9/10/2018

Simplex Method
- translates the geometric definition of the
extreme point into an algebraic definition

Linear Programming
Simplex Method (Part 1)

The Reddy Mikks Company owns a small paint factory that Graphical Solution
produces both interior and exterior house paints for wholesale
distribution. Two basic raw materials, A and B, are used to Let x = tons produced daily of exterior paint
manufacture the paints. The maximum availability of A is 6 tons a y = tons produced daily of interior paint
day; that of B is 8 tons a day. For each ton of paint, Exterior paints Objective function:
require 1 ton of A and 2 tons of B, while Interior paints require 2
maximize z = 3000x + 2000y
tons of A and a ton of B.
subject to
A market survey has established that the daily demand for
x + 2y ≤ 6
interior paint cannot exceed that of exterior paint by more than 1
ton. The survey also shows that the maximum demand for interior 2x + y ≤ 8
paint is limited to 2 tons daily. The wholesale price per ton is $3000 -x+y≤1
for exterior paint and $2000 for interior paint. How much interior y≤2
and exterior paints should the company produce daily to maximize x, y ≥ 0
gross income
10
Therefore, Reddy Mikks Company must produce or 3.33 tons of exterior paint
3
4
and or 1.33 tons of interior paint.
3

Standard LP Form Constraints

LP models may include constraints of all types (≤, ≥, =).
A constraint of the type ≤ can be converted to an equation
Moreover, the variables may be nonnegative or unrestricted in sign. by adding a slack variable or the unused amount of resource. While
To develop a general solution method, the LP problem must be put a constraint of type ≥ can be converted to an equation by
in a common format, which we will call the standard form. The subtracting a surplus variable, which represents the excess amount
properties of this form are as follows: by which the minimum specification is satisfied.
1. All the constraints are equations. On the other hand, an unrestricted variable may be
2. All the variables are nonnegative. expressed in terms of two nonnegative variables using the
substitution x = x’ – x”, where x’, x” ≥ 0. The substitution must be
3. The objective function may be maximization or effected throughout all the constraints and in the objective
minimization. function.

1
 9/10/2018

Constraints
Form Action Let x = tons produced daily of exterior paint
y = tons produced daily of interior paint
≤ add slack variable
Objective function: Objective function:
≥ subtract surplus variable
maximize z = 3000x + 2000y maximize z = 3000x + 2000y + 0𝑠1 + 0𝑠2 + 0𝑠3 + 0𝑠4
unrestricted substitute x = x’ – x”, where x’, x” ≥ 0 subject to
subject to
effected throughout all the constraints and in the objective function. x + 2y + 𝑠1= 6
x + 2y ≤ 6
2x + y ≤ 8 2x + y + 𝑠2= 8
Write the LP model in standard form:
-x+y≤1 - x + y + 𝑠3= 1
y + 𝑠4 = 2
y≤2
minimize z = 2x + 3y minimize z = 2x’ – 2x” + 3y + 0𝑠1 +0𝑠2 x, y, 𝑠1, 𝑠2, 𝑠3, 𝑠4 ≥ 0
x, y ≥ 0
subject to subject to
x + y = 10 x’ – x” + y = 10
-2x + 3y ≤ - 5 -2x + 3y + 𝑠1 = - 5 -2x’ + 2x” + 3y + 𝑠1 = - 5
7x – 4y ≥ 6 7x – 4y – 𝑠2 = 6 7x’ – 7x” – 4y – 𝑠2 = 6
x unrestricted x‘, x”, y, 𝑠1, 𝑠2 ≥ 0
y≥0

Let x = tons produced daily of exterior paint y = tons produced daily of interior paint current nonbasic variables
Objective function: Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution
maximize z = 3000x + 2000y + 0𝑠1 + 0𝑠2 + 0𝑠3 + 0𝑠4 z 1 -3000 -2000 0 0 0 0 0
maximize z - 3000x - 2000y + 0𝑠1+ 0𝑠2+ 0𝑠3+ 0𝑠4 = 0 𝑠1 0 1 2 1 0 0 0 6
subject to current 𝑠2 0 2 1 0 1 0 0 8
x + 2y + 𝑠1= 6 - x + y + 𝑠3= 1 x, y, 𝑠1, 𝑠2, 𝑠3, 𝑠4 ≥ 0 basic
variables 𝑠3 0 -1 1 0 0 1 0 1
2x + y + 𝑠2= 8 y + 𝑠4 = 2 𝑠4 0 0 1 0 0 0 1 2
Step 0: Using the standard form (with all nonnegative right-hand side), transform the Step 1: Select an entering variable among the nonbasic variables.
equations into a tableau.
Since the problem involves maximization, select the most negative coefficient in
Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution the z-equation as the entering variable. In this case, x.
z 1 -3000 -2000 0 0 0 0 0
𝑠1 0 1 2 1 0 0 0 6
𝑠2 0 2 1 0 1 0 0 8
𝑠3 0 -1 1 0 0 1 0 1
𝑠4 0 0 1 0 0 0 1 2

Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution
z 1 -3000 -2000 0 0 0 0 0
𝑠1 0 1 2 1 0 0 0 6
𝑠2 0 2 1 0 1 0 0 8
Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution Ratio
𝑠3 0 -1 1 0 0 1 0 1
z 1 -3000 -2000 0 0 0 0 0
𝑠4 0 0 1 0 0 0 1 2
𝑠1 0 1 2 1 0 0 0 6 6/1 = 6
𝑠2 0 2 1 0 1 0 0 8 8/2 = 4 Step 3: Determine the values of the new basic variables by making the entering variable
basic and the leaving variable nonbasic.
𝑠3 0 -1 1 0 0 1 0 1 -
Pivot Equation: new pivot equation = old pivot equation ÷ pivot element
𝑠4 0 0 1 0 0 0 1 2 -
Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution
Step 2: Select a leaving variable among the basic variables.
z
Find the ratio between the basic variables’ solutions and their corresponding
𝑠1
coefficient under the entering variable.
x 0 1 1/2 0 1/2 0 0 4
𝑠3 has a negative ratio – meaning that the third constraint does not limit x in the
positive direction. 𝑠4 does not intersect with x at all because its ratio is undefined. 𝑠3
Since 𝑠2 has the least positive ratio, it is the leaving variable. Additionally, the 𝑠4
intersection value (2) is the pivot element.

2
 9/10/2018

Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution

z 1 -3000 -2000 0 0 0 0 0 z 1 -3000 -2000 0 0 0 0 0
𝑠1 0 1 2 1 0 0 0 6 𝑠1 0 1 2 1 0 0 0 6
𝑠2 0 2 1 0 1 0 0 8 𝑠2 0 2 1 0 1 0 0 8
𝑠3 0 -1 1 0 0 1 0 1 𝑠3 0 -1 1 0 0 1 0 1
𝑠4 0 0 1 0 0 0 1 2 𝑠4 0 0 1 0 0 0 1 2

Step 3: Determine the values of the new basic variables by making the entering variable All other equations, including z:
basic and the leaving variable nonbasic. new equation = old equation – (its entering column coefficient)(new pivot equation)
All other equations, including z: Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution
new equation = old equation – (its entering column coefficient)(new pivot equation) z
𝑠1
Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution
x 0 1 1/2 0 1/2 0 0 4
z
𝑠3
𝑠1
𝑠4
x 0 1 1/2 0 1/2 0 0 4
𝑠3 For z:
1 – (-3000)(0) = 1
𝑠4

Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution

z 1 -3000 -2000 0 0 0 0 0 z 1 -3000 -2000 0 0 0 0 0
𝑠1 0 1 2 1 0 0 0 6 𝑠1 0 1 2 1 0 0 0 6
𝑠2 0 2 1 0 1 0 0 8 𝑠2 0 2 1 0 1 0 0 8
𝑠3 0 -1 1 0 0 1 0 1 𝑠3 0 -1 1 0 0 1 0 1
𝑠4 0 0 1 0 0 0 1 2 𝑠4 0 0 1 0 0 0 1 2
All other equations, including z: All other equations, including z:
new equation = old equation – (its entering column coefficient)(new pivot equation) new equation = old equation – (its entering column coefficient)(new pivot equation)
Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution
z z 1 0 -500 0 1500 0 0 12000 Profit
𝑠1 𝑠1
x 0 1 1/2 0 1/2 0 0 4 x 0 1 1/2 0 1/2 0 0 4
𝑠3 𝑠3
𝑠4 𝑠4
For z (not required to be shown): For z (not required to be shown):
1 – (-3000)(0) = 1 0 – (-3000)(0) = 0 0 – (-3000)(0) = 0 1 – (-3000)(0) = 1 0 – (-3000)(0) = 0 0 – (-3000)(0) = 0
-3000 –(-3000)(1) = 0 0 – (-3000)(1/2) = 1500 0 – (-3000)(4) = 12000 -3000 –(-3000)(1) = 0 0 – (-3000)(1/2) = 1500 0 – (-3000)(4) = 12000
-2000 – (-3000)(1/2) = - 500 0 – (-3000)(0) = 0 -2000 – (-3000)(1/2) = - 500 0 – (-3000)(0) = 0

Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution

z 1 -3000 -2000 0 0 0 0 0 z 1 -3000 -2000 0 0 0 0 0
𝑠1 0 1 2 1 0 0 0 6 𝑠1 0 1 2 1 0 0 0 6
𝑠2 0 2 1 0 1 0 0 8 𝑠2 0 2 1 0 1 0 0 8
𝑠3 0 -1 1 0 0 1 0 1 𝑠3 0 -1 1 0 0 1 0 1
𝑠4 0 0 1 0 0 0 1 2 𝑠4 0 0 1 0 0 0 1 2
All other equations, including z: All other equations, including z:
new equation = old equation – (its entering column coefficient)(new pivot equation) new equation = old equation – (its entering column coefficient)(new pivot equation)
Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution
z 1 0 -500 0 1500 0 0 12000 Profit z 1 0 -500 0 1500 0 0 12000
𝑠1 𝑠1
x 0 1 1/2 0 1/2 0 0 4 x 0 1 1/2 0 1/2 0 0 4
𝑠3 𝑠3
𝑠4 𝑠4
For z (alternative procedure) For 𝑠1 (not required to be shown):
old equation: (1 -3000 -2000 0 0 0 0 0) 0 – (1)(0) = 0 1 – (1)(0) = 0 0 – (1)(0) = 0
-(coefficient)(new pivot equation): -(-3000)(0 1 ½ 0 ½ 0 0 4) = (0 3000 1500 0 1500 0 0 12000) 1 –(1)(1) = 0 0 – (1)(1/2) = -1/2 6 – (1)(4) = 2
new z equation (add): (1 0 -500 0 1500 0 0 12000)
2 – (1)(1/2) = 3/2 0 – (1)(0) = 0

3
 9/10/2018

Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution

z 1 -3000 -2000 0 0 0 0 0 z 1 -3000 -2000 0 0 0 0 0
𝑠1 0 1 2 1 0 0 0 6 𝑠1 0 1 2 1 0 0 0 6
𝑠2 0 2 1 0 1 0 0 8 𝑠2 0 2 1 0 1 0 0 8
𝑠3 0 -1 1 0 0 1 0 1 𝑠3 0 -1 1 0 0 1 0 1
𝑠4 0 0 1 0 0 0 1 2 𝑠4 0 0 1 0 0 0 1 2
All other equations, including z: All other equations, including z:
new equation = old equation – (its entering column coefficient)(new pivot equation) new equation = old equation – (its entering column coefficient)(new pivot equation)
Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution
z 1 0 -500 0 1500 0 0 12000 z 1 0 -500 0 1500 0 0 12000
𝑠1 0 0 3/2 1 -1/2 0 0 2 𝑠1 0 0 3/2 1 -1/2 0 0 2
x 0 1 1/2 0 1/2 0 0 4 x 0 1 1/2 0 1/2 0 0 4
𝑠3 𝑠3
𝑠4 𝑠4
For 𝑠1 (not required to be shown): For 𝑠3 (not required to be shown):
0 – (1)(0) = 0 1 – (1)(0) = 1 0 – (1)(0) = 0 0 – (-1)(0) = 0 0 – (-1)(0) = 0 0 – (-1)(0) = 0
1 –(1)(1) = 0 0 – (1)(1/2) = -1/2 6 – (1)(4) = 2 -1 – (-1)(1) = 0 0 – (-1)(1/2) = 1/2 1 – (-1)(4) = 5
2 – (1)(1/2) = 3/2 0 – (1)(0) = 0 1 – (-1)(1/2) = 3/2 1 – (-1)(0) = 1

Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution

z 1 -3000 -2000 0 0 0 0 0 z 1 -3000 -2000 0 0 0 0 0
𝑠1 0 1 2 1 0 0 0 6 𝑠1 0 1 2 1 0 0 0 6
𝑠2 0 2 1 0 1 0 0 8 𝑠2 0 2 1 0 1 0 0 8
𝑠3 0 -1 1 0 0 1 0 1 𝑠3 0 -1 1 0 0 1 0 1
𝑠4 0 0 1 0 0 0 1 2 𝑠4 0 0 1 0 0 0 1 2
All other equations, including z: All other equations, including z:
new equation = old equation – (its entering column coefficient)(new pivot equation) new equation = old equation – (its entering column coefficient)(new pivot equation)
Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution
z 1 0 -500 0 1500 0 0 12000 z 1 0 -500 0 1500 0 0 12000
𝑠1 0 0 3/2 1 -1/2 0 0 2 𝑠1 0 0 3/2 1 -1/2 0 0 2
x 0 1 1/2 0 1/2 0 0 4 x 0 1 1/2 0 1/2 0 0 4
𝑠3 0 0 3/2 0 1/2 1 0 5 𝑠3 0 0 3/2 0 1/2 1 0 5
𝑠4 𝑠4 0 0 1 0 0 0 1 2
For 𝑠3 (not required to be shown): For 𝑠4 , since the entering column coefficient is 0, the same values will be used.
0 – (-1)(0) = 0 0 – (-1)(0) = 0 0 – (-1)(0) = 0
-1 – (-1)(1) = 0 0 – (-1)(1/2) = 1/2 1 – (-1)(4) = 5
1 – (-1)(1/2) = 3/2 1 – (-1)(0) = 1

Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution

z 1 0 -500 0 1500 0 0 12000 z 1 0 -500 0 1500 0 0 12000
𝑠1 0 0 3/2 1 -1/2 0 0 2 𝑠1 0 0 3/2 1 -1/2 0 0 2
x 0 1 1/2 0 1/2 0 0 4 x 0 1 1/2 0 1/2 0 0 4
𝑠3 0 0 3/2 0 1/2 1 0 5 𝑠3 0 0 3/2 0 1/2 1 0 5
𝑠4 0 0 1 0 0 0 1 2 𝑠4 0 0 1 0 0 0 1 2
Step 4: Repeat steps 1 to 3 until all the coefficients in the z-equation are nonnegative.
Solve for the new z – equation.
Entering variable: y (most negative in the z-equation, compared to 𝑠2 )
Ratio: 𝑠1 = 2/(3/2) = 4/3 x = 4/(1/2) = 8 Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution
𝑠3 = 5/(3/2) = 10/3 𝑠4 = 2/1 = 2 z 1 0 0 333.33 1333.33 0 0 12666.67
𝑠1 is the leaving variable (least positive ratio)
y 0 0 1 2/3 -1/3 0 0 4/3
Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution x
z 𝑠3
y 0 0 1 2/3 -1/3 0 0 4/3 𝑠4
x
𝑠3
𝑠4

4
 9/10/2018

Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution

z 1 0 -500 0 1500 0 0 12000 z 1 0 -500 0 1500 0 0 12000
𝑠1 0 0 3/2 1 -1/2 0 0 2 𝑠1 0 0 3/2 1 -1/2 0 0 2
x 0 1 1/2 0 1/2 0 0 4 x 0 1 1/2 0 1/2 0 0 4
𝑠3 0 0 3/2 0 1/2 1 0 5 𝑠3 0 0 3/2 0 1/2 1 0 5
𝑠4 0 0 1 0 0 0 1 2 𝑠4 0 0 1 0 0 0 1 2

Solve for the new x – equation. Solve for the new 𝑠3 – equation.

Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution

z 1 0 0 333.33 1333.33 0 0 12666.67 z 1 0 0 333.33 1333.33 0 0 12666.67
y 0 0 1 2/3 -1/3 0 0 4/3 y 0 0 1 2/3 -1/3 0 0 4/3
x 0 1 0 - 1/3 2/3 0 0 10/3 x 0 1 0 - 1/3 2/3 0 0 10/3
𝑠3 𝑠3 0 0 0 -1 1 1 0 3
𝑠4 𝑠4

Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution
z 1 0 -500 0 1500 0 0 12000
𝑠1 0 0 3/2 1 -1/2 0 0 2
x 0 1 1/2 0 1/2 0 0 4
𝑠3 0 0 3/2 0 1/2 1 0 5 Seatwork
𝑠4 0 0 1 0 0 0 1 2
Romans Furniture makes two products: tables and chairs,
Solve for the new 𝑠4 – equation. which must be processed through assembly and finishing
departments. Assembly department is available for 60
Basic z x y 𝑠1 𝑠2 𝑠3 𝑠4 Solution hours in every production period, which the finishing
z 1 0 0 333.33 1333.33 0 0 12666.67 department is available for 48 hours of work.
y 0 0 1 2/3 -1/3 0 0 4/3
Manufacturing one table requires 4 hours in the assembly
x 0 1 0 - 1/3 2/3 0 0 10/3
and 2 hours in the finishing. Each chair requires 2 hours in
the assembly and 4 hours in the finishing. One table
𝑠3 0 0 0 -1 1 1 0 3 contributes Php 1800 to profit, while a chair contributes
𝑠4 0 0 0 -2/3 1/3 0 1 2/3 Php 1000. Determine the number of tables and chairs to
10 make per production period in order to maximize the profit.
Therefore, Reddy Mikks Company must produce 3
or 3.33 tons of exterior paint (Ans. 12 tables & 6 chairs)
4
and 3 or 1.33 tons of interior paint. The profit is maximized to $ 12,666.67.