As per the given question,
Let there are m linear equation which has n variables,
"\\begin{cases}\n a_{11}x_1+a_{12}x_2.........+a_{1n}x_n=b_1 \\\\\n a_{21}x_1+a_{21}x_2.........+a_{2n}x_n=b_2 \\\\\na_{31}x_1+a_{32}x_3.........+a_{3n}x_n=b_3 \\\\.\n..\\\\\n...\\\\.\n..\\\\\na_{n_1 1}+a_{n2}x_2.........+a_{nn}x_n=b_n \\\\\n\\end{cases}"
Let it is Ax=b
it have the basic feasible solution if,
"Ax=b ; x \\geq0"
The Simplex Method uses the pivot procedure to move from one BFS to an “adjacent” BFSwith an equal or better objective function value.
Pivot Procedure:
"\\begin{bmatrix}\n...& a_{ij}&... & a_{il}&... \\\\\n...& ...&... &...&...\\\\\n ...&...a_{kj}&...&a_{kl}&...\\\\\n\\end{bmatrix}\\rightarrow \\begin{bmatrix}\n...& 1&... & \\frac{a_{il}}{a_{ij}}&... \\\\\n...& ...&... &...&...\\\\\n ...&...a_{kj}&...&a_{kl}&...\\\\\n\\end{bmatrix}"
3.For each row k(other than row i), add "\u2212a_{kj}x" row i to row k.
The element in row k, column l becomes "\u2212a_{kj}\u00d7a_il+a_kl"
"\\begin{bmatrix}\n...& 1&... & \\frac{a_{il}}{a_{ij}}&... \\\\\n...& ...&... &...&...\\\\\n ...&...a_{kj}&...&a_{kl}&...\\\\\n\\end{bmatrix} \\rightarrow \\begin{bmatrix}\n...& 1&... & \\frac{a_{il}}{a_{ij}}&... \\\\\n...& ...&... &...&...\\\\\n ...&...0&...&a_{kl}-\\frac{a_{kj}a_{il}}{a_{ij}}&...\\\\\n\\end{bmatrix}"
Comments
Leave a comment