set p to zero, get rid of the denominator, move to the left all but the square root, then square the entire equation, after that square the expression in brackets and then some things should cancel out, other things can be added together. Ensure that all expressions containing I are on the left and all the others on the right, then on the left extract I and after that finally divide by the factor you get within brackets after extracting I. Easy as Pi.
$\displaystyle I = \frac{\left(P-R_d\right) \left(P \Theta -A_{\max }-\Theta R_d\right)}{\phi \left(P-A_{\max }-R_d\right)}$
$\displaystyle A_{\max }= \frac{\left(P-R_d\right) \left(P \Theta -I \phi -\Theta R_d\right)}{P-I \phi -R_d}$