Skip to content

Simplex Algorithm

Solve the following optimization problem $$ \begin{alignat*}{2} & \text{maximize: } && c^Tx \ & \text{subject to: } && \begin{aligned}[t] Ax &\leq b\ x &\geq 0 \end{aligned} \end{alignat*} $$

Concrete Problem

\[ \begin{alignat*}{2} & \text{maximize: } && z = 2x + 3y\\ & \text{subject to: } && \begin{aligned}[t] x + 3y &\leq 12 \\ 3x + 2y &\leq 12 \\ x, y &\geq 0 \end{aligned} \end{alignat*} \]

1. Step I

Fromulating the problem and creating the first table for simplex algorithm.

Basic \(x\) \(y\) \(s_{1}\) \(s_{2}\) Solution Ratio
\(z\) \(-2\) \(-3\) \(0\) \(0\) \(0\)
\(s_{1}\) \(1\) \(3\) \(1\) \(0\) \(12\)
\(s_{2}\) \(3\) \(2\) \(0\) \(1\) \(12\)

2. Step II

We have to choose an entering variable which will increase the value of objective. So we choose any variable with negative coefficient. Here we choose \(y\).

Basic \(x\) \(y\) \(s_{1}\) \(s_{2}\) Solution Ratio
\(z\) \(-2\) \(-3\) \(0\) \(0\) \(0\) \(0\)
\(s_{1}\) \(1\) \(3\) \(1\) \(0\) \(12\) \(4\)
\(s_{2}\) \(3\) \(2\) \(0\) \(1\) \(12\) \(6\)

3. Step III

We have to choose a leaving variable based on ratio analysis. Here we choose the variable with least positive ratio. In this table \(s_1\) has the least positive ratio, so it will leave. Now we have to perform the pivoting step.

4. Step IV

After pivoting we get a new table. Where we can repreat the above steps again and again till the simplex algorithm terminate. In ideal case we get no entering varable.

Basic \(x\) \(y\) \(s_{1}\) \(s_{2}\) Solution Ratio
\(z\) \(-1\) \(0\) \(1\) \(0\) \(12\)
\(y\) \(\frac{1}{3}\) \(1\) \(\frac{1}{3}\) \(0\) \(4\)
\(s_{2}\) \(\frac{7}{3}\) \(0\) \(- \frac{2}{3}\) \(1\) \(4\)
Basic \(x\) \(y\) \(s_{1}\) \(s_{2}\) Solution Ratio
\(z\) \(-1\) \(0\) \(1\) \(0\) \(12\) \(-12\)
\(y\) \(\frac{1}{3}\) \(1\) \(\frac{1}{3}\) \(0\) \(4\) \(12\)
\(s_{2}\) \(\frac{7}{3}\) \(0\) \(- \frac{2}{3}\) \(1\) \(4\) \(\frac{12}{7}\)
Basic \(x\) \(y\) \(s_{1}\) \(s_{2}\) Solution Ratio
\(z\) \(0\) \(0\) \(\frac{5}{7}\) \(\frac{3}{7}\) \(\frac{96}{7}\)
\(y\) \(0\) \(1\) \(\frac{3}{7}\) \(- \frac{1}{7}\) \(\frac{24}{7}\)
\(x\) \(1\) \(0\) \(- \frac{2}{7}\) \(\frac{3}{7}\) \(\frac{12}{7}\)

Finally we got a table with all non-negaitve coefficient corresponding to \(z\) variable. The solution of this problem is \(96/7\) at the point \((12/7, 24/7)\)

5. Problem 2

Solve the following optimization problem.

\[ \begin{alignat*}{2} & \text{maximize: } && z = 2x_1 + x_2 - 3x_3 + 5x_4 \\ & \text{subject to: } && \begin{aligned}[t] x_1 + 2x_2 + 2x_3 + 4x_4 &\leq 40 \\ 2x_1 - x_2 + x_3 + 2x_4 &\leq 8 \\ 4x_1 - 2x_2 + x_3 - x_4 &\leq 10 \\ x_1, x_2, x_3, x_4 &\geq 0 \end{aligned} \end{alignat*} \]

\(\displaystyle \left[\begin{matrix}-2 & -1 & 3 & -5 & 0 & 0 & 0 & 0\\1 & 2 & 2 & 4 & 1 & 0 & 0 & 40\\2 & -1 & 1 & 2 & 0 & 1 & 0 & 8\\4 & -2 & 1 & -1 & 0 & 0 & 1 & 10\end{matrix}\right]\)

