ISL 323E Operations Research

Fall 2024

Assoc. Prof. Dr. Fuat Kosanoğlu

ISL 323E Operations Research
Week II
8 November 2024
 Minimization Problem
 Dorian Auto manufactures luxury cars and trucks. The company
believes that its most likely customers are high-income women and
men. To reach these groups, Dorian Auto has embarked on an
ambitious TV advertising campaign and has decided to purchase 1-
minute commercial spots on two types of programs: comedy shows
and football games. Each comedy commercial is seen by 7 million
high-income women and 2 million high- income men. Each football
commercial is seen by 2 million high-income women and 12 million
high-income men. A 1-minute comedy ad costs $50,000, and a 1-
minute football ad costs $100,000. Dorian would like the
commercials to be seen by at least 28 million high-income women
and 24 million high-income men. Use linear programming to deter-
mine how Dorian Auto can meet its advertising requirements at
minimum cost.
2/68
Elements of Optimization Model
 Decision Variables
◼ x1 : Number of 1-minute comedy ads purchased
◼ x2 : Number of 1-minute football ads purchased
 Objective function
◼ min z =50x1 +100x2
 Constraints:
◼ 7x1 +2x2 ≥28
◼ 2x1 +12x2 ≥ 24
◼ x1, x2 ≥ 0

3/68
4/68
Assumptions of Linear Programming
 Proportionality
 Additivity
 Divisibility
 Certainty

5/68
Proportionality Assumption
 The contribution of each activity to the objective
function is proportional to the level of the activity.
 The contribution of each activity to the left-hand side of
each functional constraint is proportional to the level of
the activity.
 In other words:
◼ The objective function term for activity j is of the form 𝑐𝑗 𝑥𝑗
◼ The constraint terms for activity j are of the form 𝑎𝑖𝑗 𝑥𝑗
◼ No exponents other than 1!

6/68
Proportionality Assumption
 Z=3𝑥𝑗

7/68
Start-Up Costs Violate Proportionality
3𝑥𝑗 − 1 𝑖𝑓 𝑥𝑗 > 0
 Z=൝
0 𝑖𝑓 𝑥𝑗 = 0

8/68
Increasing Marginal Profits Violate Proportionality
1.5
 Z=3𝑥𝑗

9/68
Decreasing Marginal Profits Violate Proportionality

10/68
Quantity Discounts Violate Proportionality

11/68
Is Proportionality a Reasonable Assumption?
 Often real-life problems violate proportionality in
one or more of these ways.
 But it’s often reasonable to assume proportionality
even if it’s violated.
 There are ways of handling some forms of non-
proportionality.
 Otherwise, use mixed-integer programming or
non-linear programming

12/68
Additivity Assumption
 Every function (objective function or left-hand
side of constraints) is the sum of the individual
contributions of the respective activities.
 In other words:
◼ The variables in the objective functions and
constraints are added together,
◼ never multiplied.
◼ For example, no terms like 5𝑥1 𝑥2 .

13/68
Non-Additive Objective Function
 Suppose the objective function for the Wyndor
Glass problem were:
 Z = 𝑥1 + 5𝑥1 𝑥2

14/68
Divisibility Assumption
 The decision variables are allowed to take on any values
that satisfy the functional and non-negativity constraints.
 In other words:
 The values of the decision variables may be fractional.

15/68
Certainty Assumption
 The value assigned to each parameter is a known
constant.
 In other words:
◼ All of the parameters 𝑐𝑗 , 𝑏𝑖 , and 𝑎𝑖𝑗 are known with
certainty.

16/68
 LP Problems
 Diet Problem
 Workforce scheduling Problems
 Blending Problems
 Production process problems
 Inventory models
 Budget Planning problems

17/68
Diet Problem
 My diet requires that all the food I eat come from one of the four
“basic food groups” (chocolate cake, ice cream, soda, and
cheesecake). At present, the following four foods are available for
consumption: brownies, chocolate ice cream, cola, and pineapple
cheesecake. Each brownie costs 50¢, each scoop of chocolate ice
cream costs 20¢, each bottle of cola costs 30¢, and each piece of
pineapple cheesecake costs 80¢. Each day, I must ingest at least
500 calories, 60 oz of chocolate, 100 oz of sugar, and 80 oz of fat.
The nutritional content per unit of each food is shown in Table
below. Formulate a linear programming model that can be used to
satisfy my daily nutritional requirements at minimum cost.

