OFFICIAL (CLOSED) \ NON-SENSITIVE

Lesson 04
Product Mix and Distribution Planning
E210 – Operations Planning
 OFFICIAL (CLOSED) \ NON-SENSITIVE

E210 Operations Planning Topic Tree

2
2
 OFFICIAL (CLOSED) \ NON-SENSITIVE

IS Lecture (1)
LP Feasibility and Optimality
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Scenario for Lesson 03 – 05:

Product Mix and Distribution Planning
• PrettyFit is an apparel manufacturer specialized in
producing high quality customized dresses for various
functions. Based in South Korea, PrettyFit has a number of
manufacturing plants in Asia.
• The main resources consumed to produce the dresses are
cloth, machine time and labour time. With limited resources
and different profitability of the products, when PrettyFit
has more than one product in the product family, it is
important to plan the right product mix (quantity of each
product) to produce to achieve a maximum profit.
• Moreover, with the increasing customer demand
worldwide, especially in the Asia Pacific region, PrettyFit
foresees that the distribution cost may also become a
concern if there is no properly developed distribution plan
considering both available supplies and customer
demands.

• How should PrettyFit plan the product mix to maximize profit with the limited resources?
• How would the available resources and the unit profit of each product affect the overall
profit from producing the products?
• How should PrettyFit develop a distribution plan to minimize the distribution cost while
considering both available supplies and customer demands?

4
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Scenario for Lesson 03 – 05: Product Mix and Distribution Planning

Additional Information for Lesson 03
• Recently, PrettyFit opened a new manufacturing plant in Myanmar. As a start,
the plant focuses on producing two products: Midi-dress and Mini-dress. As
the Operations Manager, Kate is tasked to plan the right product mix (quantity
of each product) to produce daily to achieve a maximum profit.
• Among the main resources consumed to produce the Midi-dress and Mini-
dress, Kate knows that the labour time is not a concern as it is more than
enough. Table below summarizes the amount of machine time and cloth
needed to produce one unit of Midi-dress and Mini-dress, with expected profit
earned from each of the dresses sold.

Machine Time (Hour) Cloth (m2) Unit Profit ($)

Midi-dress 0.3 1.2 24
Mini-dress 0.6 0.8 36

• Currently, the plant has a maximum of 45 hours of machine time available

daily. Kate understands that the suppliers can provide up to 84 m2 of cloth
each day.
• Kate recalls that the product mix can be determined using a mathematical
model and a graphical approach.

5
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Scenario for Lesson 03 – 05: Product Mix and Distribution Planning

Additional Information for Lesson 04

• In PrettyFit’s new manufacturing plant in Myanmar, Kate has successfully planned the
right product mix for two products: Midi-dress and Mini-dress. Now PrettyFit is planning
to further expand the product family by adding one more product: Maxi-dress.
• Kate was informed by the marketing department that there are pre-orders placed by
customers: 15 units of Midi-dress and 30 units of Mini-dress daily. Kate now has a
more challenging task of planning the product mix for three products: Midi-dress, Mini-
dress and Maxi-dress.
• To meet the customer demand and produce according to the new product mix, Kate
has to consider the availability of all resources: machine time, cloth and labour time.
The plant now has a maximum of 150 hours of machine time and 500 hours of labour
time available daily. Meanwhile, the suppliers can provide up to 198 m2 of cloth each
day.
• Table below summarizes the amount of each resource needed to produce one unit of
the products with expected profit earned from each of the products sold.

Product Machine Time (Hour) Cloth (m2) Labour Time (Hour) Unit Profit ($)
Midi-dress 0.3 1.2 1.5 24
Mini-dress 0.6 0.8 2 36
Maxi-dress 0.9 1.5 1.8 42

• Kate recalls that the new product mix can be determined using a mathematical model
and Excel Solver.

