Q 1
(a) True. In degenerate LP where two or more basic variables tie for being leaving variable, no matter which one is chosen, in the step 2 of the iteration. The one or ones not chosen to be the leaving variable also will have a value of zero in the new BF solution.
(b) True. The simplex method would stop with the result that Z is unbounded, which means that a mistake for real problem has been made and the model probably has been misformulated, alternatively, a computational mistake has occurred.
© False. When the problem has multiple optimal solution, at least one of the
n
o
n
−
b
a
s
i
c
‾
\underline{non-basic}
non−basic variables has a coefficient of zero in the final row 0.
(b) True. If the iteration has no leaving variable, the feasible region is unbounded.
Q 2
(a) 
The feasible region is the shadow part showed below the blue line in the picture.As it indicates, the feasible region is unbounded.
(b) Yes. At the line :
−
x
1
+
x
2
=
0
-x_1+x_2=0
−x1+x2=0 ,move along
x
1
x_1
x1, it could find the optional solution at the stop at
x
1
=
10
x_1=10
x1=10,
Z
m
=
−
10
Z_m=-10
Zm=−10.
© The optional solution equals to the one in (b).
(d) No,it might exist at the initial point of the feasible space. Ignore the point equal to zero.
Q 3
bv Z x 1 x 2 x 3 x 4 s 1 s 2 r h s Z 1 − 1 − 1 − 1 − 1 0 0 0 s 1 0 1 1 0 0 1 0 3 s 2 0 0 0 1 1 0 1 2 (1) \begin{array}{c|clrcccc|} \text{bv}&\text{Z}&x_1&x_2&x_3&x_4&s_1&s_2&rhs \\ \hline Z&1&-1&-1&-1&-1&0&0&0 \\ \hline s_1&0&1&1&0&0&1&0&3\\ \hline s_2&0&0&0&1&1&0&1&2\\ \end{array} \tag1 bvZs1s2Z100x1−110x2−110x3−101x4−101s1010s2001rhs032(1)
bv Z x 1 x 2 x 3 x 4 s 1 s 2 r h s Z 1 − 1 − 1 0 0 0 1 2 s 1 0 1 1 0 0 1 0 3 x 4 0 0 0 1 1 0 1 2 (2) \begin {array}{c|clrcccc|} \text{bv}&\text{Z}&x_1&x_2&x_3&x_4&s_1&s_2&rhs\\ \hline Z&1&-1&-1&0&0&0&1&2\\ \hline s_1&0&1&1&0&0&1&0&3\\ \hline x_4&0&0&0&1&1&0&1&2\\ \end{array} \tag2 bvZs1x4Z100x1−110x2−110x3001x4001s1010s2101rhs232(2)
b v Z x 1 x 2 x 3 x 4 s 1 s 2 r h s Z 1 0 0 0 0 1 1 5 x 2 0 1 1 0 0 1 0 3 x 4 0 0 0 1 1 0 1 2 (3) \begin{array}{c|clrcccc|} bv&Z&x_1&x_2&x_3&x_4&s_1&s_2&rhs\\ \hline Z&1&0&0&0&0&1&1&5\\ \hline x_2&0&1&1&0&0&1&0&3\\ \hline x_4&0&0&0&1&1&0&1&2 \end{array} \tag3 bvZx2x4Z100x1010x2010x3001x4001s1110s2101rhs532(3)
The optimal solution is ( x 1 , x 2 , x 3 , x 4 ) = ( 0 , 3 , 0 , 2 ) (x_1,x_2,x_3,x_4)=(0,3,0,2) (x1,x2,x3,x4)=(0,3,0,2), and the subsequent Z m = 5 Z_m=5 Zm=5.
Q 4
(a)
There is no cross region between the two constraint space. Consequently, the problem has no feasible region.