18/68
Diet Problem

Type of Food Calories Chocolate (oz) Sugar(oz) Fat(oz)

Brownie 400 30 20 20

Chocolate ice
200 20 20 40
cream
Cola 150 0 40 10
Pineapple
500 0 40 50
cheesecake

19/68
Elements of Optimization Problem
 Decision Variables:
◼ x1 = number of brownies eaten daily
◼ x2 = number of scoops of chocolate ice cream eaten
daily (scoop)
◼ x3 = bottles of cola drunk daily
◼ x4 = pieces of pineapple cheesecake eaten daily
 Objective Function
◼ Minimize the total cost of my diet
◼ What is the total cost?
◼ Total cos is a function of decision variables
 Constraints 20/68
Data
Type of Chocolate Sugar(oz) Fat(oz)
Calories Cost
Food (oz)

Brownie 400 30 20 20 50

Chocolate
200 20 20 40 2
ice cream
Cola 150 0 40 10 3
Pineapple
500 0 40 50 8
cheesecake

Required
500 60 100 80
Amount

21/68
Objective Function

 How much do we need to pay for three brownies,

two scoop of ice cream and one bottle of cola?
◼ 3*(0.5)+2*(0.2)+1*(0.33)+0*(0.8)=2.2
 Lets write this as a function of decision variables
◼ x1 *(0.5)+ x2 *(0.2)+ x3 *(0.3)+ x4 *(0.8)=Z
 This is a linear function xj

22/68
Constraints
 Minimum required calories, chocolate, sugar and fat
amount:
◼ Minimum calories:
400x1 + 200x2 + 150x3+ 500x4 ≥500
◼ Minimum chocolate :
30x1 + 20x2 ≥ 60
◼ Minimum sugar :
20x1 + 20x2 + 40x3+ 40x4 ≥100
◼ Minimum fat:
20x1 + 40x2 + 10x3+ 50x4 ≥80
◼ Nonnegativity
xj ≥0
23/68
Mathematical Model
 Min Z= 0.5x1+ 0.2x2+ 0.3x3+ 0.8x4
400x1 + 200x2 + 150x3+ 500x4 ≥500
30x1 + 20x2 ≥ 60
20x1 + 20x2 + 40x3+ 40x4≥100
20x1 + 40x2 + 60x3+ 50x4≥80
xj ≥0 (j=1, 2, 3, 4)
or
4
min 𝑍 = ෍ 𝑐𝑗 𝑥𝑗
𝑗=1
4
෍ 𝑎𝑖𝑗 𝑥𝑗 ≥ 𝑏𝑖 𝑓𝑜𝑟 𝑖 = 1,2,3,4
𝑗=1
xj ≥0 (j=1, 2, 3, 4)
24/68
 Work-Scheduling Problem
 A post office requires different numbers of full-time employees on different days of the
week. The number of full-time employees required on each day is given in Table below.
Union rules state that each full-time employee must work five consecutive days and then
receive two days off. For example, an employee who works Monday to Friday must be
off on Saturday and Sunday. The post office wants to meet its daily requirements using
only fulltime employees. Formulate an LP that the post office can use to minimize the
number of full-time employees who must be hired.

Day # of full time employees required

Monday 17
Tuesday 13
Wednesday 15
Thursday 19
Friday 14
Saturday 16
Sunday 11

25/68
min 𝑧 = 𝑥1 + 𝑥2 + 𝑥3 + 𝑥4 + 𝑥5 + 𝑥6 + 𝑥7
𝑠. 𝑡. 𝑥1 ≥ 17
𝑥2 ≥ 13
𝑥3 ≥ 15
𝑥4 ≥ 19
𝑥5 ≥ 14
𝑥6 ≥ 16
𝑥7 ≥ 11
𝑥𝑖 ≥ 0 (𝑖 = 1,2, … , 7)
26/68
27/68
Budget Planning Problem
 Star Oil Company is considering five different investment opportunities.
The cash out- flows and net present values (in millions of dollars) are
given in Table below. Star Oil has $40 million available for investment
now (time 0); it estimates that one year from now (time 1) $20 million
will be available for investment. Star Oil may purchase any fraction of
each investment. In this case, the cash outflows and NPV are adjusted
accordingly. Star Oil wants to maximize the NPV that can be obtained by
investing in investments 1–5. Formulate an LP that will help achieve this
goal. Assume that any funds left over at time 0 cannot be used at time 1.

