04-Some-proves-and-Applications

1 minute read

In []:

1
from scipy.optimize import minimize

Question 1

Consider the two problems \(\begin{align} \text{Maximize } f(x_1,\dots,x_n) \text{ subject to } (x_1,\dots,x_n)\in \Omega\\ \end{align}\) and \(\begin{align} \text{Minimize } -f(x_1,\dots,x_n) \text{ subject to } (x_1,\dots,x_n)\in \Omega \end{align}\) Show that \(\begin{align} f(x_1^*,\dots,x_n^*)=max\{f(x_1,\dots,x_n):(x_1,\dots,x_n) \in \Omega \}\\ \end{align}\) if and only if \(\begin{align} -f(x_1^*,\dots,x_n^*)=min\{-f(x_1,\dots,x_n):(x_1,\dots,x_n) \in \Omega \}\\ \end{align}\)

With out loss of generality we assume these problems have unique solutions.

Assuming:

\[\begin{align} f(x_1^*,\dots,x_n^*)=max\{f(x_1,\dots,x_n):(x_1,\dots,x_n) \in \Omega \}\\ \end{align}\]

We have that:

\[\begin{align} \forall (x_1,\dots,x_n) \in \Omega \backslash \{(x_1^*,\dots,x_n^*)\} &\\ f(x_1^*,\dots,x_n^*) &> f(x_1,\dots,x_n)\\ \text{and therefore}\\ -f(x_1^*,\dots,x_n^*) &< -f(x_1,\dots,x_n) \end{align}\]

which is:

\[\begin{align} -f(x_1^*,\dots,x_n^*)=min\{-f(x_1,\dots,x_n):(x_1,\dots,x_n) \in \Omega \}\\ \end{align}\]

The reverse also holds:

Assuming:

\[\begin{align} -f(x_1^*,\dots,x_n^*)=min\{-f(x_1,\dots,x_n):(x_1,\dots,x_n) \in \Omega \}\\ \end{align}\]

We have that:

\[\begin{align} \forall (x_1,\dots,x_n) \in \Omega \backslash \{(x_1^*,\dots,x_n^*)\} &\\ -f(x_1^*,\dots,x_n^*) &< -f(x_1,\dots,x_n)\\ \text{and therefore}\\ f(x_1^*,\dots,x_n^*) &> f(x_1,\dots,x_n) \end{align}\]

which is:

\[\begin{align} f(x_1^*,\dots,x_n^*)=max\{f(x_1,\dots,x_n):(x_1,\dots,x_n) \in \Omega \}\\ \end{align}\]

Question 2

An import car company has warehouses in cities $A$,$B$ and $C$ and supplies four different dealers $D_1$,$D_2$,$D_3$ and $D_4$. The cost in dollars of transporting a car from a given warehouse to a given dealer is found in the following table. Currently, $D_1$ needs $100$, $D_2$ needs $50$, $D_3$ needs $65$, and $D_4$ needs $75$ cars. How should the cars be shipped in order to minimize the shipping costs?

C=\(\begin{bmatrix} ~ & D_1 & D_2 & D_3 & D_4\\ A & 225 & 150 & 375 & 140\\ B & 105 & 110 & 400 & 200\\ C & 200 & 450 & 310 & 105 \\ \end{bmatrix}\)

X=\(\begin{bmatrix} ~ &Fabrication & Assembly\\ A & 2 & 1 \\ B & 3 & 1\\ C & 2 & 2 \\ \end{bmatrix}\)

L=\(\begin{bmatrix} ~ & \text{Limits}\\ Fabrication & 295\\ Ass & 400\\ C & 300\\ \end{bmatrix}\)

P=\(\begin{bmatrix} ~ & \text{Profits}\\ A & 7\\ B & 8\\ C & 10\\ \end{bmatrix}\)

\[\begin{align} \text{Minimize }&P^TX\\ \text{Subject to }&X_{i,:}\boldsymbol{1}<L_i & i=\{1,\dots,3\}\\ &X_{i,:}^T\boldsymbol{1}<D_i & i=\{1,\dots,4\} \end{align}\]