(b)
Big M method
m
i
n
Z
=
m
a
x
(
−
Z
)
,
l
e
t
−
Z
=
Z
‾
min Z=max( -Z),let -Z=\overline{Z}
minZ=max(−Z),let−Z=Z
Z
‾
=
−
5000
x
1
−
7000
x
2
\overline{Z}=-5000x_1-7000x_2
Z=−5000x1−7000x2
Let
M
→
∞
M\rightarrow\infty
M→∞
max
Z
‾
\overline{Z}
Z
s.t.
5000
x
1
+
7000
x
2
+
M
x
3
‾
+
M
x
4
‾
+
Z
‾
=
0
(
0
)
5000x_1+7000x_2+M\overline{x_3}+M\overline{x_4}+\overline{Z}=0\quad(0)
5000x1+7000x2+Mx3+Mx4+Z=0(0)
−
2
x
1
+
x
2
+
s
1
+
x
3
‾
=
1
(
1
)
\quad-2x_1+x_2+s_1+\overline{x_3}=1\qquad(1)
−2x1+x2+s1+x3=1(1)
x
1
−
2
x
2
+
s
2
+
x
4
‾
=
1
(
2
)
\quad x_1-2x_2+s_2+\overline{x_4}=1\qquad(2)
x1−2x2+s2+x4=1(2)
New eq.(0)=eq.(0)-eq.(1)-eq.(2)=
Z
‾
+
(
5000
+
M
)
x
1
+
(
7000
+
M
)
x
2
−
M
s
1
−
M
s
2
=
−
2
M
\overline{Z}+(5000+M)x_1+(7000+M)x_2-Ms_1-Ms_2=-2M
Z+(5000+M)x1+(7000+M)x2−Ms1−Ms2=−2M
Iteration 0
b
v
Z
‾
x
1
x
2
x
3
‾
s
1
s
2
x
4
‾
r
h
s
Z
‾
1
5000
+
M
7000
+
M
0
−
M
−
M
0
−
2
M
x
3
‾
0
−
2
1
1
1
0
0
1
x
4
‾
0
1
−
2
0
0
1
1
1
\begin{array}{c|clr} bv&\overline{Z}&x_1&x_2&\overline{x_3}&s_1&s_2&\overline{x_4}&rhs\\ \hline \overline{Z}&1&5000+M&7000+M&0&-M&-M&0&-2M\\ \hline\overline{x_3}&0&-2&1&1&1&0&0&1\\ \hline \overline{x_4}&0&1&-2&0&0&1&1&1\\ \end{array}
bvZx3x4Z100x15000+M−21x27000+M1−2x3010s1−M10s2−M01x4001rhs−2M11
Iteration 1
b
v
Z
‾
x
1
x
2
x
3
‾
s
1
s
2
x
4
‾
r
h
s
Z
‾
1
5000
−
M
7000
+
2
M
M
0
−
M
0
−
M
s
1
0
−
2
1
1
1
0
0
1
x
4
‾
0
1
−
2
0
0
1
1
1
\begin{array}{c|clr} bv&\overline{Z}&x_1&x_2&\overline{x_3}&s_1&s_2&\overline{x_4}&rhs\\ \hline\overline{Z}&1&5000-M&7000+2M&M&0&-M&0&-M\\ \hline s_1&0&-2&1&1&1&0&0&1\\ \hline\overline{x_4}&0&1&-2&0&0&1&1&1\\ \end{array}
bvZs1x4Z100x15000−M−21x27000+2M1−2x3M10s1010s2−M01x4001rhs−M11
Iteration 2
b
v
Z
‾
x
1
x
2
x
3
‾
s
1
s
2
x
4
‾
r
h
s
Z
‾
1
5000
7000
M
0
0
M
0
s
1
0
−
2
1
1
1
0
0
1
s
2
0
1
−
2
0
0
1
1
1
\begin{array}{c|clr} bv&\overline{Z}&x_1&x_2&\overline{x_3}&s_1&s_2&\overline{x_4}&rhs\\ \hline \overline{Z}&1&5000&7000&M&0&0&M&0\\ \hline s_1&0&-2&1&1&1&0&0&1\\ \hline s_2&0&1&-2&0&0&1&1&1\\ \end{array}
bvZs1s2Z100x15000−21x270001−2x3M10s1010s2001x4M01rhs011
As it indicates, the original LP is infeasible.