Investments
1 2 3 4 5
Year 0 11 53 5 5 29
Year 1 3 6 5 1 34
NPV 13 16 16 14 39

28/68
 Budget Planning Problem
 Decision variables:
𝑥𝑖 = 𝑓𝑟𝑎𝑐𝑡𝑖𝑜𝑛 𝑜𝑓 𝑖𝑛𝑣𝑒𝑠𝑡𝑚𝑒𝑛𝑡 𝑖 𝑝𝑢𝑟𝑐ℎ𝑎𝑠𝑒𝑑 𝑏𝑦 𝑆𝑡𝑎𝑟 𝑂𝑖𝑙

 Objective function: Maximize the total NPV

earned from investments
◼ NPV from investment i= (NPV from investment i)(fraction
of investment i purchased)
𝑍 = 13𝑥1 + 16𝑥2 + 16𝑥3 + 14𝑥4 + 39𝑥5

29/68
 Budget Planning Problem
 Constraints:
◼ Star cannot invest more than $40 million at time 0.
◼ Star cannot invest more than $20 million at time 1.
◼ Star cannot purchase more than 100% of investment i

11𝑥1 + 53𝑥2 + 5𝑥3 + 5𝑥4 + 29𝑥5 ≤ 40 (𝑇𝑖𝑚𝑒 0 𝑐𝑜𝑛𝑠𝑡𝑟𝑎𝑖𝑛𝑡)

3𝑥1 + 6𝑥2 + 5𝑥3 + 𝑥4 + 34𝑥5 ≤ 20 (𝑇𝑖𝑚𝑒 1 𝑐𝑜𝑛𝑠𝑡𝑟𝑎𝑖𝑛𝑡 )
𝑥𝑖 ≤ 1 𝑓𝑜𝑟 𝑖 = 1,2, … , 5
𝑥𝑖 ≥ 0 𝑓𝑜𝑟 𝑖 = 1,2, … , 5

30/68
 Blending Problems
 Sunco Oil manufactures three types of gasoline (gas 1, gas 2, and gas 3). Each
type is produced by blending three types of crude oil (crude 1, crude 2, and crude
3). Sunco can purchase up to 5,000 barrels of each type of crude oil daily. The
sales price per barrel of gasoline and the purchase price per barrel of crude oil are
given in Table below. The three types of gasoline differ in their octane rating and
sulfur content, and they must satisfy required standards:
◼ Gas 1 must have an average octane rating of at least 10 and contain at most 1% sulfur
◼ Gas 2 must have an average octane rating of at least 8 and contain at most 2% sulfur
◼ Gas 3 must have an average octane rating of at least 6 and contain at most 1% sulfur
 It costs $4 to transform one barrel of oil into one barrel of gasoline, and Sunco’s
refinery can produce up to 14,000 barrels of gasoline daily.
 Sunco’s customers require : gas 1-3,000 barrels per day; gas 2-2,000 barrels per
day; gas 3-1,000 barrels per day (at most)
 Sunco also has the option of advertising to stimulate demand for its products. Each
dollar spent daily in advertising a particular type of gas increases the daily demand
for that type of gas by 10 barrels.
 Formulate an LP that will enable Sunco to maximize daily profits
31/68
Blending Problems
Crude Octane sulfur(%) Purchase Price
($/barrel)
1 12 0.5 45
2 6 2 35
3 8 3 25

Gas Sales Price($/barrel)

