0% found this document useful (0 votes)
5 views28 pages

Big M Method in Linear Programming

Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
5 views28 pages

Big M Method in Linear Programming

Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

MATH 304 – MODULE 4

MR KENNETH N TERCERO
Introduction to the Big M Method

In this section, we will present a generalized version of the simplex method that
will solve both maximization and minimization problems with any combination
of ≤, ≥, = constraints

 In order to use the simplex method on problems with mixed constraints, we


turn to a device called an artificial variable.

 An artificial variable is a variable introduced into >= and = constraint to


ensure that we consider only basic feasible solutions, an artificial variable is
required to satisfy the nonnegative constraint.
Introduction to the Big M Method

 To prevent an artificial variable from becoming part of an optimal solution to


the original problem, a very large “penalty” is introduced into the objective
function. This penalty is created by 8 choosing a positive constant M so large
that the artificial variable is forced to be 0 in any final optimal solution of the
original problem.
Big M Method: Form the Modified
Problem
 If any problem constraints have negative constants on the right e, multiply
both sides by -1 to obtain a constraint with a nonnegative constant.
Remember to reverse the direction of the inequality if the constraint is an
inequality.
 Introduce a slack variable for each constraint of the form ≤.
 Introduce a surplus variable and an artificial variable in each ≥ constraint.
 Introduce an artificial variable in each = constraint.
 For each artificial variable a, add –Ma in case of maximizing objective
function and add +Ma in case of miniimizing objective function. Use the
same constant M for all artificial variables.
STEP BY STEP PROCESS

STEP 1: TRANSFORM THE LINEAR PROGRAMMING MODEL INTO EQUATION;


STEP 3: CHANGE THE M IN A1 AND A2 TO ZERO;
STEP 2: CONVERT THE SLE TO SIMPLEX TABLEAU;
STEP 4: CHECK THE OPTIMALITY;
STEP 5: DETERMINE THE PIVOT COLUMN;
STEP 6: DETERMINE THE PIVOT ROW (RATIO TEST);
STEP 7: PIVOTING
LINEAR PROGRAMMING MODEL
TAKE NOTE (INEQUALITIES):

PROBLEM SLE
MAXIMIZE 𝐏 = 𝒙 − 𝒚 + 𝟑𝒛 MAXIMAZATION + (-MA)
MINIMIZATION +(MA)
SUBJECT TO;

𝒙 + 𝒚 ≤ 𝟐𝟎 TAKE NOTE (CONSTRAINTS):


𝒙+𝒛=𝟓 INEQUALITIES SLE
𝒚 + 𝒛 ≥ 𝟏𝟎 ≤ +S
𝒙, 𝒚, 𝒛 ≥ 𝟎 = +A
>= -S AND +A
Introduction to the Big M Method
Example:

MAXIMIZE 𝐏 = 𝒙 − 𝒚 + 𝟑𝒛 + −𝑴𝑨𝟏 + (−𝑴𝑨𝟐 )

SUBJECT TO;

𝒙 + 𝒚 + 𝑺𝟏 = 𝟐𝟎
𝒙 + 𝒛 + 𝑨𝟏 = 𝟓
𝒚 + 𝒛 − 𝑺𝟐 + 𝑨𝟐 = 𝟏𝟎
𝒙, 𝒚, 𝒛, 𝑺𝟏 , 𝑺𝟐 , 𝑨𝟏 , 𝑨𝟐 ≥ 𝟎
STEP 1: TRANSFORM THE LINEAR PROGRAMMING MODEL INTO EQUATION

MAXIMIZE
−𝒙 + 𝒚 − 𝟑𝒛 + 𝑴𝑨𝟏 + 𝑴𝑨𝟐 + 𝑷 = 𝟎

SUBJECT TO;

𝒙 + 𝒚 + 𝑺𝟏 = 𝟐𝟎
𝒙 + 𝒛 + 𝑨𝟏 = 𝟓
𝒚 + 𝒛 − 𝑺𝟐 + 𝑨𝟐 = 𝟏𝟎
𝒙, 𝒚, 𝒛, 𝑺𝟏 , 𝑺𝟐 , 𝑨𝟏 , 𝑨𝟐 ≥ 𝟎
STEP 2: CONVERT THE SLE TO SIMPLEX TABLEAU
MAXIMIZE
−𝒙 + 𝒚 − 𝟑𝒛 + 𝑴𝑨𝟏 + 𝑴𝑨𝟐 + 𝑷 = 𝟎