© Two phase method
Phase 1：maximize W = − x 3 ‾ − x 4 ‾ W=-\overline{x_3}-\overline{x_4} W=−x3−x4
s.t.
−
2
x
1
+
x
2
+
s
1
+
x
3
‾
=
1
(
1
)
-2x_1+x_2+s_1+\overline{x_3}=1\quad(1)
−2x1+x2+s1+x3=1(1)
x
1
−
2
x
2
+
s
2
+
x
4
‾
=
1
(
2
)
\quad x_1-2x_2+s_2+\overline{x_4}=1\quad(2)
x1−2x2+s2+x4=1(2)
New eq.(0)=eq.(0)-eq.(1)-eq.(2)
W
+
x
1
+
x
2
−
s
1
−
s
2
=
−
2
(
0
)
W+x_1+x_2-s_1-s_2=-2\quad(0)
W+x1+x2−s1−s2=−2(0)
Iteration 0
b
v
W
x
1
x
2
s
1
s
2
x
3
‾
x
4
‾
r
h
s
W
1
1
1
−
1
−
1
0
0
−
2
x
3
‾
0
−
2
1
1
0
1
0
1
x
4
‾
0
1
−
2
0
1
0
1
1
\begin{array}{c|clr} bv&W&x_1&x_2&s_1&s_2&\overline{x_3}&\overline{x_4}&rhs\\ \hline W&1&1&1&-1&-1&0&0&-2\\ \hline\overline{x_3}&0&-2&1&1&0&1&0&1\\ \hline\overline{x_4}&0&1&-2&0&1&0&1&1 \end{array}
bvWx3x4W100x11−21x211−2s1−110s2−101x3010x4001rhs−211
Iteration 1
b
v
W
x
1
x
2
s
1
s
2
x
3
‾
x
4
‾
r
h
s
W
1
−
1
2
0
−
1
1
0
−
1
s
1
0
−
2
1
1
0
1
0
1
x
4
‾
0
1
−
2
0
1
0
1
1
\begin{array}{c|clr} bv&W&x_1&x_2&s_1&s_2&\overline{x_3}&\overline{x_4}&rhs\\ \hline W&1&-1&2&0&-1&1&0&-1\\ \hline s_1&0&-2&1&1&0&1&0&1\\ \hline \overline{x_4}&0&1&-2&0&1&0&1&1\\ \end{array}
bvWs1x4W100x1−1−21x221−2s1010s2−101x3110x4001rhs−111
Iteration 2
b
v
W
x
1
x
2
s
1
s
2
x
3
‾
x
4
‾
r
h
s
W
1
0
0
0
0
1
1
0
s
1
0
−
2
1
1
0
1
0
1
s
2
0
1
−
2
0
1
0
1
1
\begin{array}{c|clr} bv&W&x_1&x_2&s_1&s_2&\overline{x_3}&\overline{x_4}&rhs\\ \hline W&1&0&0&0&0&1&1&0\\ \hline s_1&0&-2&1&1&0&1&0&1\\ \hline s_2&0&1&-2&0&1&0&1&1 \end{array}
bvWs1s2W100x10−21x201−2s1010s2001x3110x4101rhs011
W
=
0
W=0
W=0,
x
3
‾
=
1
,
x
4
‾
=
1
\overline{x_3}=1,\overline{x_4}=1
x3=1,x4=1.This indicates that the original problem is infeasible.