6
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Scenario for Lesson 03 – 05: Product Mix and Distribution Planning

Additional Information for Lesson 05

• To plan for the distribution, Kate checked the

• Kate has successfully expanded the
inventories at the manufacturing plants in
product line by introducing Maxi-dress into
the Asia Pacific region and found three
the product family. With broader customer
plants with enough inventories. Table 2
base built in the Asia Pacific region, now it
below shows the three manufacturing plants
is imperative for her to consider how to
and their available supplies.
distribute the products from the
manufacturing plants to the customers to Table 2: Manufacturing Plants and Supplies
meet the customer demands while Manufacturing Plants at Supply (Box )
keeping the distribution cost at its A. Beijing 320
minimum. B. Kuala Lumpur 260
C. Sydney 480
• For the coming month, Kate has received
Total 1060
orders placed from customers in the Asia
Pacific region. Table 1 below shows the • Help Kate develop a shipping arrangement if
locations of the customers and their the shipping cost per box of products from
demands. each manufacturing plant to each customer
region is given in the following Table 3.
Table 1: Customers and Demands Table 3: Shipping Cost
Customers at Demand (Box) To (Cost per Box of Products in US $)
1. Bangkok 210 From 1 2 3 4
2. Hong Kong 350 Bangkok Hong Kong Manila Singapore
3. Manila 200 A. Beijing 32 18 42 34
4. Singapore 300 B. Kuala Lumpur 12 30 15 10
Total 1060 C. Sydney 48 55 23 36

7
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Scenario Definition Template

• What do we know?

• What do we not know?

• What do we need to find out?

8
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Scenario – Tasks of the day

Assume that you are a consultant helping Kate plan the right product mix
(quantity of each product) to produce daily. Help her
• Formulate a Linear Programming (LP) model for the product mix problem.
• Apply the Excel Solver to determine the optimum quantity to produce for
each product in order to maximize the daily profit.
• Interpret answer and sensitivity reports to
 Determine what is limiting PretyFit from making more profit and how
the profit will be affected if PretyFit could have more resources.
 Determine how sensitive the daily profit from the two products is to
the unit profit of the Midi-dress.
 In case a product quantity is zero in the optimal product mix,
determine the impact of forcing it to be non-zero on the daily profit.

• Lesson 03 helped Kate plan the product mix (two products) and conduct some sensitivity
analysis using the graphical method.
• Lesson 04 will help Kate plan the product mix (more than two products) and conduct
sensitivity analysis using Excel Solver.
• Lesson 05 will help Kate develop a distribution plan to minimize the distribution cost based
on the transportation model.

9
 OFFICIAL (CLOSED) \ NON-SENSITIVE

LP Structure
Define decision variables x1, x2 and x3

Maximize (profit/ sales/ results) or

Minimize (cost/ time/ resources):
Z = c 1x 1 + c 2x 2 + c 3x 3 (objective function)
Objective function coefficients
Subject to:
a11x1 + a12x2 + a13x3 <= b1 (limited resources)
a21x1 + a22x2 + a23x3 = b2 (exact requirement)
a31x1 + a32x2 + a33x3 >= b3 (minimum requirement)
X1, x2 , X3 >= 0 (non-negativity)

Constraint coefficients
10
 OFFICIAL (CLOSED) \ NON-SENSITIVE

How to Solve LP Problems?

• Manual graphical method (good for
illustrating the methodology; restricted to 2
variables)
• Software tools
 Now widely available!
 Solves complex problems (hundreds of variables
& constraints) in short time
 In real life applications, formulating the LP model
is the most challenging part!

