# Minimize linear function

• February 8th 2011, 06:18 AM
Tony
Minimize linear function
Hello Everyone, I'm not 100% sure I have posted this in the right section but I think it is.

My problem is as follows:

I have a linear function $AX_{1} + BX_{2} + CX_{3} +DX_{4} = Q \left ( A+B+C+D \right )
$

$A,B,C,D$ and $Q$ are Constants, with $0<=Q<=1$ and $X_{1}, X_{2} ,X_{3},X_{4}$ are variables with $0<=X_{i}<=1$

I have a starting equation (of random values).

for example,

$3\left ( .4 \right )+2\left ( .5\right )+9\left ( .6\right )+4\left ( .3\right ) = .49(3+2+9+4)$

Then I get a set of new values

$11\left ( X_{1} \right )+7\left ( X_{2}\right )+2\left ( X_{3}\right )+36\left ( X_{4}\right ) = .57(11+7+2+36)$

What I want to find are the values of $X_{i}$ that solve the second equation but at the same time minimize the difference between the new $X_{i}$ value and the old (starting) $X_{i}$ value. i.e. minimize $\left( X_{1new} -X_{1old(starting)}\right)$ so in this case $\left( X_{1new} -.4\right)$

I don't know if anyone can help me with this, but even just a suggestion of where to look / what to try would be great.