Q 5
(a) Big M method
m
i
n
Z
=
m
a
x
(
−
Z
)
,
l
e
t
(
−
Z
)
=
Z
‾
minZ=max(-Z), let (-Z)=\overline{Z}
minZ=max(−Z),let(−Z)=Z
Maximization
Z
‾
\overline{Z}
Z
(0)
Z
‾
=
3
x
1
+
2
x
2
+
4
x
3
−
M
x
4
‾
\overline{Z}=3x_1+2x_2+4x_3-M\overline{x_4}
Z=3x1+2x2+4x3−Mx4
(1)
2
x
1
+
x
2
+
3
x
3
+
x
4
‾
=
60
2x_1+x_2+3x_3+\overline{x_4}=60
2x1+x2+3x3+x4=60
(2)
3
x
1
+
3
x
2
+
5
x
3
≥
120
3x_1+3x_2+5x_3\geq120
3x1+3x2+5x3≥120
x
j
≥
0
,
j
=
1
,
2
,
3
x_j\geq0,j=1,2,3
xj≥0,j=1,2,3
New eq.(0)=eq.(0)-M eq.(1）=
Z
‾
+
(
−
2
M
−
3
)
x
1
+
(
M
−
2
)
x
2
+
(
3
M
−
4
)
x
3
=
60
\overline{Z}+(-2M-3)x_1+(M-2)x_2+(3M-4)x_3=60
Z+(−2M−3)x1+(M−2)x2+(3M−4)x3=60
Iteration 0
b
v
Z
‾
x
1
x
2
x
3
x
4
‾
x
5
r
h
s
Z
‾
1
−
2
M
−
3
−
M
−
2
−
3
M
−
4
0
0
−
60
M
x
4
‾
0
2
1
3
1
0
60
x
5
0
3
3
5
0
1
120
\begin{array}{c|clr} bv&\overline{Z}&x_1&x_2&x_3&\overline{x_4}&x_5&rhs\\ \hline \overline{Z}&1&-2M-3&-M-2&-3M-4&0&0&-60M\\ \hline \overline{x_4}&0&2&1&3&1&0&60\\ \hline x_5&0&3&3&5&0&1&120\\ \end{array}
bvZx4x5Z100x1−2M−323x2−M−213x3−3M−435x4010x5001rhs−60M60120
Iteration 1
b
v
Z
‾
x
1
x
2
x
3
x
4
‾
x
5
r
h
s
Z
‾
1
−
1
/
3
−
2
/
3
0
M
+
4
/
3
0
80
x
3
0
2
/
3
1
/
3
1
1
/
3
0
20
x
5
0
−
1
/
3
4
/
3
0
−
5
/
3
1
20
\begin{array}{c|clr} bv&\overline{Z}&x_1&x_2&x_3&\overline{x_4}&x_5&rhs\\ \hline \overline{Z}&1&-1/3&-2/3&0&M+4/3&0&80\\ \hline x_3&0&2/3&1/3&1&1/3&0&20\\ \hline x_5&0&-1/3&4/3&0&-5/3&1&20\\ \end{array}
bvZx3x5Z100x1−1/32/3−1/3x2−2/31/34/3x3010x4M+4/31/3−5/3x5001rhs802020
Iteration 2
b
v
Z
‾
x
1
x
2
x
3
x
4
‾
x
5
r
h
s
Z
‾
1
−
1
/
2
0
0
M
+
19
/
18
1
/
6
80
+
10
/
3
x
3
0
3
/
4
0
1
17
/
36
−
1
/
12
20
−
5
/
3
x
2
0
−
1
/
4
1
0
−
5
/
12
1
/
4
5
\begin{array}{c|clr} bv&\overline{Z}&x_1&x_2&x_3&\overline{x_4}&x_5&rhs\\ \hline \overline{Z}&1&-1/2&0&0&M+19/18&1/6&80+10/3\\ \hline x_3&0&3/4&0&1&17/36&-1/12&20-5/3\\ \hline x_2&0&-1/4&1&0&-5/12&1/4&5\\ \end{array}
bvZx3x2Z100x1−1/23/4−1/4x2001x3010x4M+19/1817/36−5/12x51/6−1/121/4rhs80+10/320−5/35
Iteration 3
b
v
Z
‾
x
1
x
2
x
3
x
4
‾
x
5
r
h
s
Z
‾
1
0
0
2
/
3
37
/
27
+
M
1
/
9
860
/
9
x
1
0
1
0
4
/
3
17
/
27
−
1
/
9
220
/
9
x
2
0