11
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Feasibility & Optimality

• Feasibility – Solutions that satisfy all constraints in
the model.
• Optimality – Solutions that satisfy all constraints in
the model and results in maximum (or minimum)
objective function.
• An LP problem may have
 One optimal solution
 Multiple optimal solutions or
 No optimal solution (e.g. unbounded solution)
 No solution (not feasible)

12
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Feasibility & Optimality - One Optimal Solution

X1 ≥ 0
140 The optimal
Constraint
130 solution is an
Objective Function
120 Feasible Region extreme point
110
Optimal Point
with the
100
10X1+5X2 ≤ 800 maximum
90
objective
80
70
function value
X2 60 `
50
40
0.75X1+X2 ≤ 90
30
20 X2 ≥ 0
10
0
-10-10 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140

X1 30X1+18X2

13
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Feasibility & Optimality - Multiple Optimal Solutions

X1 ≥ 0
140
Constraint
130
Objective Function
120 Feasible Region
110
Optimal Point
100
10X1+5X2 ≤ 800
90
80
70 Infinite number of solutions
X2 60
Slope of objective 50
`

function is the same 40

0.75X1+X2 ≤ 90
as the slope of a 30
constraint 20 X2 ≥ 0
10
0
-10-10 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140

X1 20X1+10X2

Solver does not alert the user about the existence of alternate optimal solutions

14
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Feasibility & Optimality - Unbounded Solution

X1 ≥ 0
140 No optimal
Constraint
130
Objective Function
solution for
120 Feasible Region maximization
110
Optimal Point problem as the
100
10X1+5X2 ≥ 800 feasible region
90
80
is unbounded
70 and extends
X2 60 towards infinity.
`
50 Optimal
40
0.75X1+X2 ≥ 90 solution is not
30
present.
20 X2 ≥ 0
10
0
-10-10 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140

X1

15
 OFFICIAL (CLOSED) \ NON-SENSITIVE
Feasibility & Optimality - Unbounded Solution
(Excel Solver Results)

16
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Feasibility & Optimality - No Feasible Solution

X1 ≥ 0
140
Constraint
130
Objective Function
120 Feasible Region
110
Optimal Point
100
10X1+5X2 ≥ 800
90
No feasible region: no
80
70
point simultaneously lies
X2 60
above line 1 and below
`
50 line 2. Constraints are in
conflict
1
40
2X1+3X2 ≤ 120
30
20 X2 ≥ 0
10
2
0
-10-10 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140

X1

17
 OFFICIAL (CLOSED) \ NON-SENSITIVE
Feasibility & Optimality - No Feasible Solution
Excel Solver Results

18
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Test Yourself: Question 1

Analyze the following LP problem and comment on feasibility
and optimality:

Max Z = 20X1+10X2

Subject to
10X1+5X2 <= 800
0.75X1+X2<= 90
X1, X2>=0

Multiple
Optimal
Solution

19
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Test Yourself: Question 2

Analyze the following LP problem and comment on feasibility
and optimality:

Max Z = 30X1+18X2

Subject to
10X1+5X2 >= 800
0.75X1+X2>= 90
X1, X2>=0

Unbounded
Solution

20
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Test Yourself: Question 3

Analyze the following LP problem and comment on feasibility
and optimality:

Max Z = 30X1+18X2

Subject to
10X1+5X2 >= 800
2X1+3X2<= 120
X1, X2>=0

No Feasible
Solution

21
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Test Yourself: Question 4

Analyze the following LP problem and comment on feasibility
and optimality:

Max Z = 30X1+18X2

Subject to
10X1+5X2 <= 800
0.75X1+X2<= 90
X1, X2>=0

Unique
Optimal
Solution

22
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Scenario for the Interactive Seminar (IS)

• Danny has successfully planned the right product mix to produce the two product types: Type
A and Type B. Now he is planning to further expand the product family by adding one more
product: Type C.
• The marketing department informed Danny that customers pre-ordered 35 units of Type C
product for the coming week.
• To meet the customer demand and produce according to the new product mix, Danny has
collected information on the availability of the resources. With the introduction of Type C
product, the main raw material used to produce the products may also limit the production
amount. For the coming week, the plant has a maximum of 160 machine hours and 300 labour
hours available, and the suppliers can provide up to 91.5 kg of plastic.
• Table below summarizes the amount of each resource needed to produce one unit of the
products with expected profit earned from each of the products sold.
Requirement of resources (unit per product )
Product Unit Profit ($)
Machine (Hours) Labour (Hours) Plastic (Gram)
Type A 0.8 3 600 120
Type B 1.6 2 900 150
Type C 1.2 1.5 1500 180

Help Danny determine the optimum quantity to make for each product in the coming week to
maximize the profit.