SUBJECT TO;

𝒙 + 𝒚 + 𝑺𝟏 = 𝟐𝟎
𝒙 + 𝒛 + 𝑨𝟏 = 𝟓
𝒚 + 𝒛 − 𝑺𝟐 + 𝑨𝟐 = 𝟏𝟎
𝒙, 𝒚, 𝒛, 𝑺𝟏 , 𝑺𝟐 , 𝑨𝟏 , 𝑨𝟐 ≥ 𝟎

ROW x y z 𝑺𝟏 𝑺𝟐 𝑨𝟏 𝑨𝟐 P RHS
R1 1 1 0 1 0 0 0 0 20
R2 1 0 1 0 0 1 0 0 5
R3 0 1 1 0 -1 0 1 0 10
R4 -1 1 -3 0 0 M M 1 0
STEP 3: CHANGE THE M IN A1 AND A2 TO ZERO

ROW x y z 𝑺𝟏 𝑺𝟐 𝑨𝟏 𝑨𝟐 P RHS

R1 1 1 0 1 0 0 0 0 20

R2 1 0 1 0 0 1 0 0 5

R3 0 1 1 0 -1 0 1 0 10

R4 -1 1 -3 0 0 M M 1 0
ROW x y z 𝑺𝟏 𝑺𝟐 𝑨𝟏 𝑨𝟐 P RHS
R1 1 1 0 1 0 0 0 0 20
CHANGE M TO ZERO R2 1 0 1 0 0 1 0 0 5
R3 0 1 1 0 -1 0 1 0 10
𝐸𝑅𝑇𝐶 ∗ RPE + RTC → RTC
R4 -1 1 -3 0 0 M M 1 0
−𝑀 ∗ 𝑅2 + 𝑅4 → 𝑅4

ROW x y z 𝑺𝟏 𝑺𝟐 𝑨𝟏 𝑨𝟐 P RHS

R1 1 1 0 1 0 0 0 0 20

R2 1 0 1 0 0 1 0 0 5

R3 0 1 1 0 -1 0 1 0 10

R4 -M-1 1 -M-3 0 0 0 M 1 -5M


RO x y z 𝑺𝟏 𝑺𝟐 𝑨𝟏 𝑨𝟐 P RHS
W
R1 1 1 0 1 0 0 0 0 20
R2 1 0 1 0 0 1 0 0 5
𝐸𝑅𝑇𝐶 ∗ RPE + RTC → RTC
R3 0 1 1 0 -1 0 1 0 10
−𝑀 ∗ 𝑅3 + 𝑅4 → 𝑅4 R4 -M-1 1 -M-3 0 0 0 M 1 -5M

ROW x y z 𝑺𝟏 𝑺𝟐 𝑨𝟏 𝑨𝟐 P RHS

R1 1 1 0 1 0 0 0 0 20

R2 1 0 1 0 0 1 0 0 5

R3 0 1 1 0 -1 0 1 0 10

R4 -M-1 -M+1 -2M-3 0 M 0 0 1 -15M


The Big M Method
x y z s1 s2 A1 A2 P RHS
1 1 0 1 0 0 0 0 20
1 0 1 0 0 1 0 0 5
0 1 1 0 -1 0 1 0 10
-M-1 -M+1 -2M-3 0 M 0 0 1 -15M