1 70
2 60
3 50

32/68
Blending Problems
 Decisions to make?
◼ How much money should be spent in advertising each
type of gas
◼ How to blend each type of gasoline from the three
types of crude oil available.
◼ yi= dollars spent daily on advertising gas i (i =1, 2, 3)
◼ xij = barrels of crude oil i used daily to produce gas j
 (i= 1, 2, 3; j =1, 2, 3)

33/68
Blending Problems
 Objective Function:
◼ Maximizing daily net profit
◼ Daily profit = daily revenue from gas sales
- daily cost of purchasing crude oil
- daily advertising costs
- daily production costs

34/68
 Blending Problems
 Daily revenue from gas sales
= 70 𝑥11 + 𝑥21 + 𝑥31 + 60 𝑥12 + 𝑥22 + 𝑥32 + 50(𝑥13 + 𝑥23 + 𝑥33 )
 Daily cost of purchasing crude oil:
= 45 𝑥11 + 𝑥12 + 𝑥13 + 35 𝑥21 + 𝑥22 + 𝑥23 + 25 𝑥31 + 𝑥32 + 𝑥33
 Daily advertising costs :
= 𝑦1 +𝑦2 + 𝑦3
 Daily production costs:
= 4(𝑥11 + 𝑥12 + 𝑥13 + 𝑥12 + 𝑥22 + 𝑥23 + 𝑥31 + 𝑥32 + 𝑥33 )
 Objective: Maximize
𝑍 = 2𝑥11 + 11𝑥12 + 𝑥13 + 31𝑥12 + 21𝑥22 + 11𝑥23 + 41𝑥31
+31𝑥32 + 21𝑥33 − 𝑦1 − 𝑦2 − 𝑦3

35/68
 Blending Problems
 Constraint 1: Gas 1 produced daily should equal its daily demand.
𝑥11 + 𝑥21 + 𝑥31 − 10𝑦1 = 3000
 Constraint 2 :Gas 2 produced daily should equal its daily demand.
𝑥12 + 𝑥22 + 𝑥32 − 10𝑦2 = 2000
 Constraint 3 :Gas 3 produced daily should equal its daily demand.
𝑥13 + 𝑥23 + 𝑥33 − 10𝑦3 = 1000
 Constraint 4 :At most 5,000 barrels of crude 1 can be purchased daily.
𝑥11 + 𝑥12 + 𝑥13 ≤ 5000
 Constraint 5 :At most 5,000 barrels of crude 2 can be purchased daily.
𝑥21 + 𝑥22 + 𝑥23 ≤ 5000
 Constraint 6 :At most 5,000 barrels of crude 3 can be purchased daily.
𝑥31 + 𝑥32 + 𝑥33 ≤ 5000
 Constraint 7: At most 14,000 barrels of gasoline can be produced daily.
𝑥11 + 𝑥12 + 𝑥13 + 𝑥12 + 𝑥22 + 𝑥23 + 𝑥31 + 𝑥32 + 𝑥33 ≤ 14
36/68
 Blending Problems
 Constraint 8 : Crude oil blended to make gas 1 must have an average
octane level of at least 10.
12𝑥11 + 6𝑥21 + 8𝑥31
Total octane value in gas 1= ≥ 10 (! 𝑛𝑜𝑡 𝑙𝑖𝑛𝑒𝑎𝑟)
𝑥11 + 𝑥21 + 𝑥31
12𝑥11 + 6𝑥21 + 8𝑥31 ≥ 10(𝑥11 + 𝑥21 + 𝑥31 )
2𝑥11 − 4𝑥21 − 2𝑥31 ≥ 0
 Constraint 9 : Crude oil blended to make gas 2 must have an average
octane level of at least 8.
4𝑥12 − 2𝑥22 ≥ 0
 Constraint 10 : Crude oil blended to make gas 3 must have an
average octane level of at least 6.
6𝑥13 + 2𝑥23 ≥ 0

37/68
 Blending Problems
 Constraint 11: Crude oil blended to make gas 1 must contain at most
1% sulfur.
0.005𝑥11 + 0.02𝑥21 + 0.03𝑥31
Total sulfur in gas 1 mixture = ≤ 0.01
𝑥11 + 𝑥21 + 𝑥31
−0.005𝑥11 + 0.01𝑥21 + 0.02𝑥31 ≤ 0
 Constraint 12: Crude oil blended to make gas 2 must contain at most
2% sulfur.
−0.015𝑥12 + 0.01𝑥32 ≤ 0