23
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Test Yourself: LP Formulation

Consider the IS scenario given, formulate the LP model by stating
the decision variables, constructing the objective function and
formulating the constraints.

Post and share your answers

in Padlet / MS Teams

24
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Test Yourself: LP Formulation

Consider the IS scenario given, formulate the LP model by stating
the decision variables, constructing the objective function and
formulating the constraints.

Decision Variables
Let X1 be the number of Type A to produce for the coming week
Let X2 be the number of Type B to produce for the coming week
Let X3 be the number of Type C to produce for the coming week

Objective Function
Maximize profit Z = 120X1 + 150X2 + 180X3

Constraints
0.8X1 + 1.6X2 + 1.2X3 ≤ 160 (Availability of Machine Hours)
3X1 + 2X2 + 1.5X3 ≤ 300 (Availability of Labour Hours)
600X1 + 900X2+ 1500X3 ≤ 91500 (Availability of Plastic)
X3 ≥ 35 (Units of Type C Product Pre-ordered)
X1 , X2, X3 ≥ 0 (Non–negativity constraints for the decision variables)
25
 OFFICIAL (CLOSED) \ NON-SENSITIVE

IS Lecture (2)
MS Excel Solver
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Installation of Excel Solver

Open Excel and Click Data, if you do not have Solver, install it by following
the steps below.

Go to ‘File’  ‘Options’  ‘Add-ins’  Manage Excel Add-ins ‘Go’  Check

Solver Add-in box  ‘OK’).

27
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Spreadsheet Design Guidelines

• Organize data, then build model around the data.
• Do not embed numeric constants in formulas. Place
numeric constants in separate cells.
• Related things (e.g. Left-Hand Side and Right-Hand
Side of constraints) should be placed in close
proximity.
• Design to allow natural left-to-right eye scanning of
information.
• Use color, shading and borders to distinguish
changeable parameters (variables) from other
elements of the model.
• Use text boxes and cell comments to document
various elements of the model.

28
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Hands-on Session: Use of Excel Solver

Consider the IS scenario, use Excel Solver to find the optimal
solution for the LP model.

Post and share your answers

in Padlet / MS Teams

29
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Hands-on Session: Setting Up the Spreadsheet

Decision variables Objective Function Coefficients

Formula for objective function

=SUMPRODUCT(E11:E13,F11:F13)

Right-hand
=$E$11*C24+$E$12*D24+$E$13*E24 Formula for Left-hand Side of constraint Side values

Constraint Coefficients
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Hands-on Session: Use of Excel Solver Objective function cell

(maximize)
Go to Excel->Data->Solver.
Range of cells for decision variables

Constraints

31
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Hands-on Session: Use of Excel Solver

 Optimal solution for an LP problem can be obtained if and only if
 the objective function and constraints are formulated correctly, and
 all constraints are satisfied.
 For multiple optimal solutions, Excel Solver will only give one solution.

Select to
generate Reports

32
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Hands-on Session: Excel Solver Solution

Optimal Solution

Optimal Objective
Function Value

Final constraint Left-hand side values 33

 OFFICIAL (CLOSED) \ NON-SENSITIVE

IS Lecture (3)
Sensitivity Analysis
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Recap: Assumptions of Linear Programming (LP)

 The Certainty assumption: Is this always
 Each model parameter is known with certainty. true???

