Faster calculation for multiple automatic goal seek

A sample spreadsheet (or at least a description of the calculations being performed) would help. A couple of thoughts:

1) Sometimes we use Goal Seek to avoid doing the algebra. In a case like this, one of the first things I would do is check the function in the target cell and see if, using algebra, I can solve the function. Then you'll be able to put 252763 into H68, and put a formula into C68 that will correspond to H68=252763 without even calling Goal Seek.

2) If the function is one that cannot readily be solved algebraically, I will often write my own algorithm for finding the solution. Goal Seek uses the Newton Raphson method, which is not overly difficult (if you are already familiar with the technique) to code into a UDF. Again, this eliminates the need to call the Goal Seek algorithm.

Can you turn screen updating off? That might speed things up a bit.

I don't think my issue can be solved by using algebraic function in column "A", because there are "if" function in column E.

I expect you already know this, but the if function in column E is there to make sure E never drops below 0 (0<=E<=$K$4). The threshold value for this condition is A=$N$16. The interesting thing -- When E=0 (when A<=$N$16), a lot of the formula in F algebraically drops out, making F (and H) very simple algebraic functions. When A<=$N$16 you should have no trouble solving for x. This could save a significant chunk of computation time, because you will not even need to run goal seek for these entries.

The main difficulty will be an algorithm for those cases where A>$N$16. If you solve for A assuming E is 0, you should get a good 1st guess at what x should be. From there, figure out a suitable "successive approximation" algorithm that will allow you to solve for x. Perhaps something like this:

1) Calculate initial guess for A from target value for H assuming E is 0 (same calculation as when A<=$N$16).
2) From this initial guess for A calculate Fest and Hest.
3) Using a suitable algorithm, compare Hest with Htarget and use that information to come up with a new guess for A. I might use a linear interpolation type routine (using the TREND() function) to get this next guess for A.
4) Compare this new A to the previous value for A. If they are the same, exit loop and end calculation. If they are different, return the step 2 with the new A and repeat 2-4.

If you are comfortable using circular references (with iteration enabled), you could very likely program this calculation without needing to use VBA or Goal Seek at all. If not, there are other options. In any case, I think this should be doable so that it calculates nearly instantaneously. I expect you will need further help building this spreadsheet, so let us know where you get stuck, and we can try to help you get this together.

If you still want to use your Goal Seek VBA procedures, you might look at them more closely (I don't use VBA to call Goal Seek, so I'm not very good at programming those procedures). I'm not certain why they should take so long to calculate. H is a fairly straightforward, almost linear function of A, so it should not need very many iterations to converge. Perhaps what you need is to use the idea I used of assuming E=0 to give Goal Seek a better 1st guess for A before calling Goal Seek.