My main finding is that I think it will be difficult to build the Supply stockpile within Japan to be larger than about 3 million.
I am about a month into my latest IJ vs AI game with my main purpose being to test this graph and the assumptions behind it.
Many details in the graph below will differ from game to game based on choices the Japan player makes about different manufacturing, but this should be enough to show some of the things that the Japan player can expect to happen as a result of those choices.
Background:
1. This is for the Japan "economic region" only, Scenario 1. The inputs have come from a separate detailed spreadsheet on each type of production and manufacturing where I have forecast different inputs required, including Supply. Production of Supply at full capacity within Japan is 19k per day, or close enough to 600k per month on average.
Major demands on Supply:
2. New industry creation, followed by repair. This graph assumes that all industry is setup in the first few days, as soon as enough Supply is available at each base. A number of experienced players have advised against doing too much industry expansion too soon, but one of the main reasons that I started building this model was to test that advice and try to figure out an answer to “how much is too much?” What I’ve found is that I can setup my pre-war plan in the first week of the game and while it does use a lot of Supply and Manpower, it doesn't outright "crash" the economy. This should let me start seeing the benefits of the industry expansion as quickly as possible. Interestingly, the Manpower pool ran out for me on day 2. So instead of creating all of your large (100+) engine/aircraft production factories in Dec 41, you can create them in +30 increments on say the 1st of each month until they reach the size you want, as they will only repair at 1k Supply per day anyway. You just need to write your pre-war plan down and refer back to it on the 1st of each month. I definitely can’t keep all the inter-related details of IJ war manufacturing in my head, so I feel that coming up with a detailed plan is essential so that you have something to adjust from as the war situation changes.
3. Engine research repair can all be completed within the first 30 days, which will start to accelerate the arrival of the later model engines. This typically needs 30k of supply per research factory to do this. I've chosen size 30 factories to suit my aircraft research/production plans. This is one area where you could delay the later war engines if you don’t need them to be accelerated to align with your aircraft research.
4. Aircraft research factories repair as a function of time until the original due date. This line on the graph will change a lot player to player, depending on your aircraft research/production plan. I've straight lined the Supply spend estimate for this for simplicity, but I know it is more of an exponential function. In reality the Supply spend I’ve shown on this graph will be less earlier/more later due to my technically incorrect straight line forecast. There are two reasons for why I have setup aircraft research factories: the first and most important one is to get certain aircraft quicker (mainly fighters). I don't want to delay this research. The second is to setup other aircraft factories (mainly non-fighters) for immediate production once they have finished research. These are the factories that I would consider delaying next time as while I want those new aircraft in time, I am happy to keep the current models until they arrive (eg: recon, medium bomber, transport).
Applicable to all of 2,3 & 4 above, as you will need to change the aircraft factories and engines throughout the war, these activities appear as jagged lines. Your spikes will change depending on aircraft types and numbers that you select, but it does show that your aircraft/engine manufacturing plan makes up a large percentage of your Supply burn.
5. LCU creation is another "lumpy" line on the graph. This shows the Supply points needed just to raise and equip the new LCUs that arrive in Japan (assuming that you have kept sufficient Manpower, Vehicle and Armament points, which are sold separately).
6. LCU "sustainment" is the Supply needed to then feed the Army. This can be seen to grow from not a lot to a fair bit by late war.
7. Supply used by Airgroups arriving and in training, but no allowance for airgroups conducting combat missions which would be more.
8. Export of Supply from Japan to support military operations in areas with large Supply deficiencies to support operations (OpsExport on the graph). My forecast is that a total of 8700 Supply / day (or about 260k/month) is required just to keep Armies fed in places like China, Burma, Philippines, 4th Fleet and SE Area Fleet. I am going to try and delay this Supply export until Apr 1942 to allow the majority of industry to be expanded first, but until then any export of Supply from Japan would mean that some form of industry expansion has to be delayed. These sorts of choices can be informed by Supply use forecasts rather than just leaving it to random guesswork.
Other relatively minor demands on Supply shown in this graph include:
9. Vehicle factory repairs. I get this done early as it links directly to my forecast demand from LCUs.
10. Naval Shipyard Repairs. I only increase a few, based on my naval ship building plan so again I get them done and then forget about naval shipyards until it comes time to start switching some off.
11. Repair Shipyard repairs. I only increase Tokyo to size 50 to allow the CS to CVL conversions. This could be delayed a bit, but I'd probably forget if I didn't just do it on day 1.
12. Oil repairs. This was just a prompt to allocate enough supply to repairing Miri oilfields after they are captured. I'm going to try and use Supply from Indochina to contribute to this and any other Oil fields that are damaged on capture. It doesn't all need to be moved in one go as the repairs can only be done 1k/day or 30k/month, but by about June 1942 this should all be done.
Things not included on this graph, which will push more months into Supply deficit than shown here, include:
13. Reductions to Supply production due to fuel shortages that may occur late war (impacting HI centres therefore Supply production)
14. As above, but anything due to enemy activity such as naval or aerial bombing HI, LI, Supply stockpiles, etc
15. Any use for military operations, such as ground combat in Japan, offensive air missions or rearming naval ships
Also, the graph does not factor in any Supply imports to Japan which may be possible, again depending on the game situation. Before I move Supply back to Japan I would first want to leave enough of it in those forward areas to support operations, but any Supply excess I generate, especially in places like Java, I will move to where I think it will best support future operations.
You do need to come up with an industry and manufacturing plan that is affordable within the constraints of the starting Supply stockpile and the 19k/day Supply production rate. If you do that, and warning lights still come on because of low Supply in Japan, you can always just toggle off a select few industry repairs to avoid the deficit, switching them back on once your higher priority factories are fully repaired.
But if you don’t carefully consider the Supply input/output equation and massively overdo your factory expansions you will waste a lot of Supply that can never be replaced, which will limit your ability to fight as well as you otherwise could have in the mid and late war.