 Model parameters here refer to objective function coefficients, right-

hand side values of model constraints, and constraint coefficients.
 The Proportionality assumption:
 The contribution to the objective function (left-hand side of each
constraint) from each decision variable is proportional to the value of
the decision variable.
 The Additivity assumption:
 Total contributions to the objective function (left-hand side of each
constraint) is the sum of individual contributions from each variable.
 The Continuity assumption:
 Variables can take on any value within a given feasible range

35
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Recap: Role of Sensitivity Analysis of Optimal solution

• When formulating LP models, the model

parameters are assumed to be known with
certainty
• In reality, model parameters are simply estimates
(best guesses) that are subject to change.
• Sensitivity analysis is the analysis of parameter
changes and their effects on the model solution.
• Sensitivity analysis provides a better picture of how
the solution to a problem will change if different
parameters in the model change.

36
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Recap: Sensitivity Analysis

• Sensitivity analysis aids in answering questions such as:
 Range of optimality - The range of values that the objective
function coefficients can assume without changing the optimal
solution.
 Range of feasibility - The range of values for the right-hand side
of a constraint, in which the shadow price for the constraint remains
unchanged - the valid range of shadow price.
 Reduced Cost - The impact of forcing a variable which is
currently zero to be non-zero on the optimal objective function
value.
 Shadow price - The impact of increasing or decreasing the right-
hand side value of various constraints on the optimal objective
function value.
 The impact of changing the constraint coefficients on the optimal
solution to the problem.

37
 OFFICIAL (CLOSED) \ NON-SENSITIVE
Hands-on Session:
Binding, Non-binding Constraints and Slacks
Consider the IS scenario, how does Danny know if the resources
have been used up? What do you call the constraints corresponding
to the resources being used up?

Post and share your answers

in Padlet / MS Teams

38
 OFFICIAL (CLOSED) \ NON-SENSITIVE
Hands-on Session:
Binding, Non-binding Constraints and Slacks
Non-binding Constraints
Eg. Availability of Machine Hours constraint: 66 hours of unused
Machine Time (non-binding constraint; slack = 160 - 94 = 66 hours)
Availability of Labour Hours constraint: 52.5 hours of unused Labour
Time (non-binding constraint; slack = 300 – 247.5 = 52.5 hours)
Final constraint Left-hand side values
Constraints
Cell Name Cell Value Formula Status Slack
$G$22 Availability of Machine Hours LHS 94.00 $G$22<=$I$22 Not Binding 66
$G$23 Availability of Labor Hours LHS 247.50 $G$23<=$I$23 Not Binding 52.5
$G$24 Availability of Plastic LHS 91500.00 $G$24<=$I$24 Binding 0
$G$25 Units of Type C Product Pre-ordered LHS 35.00 $G$25>=$I$25 Binding 0.00

Binding Constraints
Eg. Availability of Plastic constraint is binding (slack = 0, final left-hand Side value of the
constraint is equal to right-hand side value).
Increase in amount of Plastic (currently at 91,500 grams) would improve the optimal
solution (i.e. increase the total profit).
Pre-order for Type C constraint is also binding. How does the increase in Units of Type C
Product Pre-ordered affect the optimal solution? Check shadow price in Sensitivity 39
Report!
 OFFICIAL (CLOSED) \ NON-SENSITIVE
Hands-on Session:
Shadow Price of Constraints

Consider the IS scenario, assume that the amount of Plastic is

reduced from 91,500 grams to 90,500 grams.
a) Determine the shadow price of the Availability of Plastic
constraint.
b) Calculate the new profit for the coming week based on the
shadow price.
c) Determine the range of feasibility for the Availability of Plastic
constraint. What will happen if the amount of Plastic is
decreased or increased beyond this range?

