Spreadsheet Province Simulator - Input wanted

[Trommel]

I've recently returned to this game after 5 years absence and just finished updating and pimping my old spreadsheet simulator. I also wrote some instructions for it so that other players might use it as well.
I'd love to get some feedback on it, especially concerning:

• The usability (user interface, instructions etc.)
• If there's anything important missing
• Input on the "durability indicators" I've constructed in order to facilitate easier comparison between different strategies
• Help finding bugs in there (report to me if you get strange numbers when testing it and I'll look into it)

Since I couldn't attach Excel files in the thread, here's a link to the spreadsheet:
http://ge.tt/6Jo9l8r/v/0?c
(gett is a quick-share site, so the file will only be there for 30 days, but I'll re-upload if needed).


Also, if there are any Solver experts out there, I could use some help getting the Solver to work better on this sheet. It works quite well when optimizing with relatively few constraints, for example maximizing military output (H30) by changing the 6 variables TGs/Forts/Stables (C17, C20, C29) and Off specs/Def spec/Elite per acre (G13, G14, G15) with constraints set to minimum income (H9>x) and buildings (C31<100%). You can even add the durability indicators as constraints and add GS/Hosps/Homes/Banks as variables and it still does ok on optimizing the military output. Another example where it works well is optimizing science distribution between Tools/Housing/Alchemy with the objective to maximizing either BE or Income (with a constraint set on the other of those two, and to military output).
However, my ambition when constructing the spreadsheet was that the solver would be able to optimize military output by changing pretty much every variable that's in there and with some 10+ constraints. When I try that, it doesn't return an optimal solution. Quite far from it actually. I'm not an expert on the Solver, so I don't know if it's simply the formulas that are too complex or intertwined for the Solver to manage it, or if I'm using the wrong settings on the solver for this sort of optimization problem.