From how I see the WitP engine work, fuel is only used by ships. Supply points are used by everything else. Therefore, fuel does seem to only represent ship fuel.
You are correct. There are two technical problems with this:
1) The ships in the game do not all use the same "fuel." The most common cases include fuel oil, diesel oil and coal. There also are a few vessels which use gasoline. And in the case of steamships that normally burn fuel oil, in some instances they can burn crude oil - and this was important in this campaign. These factors mean that some of what we consider "supply points" (representing diesel oil and gasoline) and "oil" (representing fuel oil) and "resources" (representing coal) should be converted to "fuel" - based on some set of assumptions considered reasonable. These factors will INCREASE the ship fuel over what you think it should be.
2) Military units (and in some cases where there is heavy industry being fed non-military entities) will consume some of the fuel you think is ship fuel. This is exaserbated because some of the stored fuels at all major sites is actually "supply points" (representing diesel oil and gasoline) - and not fuel oil in the formal sense at all. These factors will DECREASE the ship fuel over what you think it should be.
All I have tried to say is that you need to consider all these factors in calculating the values - because otherwise you will not have correct values for inventories in game terms. I have not said anything about your data per se - except it is not detailed enough. Further while it may indicate the storage capacity - it probably does not even do that - capacity was probably greater than the numbers you have - because not all tanks were full - and because you are not counting non tank storage in various watercraft used to store fuels. IF your data really is entirely ship fuel oil, there is a good deal of other petroleum stored at these locations as well. More likely, your data includes those other products - and you need to figure out to what extent? Either way, you need estimates for all fuels, and all fuel usage requirements that draw on these facilities, and all other facilities and vessels used for storage not included as well.
I am not yet addressing this matter - except informally - to be helpful. But I probably have much of the data - and I will address it when I have finished with planes and ships and devices. But rather than make you wait, I tried to help you understand how to think about this information - and how raw data must be modified to get game values. If you don't think about doing this properly, you cannot get correct values. This seems likely to be what happened to the original data.
Your contract data is not necessairily bad - but it is necessairily incomplete. You must add to it all civil data - including that from foreign companies and governments. Further, if you are using Navy data, you may have to add some army material as well. There really was a B-17 route criss crossing the Pacific (by a very zig zaggy route) from Hawaii to the Philippines that did NOT use the Pan Am Clipper route - and this route was supplied with aviation fuel - and of course it could be stored at each island. If you have not identified all these islands and their facilities, you are not yet complete. Then there is the data for Japanese islands - see Nanyo for it - and data for neutrals as well - although you may reject using that data - you must know what it is before you decide. Near the map edges - e.g. in Siberia and Alaska - you may also decide to ignore some ports - but you need to understand what you are rejecting before you do that. Thus almost everyone would reject Whittier - but when you know that Elmendorf depends on it - you may change your mind.