Post and share your answers

in Padlet / MS Teams

40
 OFFICIAL (CLOSED) \ NON-SENSITIVE
Hands-on Session:
Shadow Price of Constraints
How much is the worth of the additional resource?
Constraints
Final Shadow Constraint Allowable Allowable
Cell Name Value Price R.H. Side Increase Decrease
$G$22 Availability of Machine Hours LHS 94 0 160 1E+30 66
$G$23 Availability of Labor Hours LHS 247.5 0 300 1E+30 52.5
$G$24 Availability of Plastic LHS 91500 0.2 91500 10500 39000
$G$25 Units of Type C Product Pre-ordered LHS 35 -120 35 26 8.75

 Shadow price: The increase (or decrease) in objective function given 1 unit
increase (or decrease) in resource (right-hand side value)
 The shadow price remains valid if and only if the right-hand side value
stays within the Range of Feasibility:
(Right-hand side value- Allowable Decrease, Right-hand side value +
Allowable Increase).
 Shadow price is non-zero only for binding constraints. For non-binding
constraints (surplus resources), additional resource is not valuable.
 For example, increasing the amount of Machine Hours does not change
the total profit. 41
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Hands-on Session: Shadow Price of Constraints

– Contribution to Objective Function Value
Constraints
Final Shadow Constraint Allowable Allowable
Cell Name Value Price R.H. Side Increase Decrease
$G$22 Availability of Machine Hours LHS 94 0 160 1E+30 66
$G$23 Availability of Labor Hours LHS 247.5 0 300 1E+30 52.5
$G$24 Availability of Plastic LHS 91500 0.2 91500 10500 39000
$G$25 Units of Type C Product Pre-ordered LHS 35 -120 35 26 8.75

 Assume that the available Plastic is reduced to 90,500 grams. With

the decrease of 1,000 grams (within allowable decrease) of Plastic,
the total profit (objective function) will decrease to:
$14,100 + ($0.2 x (-1000)) = $13,900
 The shadow price of Availability of Plastic constraint ($0.2) is valid
when available Plastic (Right-hand side of Availability of Plastic
constraint) is between 52,500 grams (91,500-39,000) and 102,000
grams (91,500+10,500). (Range of feasibility)
 Note that the optimal solution changes when the right-hand side
value of a binding constraint changes. (i.e. Production of Type A,
Type B and Type C would no longer be X1= 65; X2=0; X3=35.) 42
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Hands-on Session:
Sensitivity Analysis of Objective Function Coefficients

Consider the IS scenario,

a) What would happen if the unit profit of Type A is decreased from
$120 to $110?
b) What is the range of optimality for the objective function
coefficient of X1 (Type A)?
c) What would happen if the unit profit of Type A is increased or
decreased beyond the range of optimality?

Post and share your answers

in Padlet / MS Teams

43
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Hands-on Session:
Range of Optimality for Objective Function Coefficients
For the optimal solution to remain unchanged, how much can the
objective function coefficient change?

Variable Cells
Final Reduced Objective Allowable Allowable
Cell Name Value Cost Coefficient Increase Decrease
$E$11 Type A Quantity 65 0 120 1E+30 20
$E$12 Type B Quantity 0 -30 150 30 1E+30
$E$13 Type C Quantity 35 0 180 120 1E+30

 For the solution (product mix) to stay optimal,  When the unit profit of Type A is
the objective function coefficient can assume decreased from $120 to $110, the optimal
the range (‘Coefficient – Allowable Decrease’, solution does not change. The optimal
‘Coefficient + Allowable Increase’). (Range of objective function value (total profit) will
Optimality) decrease by 65*(120-110) = $650
 In other words, the optimal solution (X1= 65;  The Range of Optimality for objective
X2=0; X3=35) is valid when the unit profits function coefficient of X1 (Type A) is:
stay within the following ranges: P1 ≥ $100.
Product Unit Profit  If the unit profit of Type A is decreased
Type A P1≥ 120 - 20 = $100 below $100, the optimal solution will
Type B P2 ≤ 150 + 30 = $180
change, i.e., the optimal solution will
not be (X1= 65; X2=0; X3=35) anymore.
Type C P3 ≤ 180 + 120 = $300 44
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Hands-on Session:
Range of Optimality for Objective Function Coefficients

