DS321-Linear & Integer Programming -2021-2022- Prof. Dr. Tarek H. M.

Abou-El-Enien

Notes on Lesson (5)

The Two-Phase Method

Importance of the Two-Phase Method:

In the M-method, the use of the penalty M, which by definition must be
large relative to the actual objective coefficients of the model, can result in
roundoff error that may impair the accuracy of the simplex calculations. The
two-phase method alleviates this difficulty by eliminating the constant M
altogether.

General idea of the Two-Phase Method:

The method solves the LP in two phases:
1) Phase I attempts to find a starting basic feasible solution, and if one is
found,
2) Phase II is invoked to solve the original problem.

Summary of the steps of the Two-Phase Method:

Phase (I):
1) Put the problem in equation form, and add the necessary artificial
variables to the constraints (exactly as in the M method) to secure a
starting basic solution.
2) Find a basic solution of the resulting equations that, regardless of
whether the LP is maximization or minimization, always minimize the
sum of artificial variables.
3) If the minimum value of the sum is positive, the LP problem has no
feasible solution, which ends the process (recall that a positive artificial

Page 1 of 5
 DS321-Linear & Integer Programming -2021-2022- Prof. Dr. Tarek H. M. Abou-El-Enien

signifies that an original constraint is not satisfied). Otherwise, proceed

to Phase II.
Phase (II):
4) Use the feasible solution from Phase I as a starting basic feasible
solution for the original problem.

Example:
Use the two-phase Method to find the optimal solution for the following
linear programming (LP) problem (If exist):

Minimize F= 4x1 + x2
subject to
3x1 + x2 = 3
4 x1 + 3x2 ≤ 6
x1 + 2x2 ≤ 4
x1 , x2 ≥ 0

Solution:
Minimize F= 4x1 + x2 + 0s1 + 0s2
subject to
3x1 + x2 = 3
4 x1 + 3x2 – s1 = 6
x1 + 2x2 + s2 = 4
x1 , x2 , s1 , s2 ≥ 0

Page 2 of 5
 DS321-Linear & Integer Programming -2021-2022- Prof. Dr. Tarek H. M. Abou-El-Enien

Minimize F= 4x1 + x2 + 0s1 + 0s2 + MR1 + MR2

subject to
3x1 + x2 + R1 =3
4 x1 + 3x2 – s1 + R2 = 6
x1 + 2x2 + s2 = 4
x1 , x2 , s1 , R1 , R2 , s2 ≥ 0 ,

Phase (I):
Minimize r = R1 +R2
subject to
3x1 + x2 + R1 =3
4 x1 + 3x2 – s1 + R2 = 6
x1 + 2x2 + s2 = 4
x1 , x2 , s1 , R1 , R2 , s2 ≥ 0 ,

n = 6, m = 3 , n – m =6 – 3 =3
x1 , x2 , s1 = 0 ⟹ R1 = 3 , R2 = 6 , s2 = 4 ,

Basic x1 x2 s1 R1 R2 s2 Solution
r 0 0 0 -1 -1 0 0
R1 3 1 0 1 0 0 3
R2 4 3 -1 0 1 0 6
s2 1 2 0 0 0 1 4

R1=3 - 3 x1 - x2
R2 = 6 - 4 x1 - 3 x2 + s1
r = R1 +R2 = (3-3 x1 - x2) + (6 - 4 x1 - 3 x2 + s2) = -7 x1 - 4 x2 + s2 + 9
Page 3 of 5
 DS321-Linear & Integer Programming -2021-2022- Prof. Dr. Tarek H. M. Abou-El-Enien

Basic x1 x2 s1 R1 R2 s2 Solution
r 7 4 -1 0 0 0 9
R1 3 1 0 1 0 0 3
R2 4 3 -1 0 1 0 6
s2 1 2 0 0 0 1 4

Basic x1 x2 s1 R1 R2 s2 Solution
r 0 1.67 -1 -2.33 0 0 2
x1 1 0.33 0 0.33 0 0 1
R2 0 1.67 -1 -1.33 1 0 2
s2 0 1.67 0 -0.33 1 0 3

Basic x1 x2 s1 R1 R2 s2 Solution
r 0 0 0 -1 -1 0 0
x1 1 0 0.2 -0.2 -0.2 0 0.6
x2 0 1 -0.6 -0.8 0.6 0 1.2
s2 0 0 1 1 -1 1 1

Phase (II):
Minimize F= 4x1 + x2
subject to
x1 + 0.2 s1 = 0.6
x2 – 0.6 s1 = 1.2
s1 + s2 = 1
x1 , x2 , s1 , s2 ≥ 0

Page 4 of 5
 DS321-Linear & Integer Programming -2021-2022- Prof. Dr. Tarek H. M. Abou-El-Enien

Basic x1 x2 s1 s2 Solution
F -4 -1 0 0 0
x1 1 0 0.2 0 0.6
x2 0 1 -0.6 0 1.2
s2 0 0 1 1 1

F = 4x1 + x2 = 4(0.6 – 0.2 s1) + (1.2 + 0.6 s1) = -0.2 s1 + 3.6

Basic x1 x2 s1 s2 Solution
F 0 0 0.2 0 3.6
x1 1 0 0.2 0 0.6
x2 0 1 -0.6 0 1.2
s2 0 0 1 1 1

Basic x1 x2 s1 s2 Solution
F 0 0 0 -0.2 3.4
x1 1 0 0 -0.2 0.4
x2 0 1 0 0.6 1.8
s1 0 0 1 1 1
x1 = 0.4 , x2=1.8 , F = 3.4

Page 5 of 5