Big-M Method

Consider the given linear programming (LP) problem.

Maximise P=x +5 y
subject to 3 x+ 4 y ≤ 6

x +3 y ≥ 2x , y ≥ 0

Introducing slack variable, S1, surplus variables, S2and artificial variable, A1:
Maximize P=x +5 y +0 S1 +0 S 2 – M A 1 3 x+ 4 y + S1 =6

x +3 y−S2 + A1=2x , y , S 1 , S2 , A 1 ≥ 0

Cj 1 5 0 0 −M
CB Basis x y S1 S2 A1 Solution Replacement
variables, B values, b b
ratio,
y
0 S1 3 4 1 0 0 6 6
=1.5
4
−M A1 1 3 0 −1 1 2 2
=0.6667
3
Pj −M −3 M 0 M −M −2 M
C j −Pi 1+ M 5+3 M 0 −M 0

The pivot element is 3. Replace the leaving basis variable, A1 with the entering variable, y .
Cj 1 5 0 0 −M
CB Basis x y S1 S2 A1 Solution
variables, B values, b
0 S1 5 0 1 4 −4 10 R1 (new) → R 1(old )−4 R2 (new)
3 3 3 3
5 y 1 1 0 −1 1 2
1
3 3 3 3 R2 (new) → R 2 (old)
3
Pj 5 5 0 −5 5 10
3 3 3 3
C j −Pi −2 0 0 5 5
−M −
3 3 3

Cj 1 5 0 0 −M
CB Basis x y S1 S2 A1 Solution Replacement
variables, B values, b b
ratio,
y
 0 S1 5 0 1 4 −4 10 10 4
÷ =2.5
3 3 3 3 3 3
5 y 1 1 0 −1 1 2 2 −1
÷( )=−2
3 3 3 3 3 3
Pj 5 5 0 −5 5 10
3 3 3 3
C j −Pi −2 0 0 5 5
−M −
3 3 3

4
The pivot element is . Replace the leaving basis variable, S1 with the entering variable, S2 .
3
Cj 1 5 0 0 −M
CB Basis x y S1 S2 A1 Solution
variables, values,
B b 3
0 S2 1.25 0 0.75 1 −2.5 2.5 R1 (new)→ R ( old)
4 2
5 y 0.75 1 0.25 0 −0.5 1.5 1
R2 (new) → R 2 (old )+ R (new)
Pj 3.75 5 1.25 0 −2.5 7.5 3 1
C j −Pi −2.75 0 −1.25 0 −M −2.5

Since all C j−Pi ≤ 0, optimal solution is arrived with value of variables as x=0 and y=1.5
and the maximum of P is7.5.

The problem is converted to canonical form by adding slack, surplus and artificial variables
as appropriate:
1. As the constraint-1 is of type '≤' we should add slack variable S1
2. As the constraint-2 is of type '≥' we should subtract surplus variable S2 and add artificial

variable A1
Cj contribution of

Cb coefficient

Algorithm Example
Step 1 Formulate the mathematical model of Maximize: z=2 x+ 5 y
the given linear programming (LP) Subject to 2 x+ y ≤ 5
problem. x +2 y ≥ 4
x , y ≥0
Step 2 Convert LP problem into canonical form Adding slack variable, S1, surplus
by adding slack, surplus and artificial variables, S2and artificial variable, A1:
variables for each constraint.
2 x+ y + S 1=5
In the objective function, assign a “0”
x +2 y −S 2 + A 1=4
coefficient to slack and surplus
z=2 x+ 5 y + 0 S1 +0 S2 – M A 1
variables. Then, add −M (for
Maximization case) or +M (for
Minimization case) to artificial variables
where M is a very large positive value.
Step 3 Construct a simplex table.
Find the initial basic feasible solution by
setting zero value to the decision
variables.
Step 4 Select the pivot column.
a) Calculate the values ofc j−z j in the
last row of simplex table.
b) For maximization case:
- If all c j−z j <0 , then the current
basic feasible solution is the optimal
solution.
- In c j−z j >0, then a column with the

largest value of c j−z j is the pivot

column.
For minimization case:
- If all c j−z j >0 , then the current
basic feasible solution is the optimal
solution.
- In c j−z j <0, then a column with the

smallest value of c j−z j is the pivot

column.
Step 5 Select the pivot row.
a) Find the ratio by dividing the values
of b column by the positive value of
pivot column.
b) Find the minimum ratio and this row
is the pivot row and corresponding
 variable would be the leaving basis
variable.
Step 6 Select the pivot element.
The intersection element of pivot row
and pivot column is the pivot element.
Step 7 Determine the new solution. eg. R1(new)=R1(old) ÷3
Perform row operations to make the eg. R2(new)=R2(old) -2 R1(new)

pivot element equal to 1 and the

remaining elements in the pivot column
equal to zero.
Step 8 Go to step 3 and repeat the procedure
until all the values of c j−z j ≤0  (for

Maximization case) or c j−z j ≥0  (for

Minimization case).