Solve Lagrange Multiplier Problem | f(X,Y,Z) = 2XY + 6YZ + 8XZ

Baumer8993
Messages
45
Reaction score
0

Homework Statement



Minimize f(X, Y, Z) = 2XY + 6YZ + 8XZ subject to the constraint XYZ = 12.


Homework Equations


The gradients of the equations, and XYZ = 12.


The Attempt at a Solution



I have the gradients for both of the equations.

∇f = <2Y + 8Z, 2X + 6Z, 6Y + 8X> ∇g = < YZ, XZ, XY> I also have XYZ 12.

I have set them up in there separate equations with the λ in them. I have just stuck on how to solve them. Isn't there a way to plug the equations in the my TI - 83 Plus to solve them?
 
Physics news on Phys.org
Baumer8993 said:

Homework Statement



Minimize f(X, Y, Z) = 2XY + 6YZ + 8XZ subject to the constraint XYZ = 12.


Homework Equations


The gradients of the equations, and XYZ = 12.


The Attempt at a Solution



I have the gradients for both of the equations.

∇f = <2Y + 8Z, 2X + 6Z, 6Y + 8X> ∇g = < YZ, XZ, XY> I also have XYZ 12.

I have set them up in there separate equations with the λ in them. I have just stuck on how to solve them. Isn't there a way to plug the equations in the my TI - 83 Plus to solve them?

No, don't do that; you won't learn anything that way. Use such tools later, to save time, after you have learned the basics.

In this case: use the first equation to solve for Y in terms of Z, use the second equation to solve for X in terms of Z. Now plug those tow expressions into the third equation. This will give you an equation in Z and λ alone. One way to proceed would be to try to solve that equation for Z in terms of λ; there would be more than one solution, corresponding to multiple roots of a higher-order equation. Another way would be to also substitute your expressions for X and Y into the constraint equation, giving a second equation involving z and λ. Now you could try to solve those two equations in the two unknowns in some way.
 
So you have 2Y+ 8Z= \lambda YZ, 2X+ 6Z= \lambda XZ, 6Y+ 8X= \lambda XY, together with XYZ= 12.<br /> <br /> Since a specific value for \lambda is not necessary to solve this problem, it is often simplest to first eliminate \lambda by <b>dividing</b> one equation by another. Here, dividing the first equation by the second, \frac{2Y+ 8Z}{2X+ 6Z}= \frac{Y}{X} which is the same as X(Y+ 4Z)= Y(X+ 3Z) or XY+ 4XZ= XY+ 3YZ which reduces to 4XZ= 3YZ, 4X= 3Y.<br /> <br /> Similarly, dividing the third equation by the second, \frac{6Y+ 8Z}{2X+ 6Z}= \frac{Y}{Z} which is the same as Z(3Y+ 4Z)= Y(X+ 3Z) or 3YZ+ 4Z^2= XY+ 3YZ which reduces to 4Z^2= XY. <br /> <br /> Since Y= (4/3)X, that last equation becomes 4Z^2= (4/3)X^2 or Z^2= 4X^2 and Z= \pm 2X. Continue from there.<br /> <br /> I agree with Ray Vickson- the last thing you want is to have someone or something do those solutions for you. You need to <b>learn</b> the ideas so that you will understand, when you do use &quot;technology&quot;, what is going on.
 
Last edited by a moderator:
There are two things I don't understand about this problem. First, when finding the nth root of a number, there should in theory be n solutions. However, the formula produces n+1 roots. Here is how. The first root is simply ##\left(r\right)^{\left(\frac{1}{n}\right)}##. Then you multiply this first root by n additional expressions given by the formula, as you go through k=0,1,...n-1. So you end up with n+1 roots, which cannot be correct. Let me illustrate what I mean. For this...
Back
Top