0
1
1
/
3
−
7
/
27
2
/
9
85
/
9
\begin{array}{c|clr} bv&\overline{Z}&x_1&x_2&x_3&\overline{x_4}&x_5&rhs\\ \hline \overline{Z}&1&0&0&2/3&37/27+M&1/9&860/9\\ \hline x_1&0&1&0&4/3&17/27&-1/9&220/9\\ \hline x_2&0&0&1&1/3&-7/27&2/9&85/9\\ \end{array}
bvZx1x2Z100x1010x2001x32/34/31/3x437/27+M17/27−7/27x51/9−1/92/9rhs860/9220/985/9
Let
x
4
‾
=
0
\overline{x_4}=0
x4=0, we could get the optimal solution
(
x
1
,
x
2
,
x
3
)
=
(
220
/
9
,
85
/
9
,
5
/
9
)
(x_1,x_2,x_3)=(220/9,85/9,5/9)
(x1,x2,x3)=(220/9,85/9,5/9), and the
Z
‾
m
a
x
=
860
/
9
\overline{Z}_{max}=860/9
Zmax=860/9 , that is,
Z
m
i
n
=
−
860
/
9
Z_{min}=-860/9
Zmin=−860/9
(b)Two phase method
Phase 1:maximize W = − x 4 ‾ − x 6 ‾ W=-\overline{x_4}-\overline{x_6} W=−x4−x6 ( u n t i l x 4 ‾ = 0 , x 6 ‾ = 0 ) (until \overline{x_4}=0,\overline{x_6}=0) (untilx4=0,x6=0)
Phase 2:maximize Z ‾ = − 3 x 1 − 2 x 2 − 4 x 3 \overline{Z}=-3x_1-2x_2-4x_3 Z=−3x1−2x2−4x3 ( w i t h x 4 ‾ = 0 , x 6 ‾ = 0 ) (with \overline{x_4}=0,\overline{x_6}=0) (withx4=0,x6=0)
Phase 1 problem:
Maximize
W
=
−
x
4
‾
−
x
6
‾
W=-\overline{x_4}-\overline{x_6}
W=−x4−x6
subject to:
(1)
2
x
1
+
x
2
+
3
x
3
+
x
4
‾
=
60
2x_1+x_2+3x_3+\overline{x_4}=60
2x1+x2+3x3+x4=60
(2)
3
x
1
+
3
x
2
+
5
x
3
−
x
5
+
x
6
‾
=
120
3x_1+3x_2+5x_3-x_5+\overline{x_6}=120
3x1+3x2+5x3−x5+x6=120
New eq.(0)=
W
−
5
x
1
−
4
x
2
−
8
x
3
+
x
5
=
−
180
W-5x_1-4x_2-8x_3+x_5=-180
W−5x1−4x2−8x3+x5=−180
Iteration 0
b
v
W
x
1
x
2
x
3
x
4
‾
x
5
x
6
‾
r
h
s
W
1
−
5
−
4
−
8
0
1
0
−
180
x
4
‾
0
2
1
3
1
0
0
60
x
6
‾
0
3
3
5
0
−
1
1
120
\begin{array}{c|clr} bv&W&x_1&x_2&x_3&\overline{x_4}&x_5&\overline{x_6}&rhs\\ \hline W&1&-5&-4&-8&0&1&0&-180\\ \hline \overline{x_4}&0&2&1&3&1&0&0&60\\ \hline \overline{x_6} &0&3&3&5&0&-1&1&120\\ \end{array}
bvWx4x6W100x1−523x2−413x3−835x4010x510−1x6001rhs−18060120
Iteration 1
b
v
W
x
1
x
2
x
3
x
4
‾
x
5
x
6
‾
r
h
s
W
1
1
/
3
−
4
/
3
0
0
1
0
20
x
3
0
2
/
3
1
/
3
1
1
/
3
0
0
20
x
6
‾
0
−
1
/
3
4
/
3
0
−
5
/
3
−
1
1
20
\begin{array}{c|clr} bv&W&x_1&x_2&x_3&\overline{x_4}&x_5&\overline{x_6}&rhs\\ \hline W&1&1/3&-4/3&0&0&1&0&20\\ \hline x_3&0&2/3&1/3&1&1/3&0&0&20\\ \hline \overline{x_6}&0&-1/3&4/3&0&-5/3&-1&1&20\\ \end{array}