NOTE
 For multiple optimal solutions, the allowable increase or allowable
decrease for some objective function coefficients will be zero. See
an example below:

 To find an alternate optimal solution:

Add a constraint that holds the objective function at the current optimal
value, then try to optimize other possible objectives.
45
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Hands-on Session:
Reduced Cost of a Zero-Value Decision Variable

Consider the IS scenario,

a) What is the reduced cost for the decision variable X2 (the number of
Type B to produce)? How do you interpret this reduced cost?
b) Use the reduced cost for the decision variable X1 (the number of Type A
to make) as an example, illustrate how reduced cost is calculated.
c) Use the reduced cost for the decision variable X2 (the number of Type B
to make) as an example, illustrate how reduced cost is calculated.
d) Determine the reduction in the amount of Plastic required to produce 1
unit of Type B so that producing Type B will become worthwhile.

Post and share your answers

in Padlet / MS Teams
46
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Hands-on Session:
Reduced Cost of a Zero-Value Decision Variable

What does it take to produce one unit of Type B product?

Reduced Cost: An estimate of how much the objective function will change if a decision
variable which is currently 0 is forced to be non-zero. For example,
 Number of Type B (X2) has a final value of 0 unit to produce. To produce 1 unit of Type
B, the objective function (total profit for the coming week) will decrease by $30.
 How to make the production of Type B worthwhile? The unit profit must be reduced by
at least $-30 (i.e., the unit profit has to be increased by more than $30).
 So, in order to be worthwhile to produce 1 unit of Type B, its unit profit should be at
least: $150 + $30 = $180.

47
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Hands-on Session:
Reduced Cost of a Zero-Value Decision Variable

Reduced cost is the amount per unit of the product that contributes to overall
objective function minus the value (shadow price) of the resources it consumes.

Objective coeff. Shadow price Constraints

Final Shadow
Cell Name Value Price
Reduced cost for Type A toy, X1 $G$22 Availability of Machine Hours LHS 94 0
= 120 – (0.8 x 0) – (3 x 0) – (600 x 0.2) – (0 x (-120)) $G$23 Availability of Labor Hours LHS 247.5 0
$G$24 Availability of Plastic LHS 91500 0.2
=0 $G$25 Units of Type C Product Pre-ordered LHS 35 -120

Constraints
Constraint coefficient for X1 X1 X2 X3 Constraint
0.8 1.6 1.2 Availability of Machine Hours
3 2 1.5 Availability of Labor Hours
600 900 1500 Availability of Plastic
1 Units of Type C Product Pre-ordered
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Hands-on Session:
Reduced Cost of a Zero-Value Decision Variable

Reduced cost is the amount per unit of the product that contributes to overall
objective function minus the value (shadow price) of the resources it consumes.

Objective coeff. Shadow price

Reduced cost for Type B toy, X2

= 150 – (1.6 x 0) – (2 x 0) – (900 x 0.2) – (0 x (-120))
= -30

Constraints

Constraint coefficient for X2

 OFFICIAL (CLOSED) \ NON-SENSITIVE

Hands-on Session: Determine the Requirement to Change

Decision Variable from Zero Value to Non-Zero

The reduction in the amount of Plastic required to produce 1 unit of Type

B product so that producing Type B will become worthwhile.
Variable Cells
Final Reduced Objective Allowable Allowable
Cell Name Value Cost Coefficient Increase Decrease
$E$11 Type A Quantity 65 0 120 1E+30 20
$E$12 Type B Quantity 0 -30 150 30 1E+30
$E$13 Type C Quantity 35 0 180 120 1E+30

 Let P be the amount of “revised” Plastic needed for producing 1 unit of

Type B product.
Let reduced cost of Type B = 0
150 - (1.6*0) - (2*0) - (P*0.2) - (0*(-120)) = 0
P = 150/0.2 = 750 gram
 From the computation, to make the
production of Type B worthwhile, the
amount of Plastic required has to be
reduced by more than 150 gm
(900 - 750).
50
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Hands-on Session: The Answer and Sensitivity Report

Consider the IS scenario,

a) Summarize and interpret the critical information you can obtain
from the Answer Report.
b) Summarize and interpret the critical information you can obtain
from the Sensitivity Report.