Basic \(x_{1}\) \(x_{2}\) \(x_{3}\) \(x_{4}\) \(s_{1}\) \(s_{2}\) \(s_{3}\) Solution Ratio
\(z\) \(-2\) \(-1\) \(3\) \(-5\) \(0\) \(0\) \(0\) \(0\)
\(s_{1}\) \(1\) \(2\) \(2\) \(4\) \(1\) \(0\) \(0\) \(40\)
\(s_{2}\) \(2\) \(-1\) \(1\) \(2\) \(0\) \(1\) \(0\) \(8\)
\(s_{3}\) \(4\) \(-2\) \(1\) \(-1\) \(0\) \(0\) \(1\) \(10\)
Basic \(x_{1}\) \(x_{2}\) \(x_{3}\) \(x_{4}\) \(s_{1}\) \(s_{2}\) \(s_{3}\) Solution Ratio
\(z\) \(-2\) \(-1\) \(3\) \(-5\) \(0\) \(0\) \(0\) \(0\) \(0\)
\(s_{1}\) \(1\) \(2\) \(2\) \(4\) \(1\) \(0\) \(0\) \(40\) \(10\)
\(s_{2}\) \(2\) \(-1\) \(1\) \(2\) \(0\) \(1\) \(0\) \(8\) \(4\)
\(s_{3}\) \(4\) \(-2\) \(1\) \(-1\) \(0\) \(0\) \(1\) \(10\) \(-10\)
Basic \(x_{1}\) \(x_{2}\) \(x_{3}\) \(x_{4}\) \(s_{1}\) \(s_{2}\) \(s_{3}\) Solution Ratio
\(z\) \(3\) \(- \frac{7}{2}\) \(\frac{11}{2}\) \(0\) \(0\) \(\frac{5}{2}\) \(0\) \(20\)
\(s_{1}\) \(-3\) \(4\) \(0\) \(0\) \(1\) \(-2\) \(0\) \(24\)
\(x_{4}\) \(1\) \(- \frac{1}{2}\) \(\frac{1}{2}\) \(1\) \(0\) \(\frac{1}{2}\) \(0\) \(4\)
\(s_{3}\) \(5\) \(- \frac{5}{2}\) \(\frac{3}{2}\) \(0\) \(0\) \(\frac{1}{2}\) \(1\) \(14\)
Basic \(x_{1}\) \(x_{2}\) \(x_{3}\) \(x_{4}\) \(s_{1}\) \(s_{2}\) \(s_{3}\) Solution Ratio
\(z\) \(3\) \(- \frac{7}{2}\) \(\frac{11}{2}\) \(0\) \(0\) \(\frac{5}{2}\) \(0\) \(20\) \(- \frac{40}{7}\)
\(s_{1}\) \(-3\) \(4\) \(0\) \(0\) \(1\) \(-2\) \(0\) \(24\) \(6\)
\(x_{4}\) \(1\) \(- \frac{1}{2}\) \(\frac{1}{2}\) \(1\) \(0\) \(\frac{1}{2}\) \(0\) \(4\) \(-8\)
\(s_{3}\) \(5\) \(- \frac{5}{2}\) \(\frac{3}{2}\) \(0\) \(0\) \(\frac{1}{2}\) \(1\) \(14\) \(- \frac{28}{5}\)
Basic \(x_{1}\) \(x_{2}\) \(x_{3}\) \(x_{4}\) \(s_{1}\) \(s_{2}\) \(s_{3}\) Solution Ratio
\(z\) \(\frac{3}{8}\) \(0\) \(\frac{11}{2}\) \(0\) \(\frac{7}{8}\) \(\frac{3}{4}\) \(0\) \(41\)
\(x_{2}\) \(- \frac{3}{4}\) \(1\) \(0\) \(0\) \(\frac{1}{4}\) \(- \frac{1}{2}\) \(0\) \(6\)
\(x_{4}\) \(\frac{5}{8}\) \(0\) \(\frac{1}{2}\) \(1\) \(\frac{1}{8}\) \(\frac{1}{4}\) \(0\) \(7\)
\(s_{3}\) \(\frac{25}{8}\) \(0\) \(\frac{3}{2}\) \(0\) \(\frac{5}{8}\) \(- \frac{3}{4}\) \(1\) \(29\)

Since all the coefficient corresponding to \(z\) is non-negative, hence there will be no entering variable. The simplex algorithm will terminate. The objective is \(41\) at the point \((0,6,0,7)\).

Last modified on: 2023-01-06 22:22:45