bvWx3x6W100x11/32/3−1/3x2−4/31/34/3x3010x401/3−5/3x510−1x6001rhs202020
Iteration 2
b
v
W
x
1
x
2
x
3
x
4
‾
x
5
x
6
‾
r
h
s
W
1
0
0
0
−
5
/
3
0
1
0
x
3
0
3
/
4
0
1
3
/
4
1
/
4
−
1
/
4
15
x
2
0
−
1
/
4
1
0
−
5
/
4
−
3
/
4
3
/
4
15
\begin{array}{c|clr} bv&W&x_1&x_2&x_3&\overline{x_4}&x_5&\overline{x_6}&rhs\\ \hline W&1&0&0&0&-5/3&0&1&0\\ \hline x_3&0&3/4&0&1&3/4&1/4&-1/4&15\\ \hline x_2&0&-1/4&1&0&-5/4&-3/4&3/4&15\\ \end{array}
bvWx3x2W100x103/4−1/4x2001x3010x4−5/33/4−5/4x501/4−3/4x61−1/43/4rhs01515
Phase 2 problem:
Maximize
Z
‾
=
−
3
x
1
−
2
x
2
−
4
x
3
\overline{Z}=-3x_1-2x_2-4x_3
Z=−3x1−2x2−4x3
subject to:
(1)
2
x
1
+
x
2
+
3
x
3
=
60
2x_1+x_2+3x_3=60
2x1+x2+3x3=60
(2)
3
x
1
+
3
x
2
+
5
x
3
−
x
5
=
120
3x_1+3x_2+5x_3-x_5=120
3x1+3x2+5x3−x5=120
(1) drop
x
4
‾
,
x
6
‾
\overline{x_4},\overline{x_6}
x4,x6
b
v
Z
‾
x
1
x
2
x
3
x
5
r
h
s
Z
‾
1
0
0
0
0
0
x
3
0
3
/
4
0
1
1
/
4
15
x
2
0
−
1
/
4
1
0
−
3
/
4
15
(1)
\begin{array}{c|clr} bv&\overline{Z}&x_1&x_2&x_3&x_5&rhs\\ \hline \overline{Z}&1&0&0&0&0&0\\ \hline x_3&0&3/4&0&1&1/4&15\\ \hline x_2&0&-1/4&1&0&-3/4&15\\ \end{array}\tag1
bvZx3x2Z100x103/4−1/4x2001x3010x501/4−3/4rhs01515(1)
(2) Substitute phase-2 objective function
b
v
Z
‾
x
1
x
2
x
3
x
5
r
h
s
Z
‾
1
3
2
4
0
0
x
3
0
3
/
4
0
1
1
/
4
15
x
2
0
−
1
/
4
1
0
−
3
/
4
15
(2)
\begin{array}{c|clr} bv&\overline{Z}&x_1&x_2&x_3&x_5&rhs\\ \hline \overline{Z}&1&3&2&4&0&0\\ \hline x_3&0&3/4&0&1&1/4&15\\ \hline x_2&0&-1/4&1&0&-3/4&15\\ \end{array}\tag2
bvZx3x2Z100x133/4−1/4x2201x3410x501/4−3/4rhs01515(2)
(3) Restore proper form
b
v
Z
‾
x
1
x
2
x
3
x
5
r
h
s
Z
‾
1
0
0
7
/
3
−
2
/
3
−
80
x
1
0
1
0
4
/
3
1
/
3
20
x
2
0
0
1
1
/
3
−
2
/
3
20
\begin{array}{c|clr} bv&\overline{Z}&x_1&x_2&x_3&x_5&rhs\\ \hline \overline{Z}&1&0&0&7/3&-2/3&-80\\ \hline x_1&0&1&0&4/3&1/3&20\\ \hline x_2&0&0&1&1/3&-2/3&20 \end{array}
bvZx1x2Z100x1010x2001x37/34/31/3x5−2/31/3−2/3rhs−802020
Q 6
(a) True. We introduce an artificial variable for the constraint with equality relationship to work as slack variable.