Post and share your answers

in Padlet / MS Teams

51
 OFFICIAL (CLOSED) \ NON-SENSITIVE
Hands-on Session:
Overview of Answer Report in Excel Solver

Optimal solution:
number of Type A,
Type B and Type C to
Optimal objective function Value produce
– Max Profit

Constraints’ status 52
Hands-on Session:
OFFICIAL (CLOSED) \ NON-SENSITIVE

Overview of Sensitivity Report in Excel Solver

Reduced Cost The minimum amount by Range of Optimality

which the Objective Function Coefficient of Range of unit profit contributions that the
a variable should change to cause that current optimal solution (65 units of Type A
variable to become non-zero. and 35 units of Type C) will retain optimal.

Range of feasibility
Shadow Price -Range of Shadow Price Validity.
The change in the objective -For example, if the increase in the right-hand side value of
function value per unit increase Plastic constraint is beyond the allowable increase of
(decrease) in the right-hand side 10500 gram, then the shadow price will change.
value of the constraint.
53
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Learning Outcomes
At the end of the lesson, students should be able to:
• Identify problems that can be solved using linear programming.
• Formulate a Linear Programming (LP) model by establishing the decision variables, objective
function and constraints.
• Solve a Linear programming (LP) problem with two decision variables using graphical method.
• Perform sensitivity analysis on the objective function coefficients and right-hand side values of the
constraints using graphical method.
• Interpret feasibility and optimality of an LP solution.
• Solve an LP model using MS Excel Solver.
• Perform sensitivity analysis using reports generated from MS Excel Solver.
• Interpret shadow price to determine the impact of changing constraint right-hand side value on the
optimal objective value.
• Explain range of optimality: the range of values that the objective function coefficients can assume
without changing the optimal solution.
• Explain range of feasibility: the range of values for the right-hand side of a constraint, in which the
shadow price for the constraint remains unchanged.
• Calculate Reduced Cost to determine the impact of forcing a decision variable which has zero value
to be non-zero on the objective function value.
• Identify a transportation problem that deals with the distribution of goods from multiple supply
sources to multiple demand destinations.
• Apply Northwest Corner Rule and Lowest Cost Method to determine a feasible solution for a
transportation problem.
• Formulate the transportation problem that involves minimizing the cost of shipping goods from a set
of origins to a set of destinations as a linear programming model.
• Apply Excel Solver to find an optimal shipment allocation for the transportation problem to minimize
the total transportation cost.
• Compare the heuristic solutions with the solution of linear programming method.
54
 OFFICIAL (CLOSED) \ NON-SENSITIVE

E-learning Video
• You may wish to access the e-learning
video from : https://docs.google.com/file/d/0Bz-
Yttbj61bxTXAtQXplQ2RxN2c/edit?usp=sharing

55
 OFFICIAL (CLOSED) \ NON-SENSITIVE

Overview of E210 Operation Planning Module

Decision
Operations LP – Integer Making using
Planning Programming Decision Tree
Overview
LP -
Transportation Network
Model Diagram and
Decision
Making with Min. Spanning
Dependent Tree
Decisions

LP formulation LP – Binary
Process Strategy and EXCEL Integer
& Capacity Solver with Programming Decision
Requirements Sensitivity Making using
analysis Utility Dissimilarity
Function Index and MST

Linear
Programming Decision
(LP) Graphical Making under
Method uncertainty
and risk

We are
here !
56