cost minimization

**mech.engineer.major** · December 14th 2008, 11:51 AM

I need some help with a problem involving the optimization of area.

This is the prompt to the problem:

"A closed box with square base is to be built to house an ant colony. The bottom of the box and all four sides will be made of material costing $1 per square foot, and the top is to be constructed of glass costing $5 per square foot. What are the dimensions of the box of greatest volume than can be constructed for $72?"

I have worked several optimization problems both involving cost and not but in this problem I am given the cost and not the area. I understand the process of optimization: Construct an equation that models the problem (hopefully with a sketch), use the given value in its respective equation and solve for one of the variables, plug this quantity into the main equation and solve for the other variable.

I guess I should start with checking if my equation is correct.

Surface Area = sides + top + bottom

SA = 4xy + x^2 + x^2 Volume = x^2*y

Cost = 4xy(1) + x^2(5) + x^2(1) Given value : $72 (max cost)

Cost = 4xy + 5x^2 + x^2

With problems like this I usually just take the derivative and doing that gave me the wrong dimensions. So going back I think that I need to set the cost equation equal to 72. However that would then incorporate implicit differentiation into an optimization problem and I have never done that before, just implicit by itself. Am I headed in the right direction and if so then what?

**Greengoblin** · December 14th 2008, 02:58 PM

Let x be depth, y be width, and z be height. Then,

$xy+2xz+2yz+5xy=72$

$2xz+2yz+6xy=72$

You can solve this using partial derivatives, by writing volume as a function of only two variables x, and y, and taking $\nabla f =0$ .

Or you can use the AM-GM inequality:

$\frac{2xz+2yz+6xy}{3}\ge(2xz\cdot 2yz\cdot 6xy)^{\frac{1}{3}}$

$\frac{72}{3}\ge(2xz\cdot 2yz\cdot 6xy)^{\frac{1}{3}}$

$24^3\ge 2xz\cdot 2yz\cdot 6xy$

$24^3\ge 24x^2y^2z^2$

$24^2\ge x^2y^2z^2$

$24\ge xyz$

$V \le 24 \implies V_{max}=24\text{ft}^3$

(The dimensions of your box are the set $\{x,y,z\in\mathbb{R}:xyz=24\}$ )

**mech.engineer.major** · December 14th 2008, 04:09 PM

Thank you for your help Greengoblin. Your final answer is correct according to my answer sheet but you created one more variable than is needed Greengoblin. I always learned that when a square base is involved one variable should be used for both of them. that is why my equation for volume is V=x^2*y. Also I have not learned about the AM-GM inequality so that is probably not a method I need to know yet. Also I'm not sure as to how I would solve it with partial derivatives.

**Greengoblin** · December 14th 2008, 04:38 PM

You're welcome. I mis-read the part about having a square base as having a rectangular base, so replacing any instances of xy in the above calculation with $x^2$ , should give more appropriate expressions as you pointed out. (This is also why I though partial derivatives would be needed..If you haven't heard of them yet, don't worry..their concerned with differentiating functions of several variables, which would have been the case had we had three independent dimensions.)

The calculus way is:

$x^2+4xy+5x^2=72$

$6x^2+4xy=72$

Now you can solve for y,

$y=\frac{72-6x^2}{4x}$

...and get volume as a function of x only, by substitution:

$V(x,y)=x^2y\implies V(x)=\frac{72x^2-6x^4}{4x}=18x-\frac{3}{2}x^3$

Now you can find f'(x)...

**mech.engineer.major** · December 14th 2008, 04:56 PM

you know I just realized why I wasn't getting the problem right. I was taking the right approach but seeing the way you set up the equation allowed me to see where my mistake was.

I must have confused the surface area equation with the cost equation after incorporating the $1 and $5 values for their respective faces. I took the derivative of this equation thinking the top and bottom quantities need to stay separate:

[IMG]file:///C:/Users/Robert/AppData/Local/Temp/moz-screenshot-2.jpg[/IMG]

Instead of adding them like you did and taking the derivative of this equation:

That's what threw me off all along. I considered doing that earlier apparently didn't follow through. Wow definitely one of those "oh......duh" moments. Thank you again I appreciate the help.

**Greengoblin** · December 14th 2008, 05:07 PM

Ok, glad you got it

Thread: cost minimization

LinkBack

Thread Tools

Display

cost minimization

Similar Math Help Forum Discussions

Cost minimization while retaining volume

Process cost systems,total cost under the FIFO Method.

Marginal Cost/Demand and Cost

Cost Minimization function with exponent

Minimization

Search Tags