Big M Minimization 2 B1Ddc6
1. **State the problem:**
Minimize $$z = 2x_1 + 3x_2$$
subject to
$$2x_1 + x_2 \geq 4$$
$$x_1 - x_2 \geq -1$$
$$x_1, x_2 \geq 0$$
2. **Convert inequalities to equalities for Big M method:**
For constraints with $$\geq$$, subtract surplus variables and add artificial variables:
$$2x_1 + x_2 - s_1 + a_1 = 4$$
$$x_1 - x_2 - s_2 + a_2 = -1$$
where $$s_1, s_2 \geq 0$$ are surplus variables and $$a_1, a_2 \geq 0$$ are artificial variables.
3. **Set up the Big M objective function:**
Add a large penalty $$M$$ for artificial variables to force them out:
$$Z = 2x_1 + 3x_2 + M a_1 + M a_2$$
4. **Initial tableau variables:**
Basic variables: $$a_1, a_2$$
Non-basic variables: $$x_1, x_2, s_1, s_2$$
5. **Initial tableau:**
\[
\begin{array}{c|cccccc|c}
& x_1 & x_2 & s_1 & s_2 & a_1 & a_2 & RHS \\
\hline
a_1 & 2 & 1 & -1 & 0 & 1 & 0 & 4 \\
a_2 & 1 & -1 & 0 & -1 & 0 & 1 & -1 \\
\hline
Z & -2 & -3 & 0 & 0 & -M & -M & 0 \\
\end{array}
\]
Note: Objective row coefficients are negative of objective function coefficients.
6. **Apply pivot operations (only allowed are):**
- Swap rows (Ri ↔ Rj)
- Multiply a row by nonzero scalar (kRi)
- Replace a row by itself plus scalar multiple of another (Ri → kRj + Ri)
7. **Goal:**
Remove artificial variables $$a_1, a_2$$ from basis by driving their coefficients in Z-row to zero and then optimize.
8. **Step 1: Remove $$a_1$$ from Z-row**
Add $$M \times$$ row 1 to Z-row:
New Z-row = Z + M * (row 1):
$$(-2 + 2M) x_1 + (-3 + M) x_2 + (-M) s_1 + 0 s_2 + 0 a_1 + (-M) a_2 = 4M$$
9. **Step 2: Remove $$a_2$$ from Z-row**
Add $$M \times$$ row 2 to Z-row:
New Z-row = previous Z-row + M * (row 2):
$$(-2 + 3M) x_1 + (-3) x_2 + (-M) s_1 + (-M) s_2 + 0 a_1 + 0 a_2 = 3M$$
10. **Choose entering variable:**
Most negative coefficient in Z-row is for $$x_2$$ with coefficient $$-3$$.
11. **Determine leaving variable:**
Calculate ratios for positive coefficients in $$x_2$$ column:
Row 1: $$\frac{4}{1} = 4$$
Row 2: $$\frac{-1}{-1}$$ negative denominator, ignore.
So, row 1 leaves.
12. **Pivot on row 1, column $$x_2$$:**
Make pivot element 1 by dividing row 1 by 1.
Row 1: $$[2,1,-1,0,1,0,4] \to [2,1,-1,0,1,0,4]$$ (already 1 in $$x_2$$ column)
13. **Eliminate $$x_2$$ from other rows:**
Row 2: new row 2 = row 2 + row 1 (since coefficient is -1):
$$[1,-1,0,-1,0,1,-1] + [2,1,-1,0,1,0,4] = [3,0,-1,-1,1,1,3]$$
Z-row: new Z-row = Z-row + 3 * row 1 (since coefficient is -3):
$$(-2 + 3M, -3, -M, -M, 0, 0, 3M) + 3 * (2,1,-1,0,1,0,4) = (-2 + 3M + 6, 0, -M - 3, -M, 3, 0, 3M + 12)$$
Simplify:
$$Z = (4 + 3M) x_1 + 0 x_2 + (-M - 3) s_1 + (-M) s_2 + 3 a_1 + 0 a_2 = 3M + 12$$
14. **Next entering variable:**
Most negative coefficient in Z-row is $$-M - 3$$ for $$s_1$$ (assuming large $$M$$, negative).
15. **Determine leaving variable:**
Check ratios for positive $$s_1$$ coefficients:
Row 1: $$s_1 = -1$$ negative, ignore.
Row 2: $$s_1 = -1$$ negative, ignore.
No positive coefficients for $$s_1$$, so no leaving variable.
16. **Next entering variable:**
Check $$s_2$$ coefficient $$-M$$ negative, no positive coefficients in constraints.
17. **Next entering variable:**
Check $$a_1$$ coefficient $$3$$ positive, so no improvement.
18. **Next entering variable:**
Check $$x_1$$ coefficient $$4 + 3M$$ positive, no improvement.
19. **Since no further pivots possible to reduce Z, and artificial variables remain in basis, problem is infeasible or requires further analysis.**
20. **Summary:**
The Big M method tableau shows artificial variables cannot be removed with given pivot operations and constraints, indicating no feasible solution under current setup.
**Final answer:** No feasible solution found using Big M method tableau with given pivot operations.