Basic Variables: s1, A1, A2 and P

Non Basic Variables: x, y, z, s2


STEP 4: CHECK THE OPTIMALITY

ROW BV X Y z s1 s2 A1 A2 P RHS RATIO TEST

1 s1 1 1 0 1 0 0 0 0 20 20/0= UNDEFINED

2 A1 1 0 1 0 0 1 0 0 5 5/1 =5

3 A2 0 1 1 0 -1 0 1 0 10 10/1 = 10

4 P -M-1 -M+1 -2M-3 0 M 0 0 1 -15M

NOT OPTIMAL
ASSUME THAT M=10

X -> -M-1 = -10 -1 = -11


Y - > -M+1 = -10+1 = -9
Z -> -2(10) -3 = -23
STEP 5: DETERMINE THE PIVOT COLUMN

ROW BV X Y z s1 s2 A1 A2 P RHS

1 s1 1 1 0 1 0 0 0 0 20

2 A1 1 0 1 0 0 1 0 0 5

3 A2 0 1 1 0 -1 0 1 0 10

4 P -M-1 -M+1 -2M-3 0 M 0 0 1 -15M


STEP 6: DETERMINE THE PIVOT ROW
(RATIO TEST)
ROW BV X Y z s1 s2 A1 A2 P RHS RATIO TEST

1 s1 1 1 0 1 0 0 0 0 20 20/0= UNDEFINED

2 A1 1 0 1 0 0 1 0 0 5 5/1 =5

3 A2 0 1 1 0 -1 0 1 0 10 10/1 = 10

4 P -M-1 -M+1 -2M-3 0 M 0 0 1 -15M


STEP 7: PIVOTING
ROW BV X Y z s1 s2 A1 A2 P RHS RATIO TEST

1 s1 1 1 0 1 0 0 0 0 20 20/0= UNDEFINED

2 A1 1 0 1 0 0 1 0 0 5 5/1 =5

3 A2 0 1 1 0 -1 0 1 0 10 10/1 = 10

4 P -M-1 -M+1 -2M-3 0 M 0 0 1 -15M

Z = Entering Variable; A1 = Leaving Variable

ROW BV X Y z s1 s2 A1 A2 P RHS RATIO TEST

1 s1 1 1 0 1 0 0 0 0 20 20/0= UNDEFINED

2 z 1 0 1 0 0 1 0 0 5 5/1 =5

3 A2 0 1 1 0 -1 0 1 0 10 10/1 = 10

4 P -M-1 -M+1 -2M-3 0 M 0 0 1 -15M


ROW BV X Y z s1 s2 A1 A2 P RHS

R1 s1 1 1 0 1 0 0 0 0 20

𝐸𝑅𝑇𝐶 ∗ RPE + RTC → RTC R2 z 1 0 1 0 0 1 0 0 5

R3 A2 0 1 1 0 -1 0 1 0 10
−1 ∗ 𝑅2 + 𝑅3 → 𝑅3
R4 P -M-1 -M+1 -2M- 0 M 0 0 1 -15M
3

ROW BV X Y z s1 s2 A1 A2 P RHS

R1 s1 1 1 0 1 0 0 0 0 20

R2 z 1 0 1 0 0 1 0 0 5

R3 A2 -1 1 0 0 -1 -1 1 0 5

R4 P -M-1 -M+1 -2M-3 0 M 0 0 1 -15M


ROW BV X Y z s1 s2 A1 A2 P RHS

R1 s1 1 1 0 1 0 0 0 0 20

R2 z 1 0 1 0 0 1 0 0 5
𝐸𝑅𝑇𝐶 ∗ RPE + RTC → RTC
R3 A2 -1 1 0 0 -1 -1 1 0 5
(2𝑀 + 3) ∗ 𝑅2 + 𝑅4 → 𝑅4
R4 P -M-1 -M+1 -2M-3 0 M 0 0 1 -15M