(b) False. If any artificial variables is
p
o
s
i
t
i
v
e
‾
\underline{positive}
positive (not equal to zreo) in optimal Big-M tableau, the original LP has no feasible solution.
© True.
Q 7
(a) The problem could be reformulated as follows:
m
a
x
Z
max Z
maxZ
s
.
t
.
s.t.
s.t.
Z
+
2
x
1
+
x
2
+
4
x
3
−
3
x
4
=
0
(
0
)
x
1
+
x
2
+
3
x
3
+
2
x
4
+
s
1
=
4
(
1
)
x
1
−
x
3
+
x
4
−
e
+
x
5
‾
=
−
1
(
2
)
2
x
1
+
x
2
+
s
2
=
2
(
3
)
x
1
+
2
x
2
+
x
3
+
2
x
4
+
x
6
‾
=
2
(
4
)
Z+2x_1+x_2+4x_3-3x_4=0\qquad(0)\\ x_1+x_2+3x_3+2x_4+s_1=4\qquad(1)\\ x_1-x_3+x_4-e+\overline{x_5}=-1\qquad(2)\\ 2x_1+x_2+s_2=2\qquad(3)\\ x_1+2x_2+x_3+2x_4+\overline{x_6}=2\qquad(4)
Z+2x1+x2+4x3−3x4=0(0)x1+x2+3x3+2x4+s1=4(1)x1−x3+x4−e+x5=−1(2)2x1+x2+s2=2(3)x1+2x2+x3+2x4+x6=2(4)
and
x
2
≥
0
,
x
3
≥
0
,
x
4
≥
0
,
x
5
‾
≥
0
,
x
6
‾
≥
0
,
e
≥
0
,
s
1
≥
0
,
s
2
≥
0
x_2\geq0,x_3\geq0,x_4\geq0,\overline{x_5}\geq0,\overline{x_6}\geq0,e\geq0,s_1\geq0,s_2\geq0
x2≥0,x3≥0,x4≥0,x5≥0,x6≥0,e≥0,s1≥0,s2≥0
(b)Big M method
Let
M
→
∞
M\rightarrow\infty
M→∞
Z
=
−
2
x
1
+
x
2
−
4
x
3
+
3
x
4
−
M
x
5
‾
−
M
x
6
‾
Z=-2x_1+x_2-4x_3+3x_4-M\overline{x_5}-M\overline{x_6}
Z=−2x1+x2−4x3+3x4−Mx5−Mx6
New eq.(0)=eq.(0)-Meq.(2)-Meq.(4)
Z
+
(
2
−
2
M
)
x
1
+
(
−
2
M
−
2
)
x
2
+
4
x
3
+
(
−
3
M
−
3
)
x
4
+
M
e
=
−
M
(
0
)
Z+(2-2M)x_1+(-2M-2)x_2+4x_3+(-3M-3)x_4+Me=-M\qquad(0)
Z+(2−2M)x1+(−2M−2)x2+4x3+(−3M−3)x4+Me=−M(0)
Iteration 0
b
v
Z
x
1
x
2
x
3
x
4
e
x
5
‾
x
6
‾
s
1
s
2
r
h
s
Z
1
2
−
2
M
−
2
M
−
2
4
−
3
M
−
3
M
0
0
0
0
−
M
s
1
0
1
1
3
2
0
0
0
1
0
4
x
5
‾
0
1
0
−
1
1
−
1
1
0
0
0
−
1
s
2
0
2
1
0
0
0
0
0
0
1
2
x
6
‾
0
1
2
1
2
0
0
1
0
0
2
\begin{array}{c|clr} bv&Z&x_1&x_2&x_3&x_4&e&\overline{x_5}&\overline{x_6}&s_1&s_2&rhs\\ \hline Z&1&2-2M&-2M-2&4&-3M-3&M&0&0&0&0&-M\\ \hline s_1&0&1&1&3&2&0&0&0&1&0&4\\ \hline \overline{x_5}&0&1&0&-1&1&-1&1&0&0&0&-1\\ \hline s_2&0&2&1&0&0&0&0&0&0&1&2\\ \hline\overline{x_6}&0&1&2&1&2&0&0&1&0&0&2\\ \end{array}
