in 1949.  This data supports that notion.

Sean, when you go back to your service bulletins and look at the kinds of information I have referenced, doesn't it make you think that the History page data and the later Master Parts List data are citing model year data?  I find the data for 1948 quite compelling in that respect.

The other fact that I think bears on 1948 production is the Sales Bulletins we have dated October 30, 1948, identifying #219589 as the first 1949 model year vehicle.  I have to believe that 219589 was already produced on or about 10/30/1948.  It's odd to think that production would stop after 20 more days of production when there were 61 more calendar days in the year.

Two things that I can think of might impact my thoughts.  One is the possibility of work stoppages in late 1947 and late 1948.  I thought those and the parts shortages occurred earlier.  the other thing is that there was a second production facility in California that might have allowed easier overlapped production of 2As and 3As in 1949.  Other than that, I'm inclined to believe that there was no overlap and that 2A production stopped in 1948.

One other tidbit is that in 2006 I copied a page from http://www.off-road.com/jeep/tech/history/serial.html which is no longer there, but I saved the URL and have the whole thing in a Word document.  It was an article by Terry L. Howe with the first paragraph saying,  You won't find more accurate Willys-Overland production information anywhere else. This production information was transcribed into electronic form by Charlie Weaver from a hand written a compilation Norton Young did between 1959 and 1961. Norton Young was an employee of Willys/Kaiser at the time and did the compilations so that executives at Kaiser would have yearly production information.  The article goes on to cite identical production data as to what is on the History page here including the incorrect annual production quantities.  I believe that we have seen Norton Young cited as the source for the production data and suspect that Harold picked up the same table with the incorrect production quantities.

I can post a copy of the page from the 1950 version of the Master Parts List, but for now, I just extracted the table information and ask you what you think of it in light of what you have in the service bulletins.

Thoughts?

Sean,

Thanks.  If you come across the compelling post or thread, you should place a link to it in this thread for future reference.

That would be good.

  And when do you believe yours were produced - October? 

Here are a few quick random thoughts on this topic. Unfortunately I don't have enough time right now to do justice to this topic. Much of this information is included in my new CJ-3A book.

Model year and calendar year are often misunderstood.

There are mistakes and omissions in Willys-Overland published information. This can be easily seen by comparing serial numbers and many other items.

Bulletin publication dates are only a clue to vehicle change dates unless an actual change date is included. For example, Bulletin 49-7 dated March 14, 1949 references engine J264163. Were CJ-2As were still being produced in March 1949? Maybe, maybe not, but it is a clue. Of course "J" prefix engines were used in other applications besides a CJ-2A vehicle.

While monthly, weekly and daily production averages can be calculate from the yearly total they are not accurate. See the newspaper article Sam referenced for just one reason.
http://www.dcent.net/cj2a/images/info/coshocton_12_20_1945.jpg

Check out this scan below of an early 1950 Master Parts List. I believe they made a mistake and inadvertently used calendar year data for the 1948 to 1949 serial number transition. This mistake was corrected to model year data in later editions of the same manual up through the version posted by Sean http://www.cj-2a.com/documentation/Willys/production/Master_serial_numbers.pdf Compare the two versions for some interesting data.



I think the production of 1949 model year CJ-2As gradually tapered off late in calendar year 1948 but did continue on into the early 1949 calendar year. I don't have much actual data from late CJ-2As, that's difficult to come by. One that stands out is CJ-2A 222638 with a front axle date of 11 29 48 and a distributor date of 11C, so some of the components of this 1949 model year CJ-2A were manufactured late in 1948. Another CJ-2A close to that serial number has a front axle date of 12 8 48.