ROW BV X Y z s1 s2 A1 A2 P RHS

R1 s1 1 1 0 1 0 0 0 0 20

R2 z 1 0 1 0 0 1 0 0 5

R3 A2 -1 1 0 0 -1 -1 1 0 5

R4 P M+2 -M+1 0 0 M 2M+3 0 1 -5M+15


STEP 4: CHECK THE OPTIMALITY
ROW BV X Y z s1 s2 A1 A2 P RHS

R1 s1 1 1 0 1 0 0 0 0 20

R2 z 1 0 1 0 0 1 0 0 5

R3 A2 -1 1 0 0 -1 -1 1 0 5

R4 P M+2 -M+1 0 0 M 2M+3 0 1 -5M+15

NOT OPTIMAL
STEP 5: DETERMINE THE PIVOT COLUMN
ROW BV X Y z s1 s2 A1 A2 P RHS

R1 s1 1 1 0 1 0 0 0 0 20

R2 z 1 0 1 0 0 1 0 0 5

R3 A2 -1 1 0 0 -1 -1 1 0 5

R4 P M+2 -M+1 0 0 M 2M+3 0 1 -5M+15

ASSUMING M = 10

X -> M+2 = 10+2 = 12


Y -> -M+1 = -10 +2 = -8
STEP 6: DETERMINE THE PIVOT ROW
(RATIO TEST)
ROW BV X Y z s1 s2 A1 A2 P RHS RATIO TEST

R1 s1 1 1 0 1 0 0 0 0 20 20/1 =20

R2 z 1 0 1 0 0 1 0 0 5 5/0 =
UNDEFINED
R3 A2 -1 1 0 0 -1 -1 1 0 5 5/1 =5

R4 P M+2 -M+1 0 0 M 2M+3 0 1 -5M+15


ROW BV X Y z s1 s2 A1 A2 P RHS RATIO TEST

R1 s1 1 1 0 1 0 0 0 0 20 20/1 =20
R2 z 1 0 1 0 0 1 0 0 5 5/0 =
UNDEFINED

R3 A2 -1 1 0 0 -1 -1 1 0 5 5/1 =5
R4 P M+2 -M+1 0 0 M 2M+3 0 1 -
5M+15

Y = ENTERING VARIABLE ; A2 = LEAVING VARIABLE


ROW BV X Y z s1 s2 A1 A2 P RHS RATIO
TEST
R1 s1 1 1 0 1 0 0 0 0 20 20/1 =20
R2 z 1 0 1 0 0 1 0 0 5 5/0 =
UNDEFIN
ED
R3 Y -1 1 0 0 -1 -1 1 0 5 5/1 =5
R4 P M+2 -M+1 0 0 M 2M+ 0 1 -
3 5M+15
STEP 7: PIVOTING
ROW BV X Y z s1 s2 A1 A2 P RHS

R1 s1 1 1 0 1 0 0 0 0 20

𝐸𝑅𝑇𝐶 ∗ RPE + RTC → RTC R2 z 1 0 1 0 0 1 0 0 5

(−1) ∗ 𝑅3 + 𝑅1 → 𝑅1 R3 y -1 1 0 0 -1 -1 1 0 5

R4 P M+2 -M+1 0 0 M 2M+3 0 1 -5M+15

ROW BV X Y z s1 s2 A1 A2 P RHS

R1 s1 2 0 0 1 1 1 -1 0 15

R2 z 1 0 1 0 0 1 0 0 5

R3 y -1 1 0 0 -1 -1 1 0 5

R4 P M+2 -M+1 0 0 M 2M+3 0 1 -5M+15


STEP 7: PIVOTING
ROW BV X Y z s1 s2 A1 A2 P RHS

R1 s1 2 0 0 1 1 1 -1 0 15
𝐸𝑅𝑇𝐶 ∗ RPE + RTC → RTC
R2 z 1 0 1 0 0 1 0 0 5
(𝑀 − 1) ∗ 𝑅3 + 𝑅4 → 𝑅4
R3 y -1 1 0 0 -1 -1 1 0 5

R4 P M+2 -M+1 0 0 M 2M+3 0 1 -5M+15

ROW BV X Y z s1 s2 A1 A2 P RHS

R1 s1 2 0 0 1 1 1 -1 0 15

R2 z 1 0 1 0 0 1 0 0 5

R3 y -1 1 0 0 -1 -1 1 0 5

R4 P 3 0 0 0 1 M+4 M-1 1 10
STEP 4: CHECK THE OPTIMALITY

ROW BV X Y z s1 s2 A1 A2 P RHS

R1 s1 2 0 0 1 1 1 -1 0 15

R2 z 1 0 1 0 0 1 0 0 5

R3 y -1 1 0 0 -1 -1 1 0 5

R4 P 3 0 0 0 1 M+4 M-1 1 10

OPTIMAL
ROW BV X Y z s1 s2 A1 A2 P RHS

R1 s1 2 0 0 1 1 1 -1 0 15

R2 z 1 0 1 0 0 1 0 0 5

R3 y -1 1 0 0 -1 -1 1 0 5

R4 P 3 0 0 0 1 M+4 M-1 1 10

BASIC VARIABLE NON BASIC VARIABLE X=0


S1 =15 X=0 Y=5
Z=5 S2 = 0 Z=5
Y=5 A1=0 P = 10
P =10 A2=0
SEATWORK #3 : BY PAIR

You might also like