D(50) = 350
D(75) = 300
Linear (y = sale price)
D(y) = Ay + b
A = -2
b = 450
D(y) = 450 - 2y
Revenue would be yD(y)
R(y) = 450y - 2y^2
R'(y) = 450 - 4y
R'(y) = 0 @ 112.5
Max revenue @ 112.5 boxes
Max revenue + 450(112.5) - 2(112.5)^2
Max revenue = 25,313
So if C'(x) = 0.5x + 50 then you integrate to get C(x) which is what you really care about.
C(x) = 0.25x^2 + 50x + c
C(200) = 22,500 = 0.25(200)^2 + 50(200) + c
c = 2500
C(x) = 0.25x^2 + 50x + 2500
Average cost would be C(x)/x
AvgC(x) = 0.25x + 50 + 2500/x
Average cost has no global minimum, it decreases steadily as production increases.
Profit is revenue - cost
P(y) = 450y - 2y^2 - (C(D(y))
P(y) = 450y - 2y^2 - [0.25(450-2y)^2 + 50(450-2y) + 2500] which reduces to
P(y) = -3y^2 + 1000y - 75625
P'(y) = -6y + 1000 which is zero @ y = 167
As defined from above, y = sale price
D(167) = 450 - 2(167) = 116 boxes
Max Profit = -3(167)^2 + 1000(167) - 75625
Max Profit = $7708 @ $167/box
2 and 3 require you to do this whole process over again with different constraints. Also I rounded above so the answers are not perfect, which I guess #2 leads to.