bvZs1x5s2x6Z10000x12−2M1121x2−2M−21012x343−101x4−3M−32102eM0−100x500100x600001s101000s200010rhs−M4−122
Iteration 1
b
v
Z
x
1
x
2
x
3
x
4
e
x
5
‾
x
6
‾
s
1
s
2
r
h
s
Z
1
(
7
−
M
)
/
2
M
+
1
(
3
M
+
11
)
/
2
0
M
0
(
3
M
+
3
)
/
2
0
0
2
M
+
3
s
1
0
0
−
1
2
0
0
0
−
1
1
0
2
x
5
‾
0
1
/
2
−
1
3
/
2
0
−
1
1
−
1
/
2
0
0
−
2
s
2
0
2
1
0
0
0
0
0
0
1
2
x
4
0
1
/
2
1
1
/
2
1
0
0
1
/
2
0
0
1
\begin{array}{c|clr} bv&Z&x_1&x_2&x_3&x_4&e&\overline{x_5}&\overline{x_6}&s_1&s_2&rhs\\ \hline Z&1&(7-M)/2&M+1&(3M+11)/2&0&M&0&(3M+3)/2&0&0&2M+3\\ \hline s_1&0&0&-1&2&0&0&0&-1&1&0&2\\ \hline \overline{x_5}&0&1/2&-1&3/2&0&-1&1&-1/2&0&0&-2\\ \hline s_2&0&2&1&0&0&0&0&0&0&1&2\\ \hline x_4&0&1/2&1&1/2&1&0&0&1/2&0&0&1 \end{array}
bvZs1x5s2x4Z10000x1(7−M)/201/221/2x2M+1−1−111x3(3M+11)/223/201/2x400001eM0−100x500100x6(3M+3)/2−1−1/201/2s101000s200010rhs2M+32−221
In the first simplex tableau, the entering variable is
x
1
x_1
x1,and the leaving variable is
s
2
s_2
s2.
The initial artificial solution is
(
x
1
,
x
2
,
x
3
,
x
4
)
=
(
0
,
0
,
0
,
1
)
(x_1,x_2,x_3,x_4)=(0,0,0,1)
(x1,x2,x3,x4)=(0,0,0,1)
©
Phase 1:maximize Z = − x 5 ‾ − x 6 ‾ Z=-\overline{x_5}-\overline{x_6} Z=−x5−x6
Phase 1 problem:
Max Z, s.t.:
x 1 + x 2 + 3 x 3 + 2 x 4 + s 1 = 4 ( 1 ) x 1 − x 3 + x 4 − e + x 5 ‾ = − 1 ( 2 ) 2 x 1 + x 2 + s 2 = 2 ( 3 ) x 1 + 2 x 2 + x 3 + 2 x 4 + x 6 ‾ = 2 ( 4 ) x_1+x_2+3x_3+2x_4+s_1=4\qquad(1)\\ x_1-x_3+x_4-e+\overline{x_5}=-1\qquad(2)\\ 2x_1+x_2+s_2=2\qquad(3)\\ x_1+2x_2+x_3+2x_4+\overline{x_6}=2\qquad(4) x1+x2+3x3+2x4+s1=4(1)x1−x3+x4−e+x5=−1(2)2x1+x2+s2=2(3)x1+2x2+x3+2x4+x6=2(4)
row(0)=eq.(0)-eq.(2)-eq.(4)
row(0):
Z
−
2
x
1
−
2
x
2
−
3
x
4
+
e
=
−
1
Z-2x_1-2x_2-3x_4+e=-1
Z−2x1−2x2−3x4+e=−1
460
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