 Constraint 13: Crude oil blended to make gas 3 must contain at most
1% sulfur.
−0.005𝑥11 + 0.01𝑥23 + 0.02𝑥33 ≤ 0

38/68
39/68
 Production Process Models
 Rylon Corporation manufactures Brute and Chanelle perfumes. The raw
material needed to manufacture each type of perfume can be purchased
for $3 per pound. Processing 1 lb of raw material requires 1 hour of
laboratory time. Each pound of processed raw material yields 3 oz of
Regular Brute Perfume and 4 oz of Regular Chanelle Perfume. Regular
Brute can be sold for $7/oz and Regular Chanelle for $6/oz. Rylon also
has the option of further processing Regular Brute and Regular Chanelle
to produce Luxury Brute, sold at $18/oz, and Luxury Chanelle, sold at
$14/oz. Each ounce of Regular Brute processed further requires an
additional 3 hours of laboratory time and $4 processing cost and yields 1
oz of Luxury Brute. Each ounce of Regular Chanelle processed further
requires an additional 2 hours of laboratory time and $4 processing cost
and yields 1 oz of Luxury Chanelle. Each year, Rylon has 6,000 hours of
laboratory time available and can purchase up to 4,000 lb of raw material.
Formulate an LP that can be used to determine how Rylon can maximize
profits. Assume that the cost of the laboratory hours is a fixed cost.
40/68
 Solution
 Decision variables:
◼ X1 :number of ounces of Regular Brute sold annually
◼ X2 :number of ounces of Luxury Brute sold annually
◼ X3 :number of ounces of Regular Chanelle sold annually
◼ X4 :number of ounces of Luxury Chanelle sold annually
◼ X5 :number of pounds of raw material purchased annually
 Objective Function: Maximize the profit:
◼ Profit = revenues from perfume sales - processing costs - costs of
purchasing raw material =7x1 +18x2 +6x3 +14x4 -(4x2 +4x4)-3x5
◼ Max Z = 7x1 +14x2 +6x3 +10x4-3x5

41/68
 Solution
 Constraints:
◼ Constraint 1: No more than 4,000 lb of raw material can be
purchased annually.
◼ x5≤ 4,000
◼ Constraint 2: No more than 6,000 hours of laboratory time can
be used each year.
◼ Total lab time used annually = time used annually to process raw
material+ time used annually to process Luxury Brute + time
used annually to process Luxury Chanelle
◼ =x5 +3x2 +2x4

42/68
 Solution
 LP Model:
Max Z = 7x1 +14x2 +6x3 +10x4-3x5
x5≤ 4,000
3x2 + 2x4 + x5≤ 6,000
xi ≥ 0 for (i = 1, 2, 3, 4, 5)

 Is this model correct?

 The amount of raw material purchased determines the
amount of Brute and Chanelle that is available for sale or
further processing
 x1 +x2 ≤ 3x5 and x3 +x4 ≤ 4x5
43/68
 Solution
 LP Model:
Max Z = 7x1 +14x2 +6x3 +10x4-3x5
x5≤ 4,000
3x2 + 2x4 + x5≤ 6,000
x1 + x2 -3x5≤ 0
x3 + x4 -4x5 ≤0
xi ≥ 0 for (i = 1, 2, 3, 4, 5)

44/68
Multi-period Decision Problems: An Inventory
Model
 One of the most important uses of optimization is in multi-period
planning.
 Partition time into a number of periods.
 Usually distinguished by Inventory or Carry-Over variables.
 Suppose there is a "planning horizon" 𝑇 = 1, 2 … 𝑇 .
 Also suppose there is a known demand 𝑑𝑡 𝑓𝑜𝑟 𝑒𝑣𝑒𝑟𝑦 𝑡 ∈ 𝑇
 Define... :
◼ 𝑃𝑡 : Production level in period 𝑡, ∀𝑡 ∈ 𝑇
◼ 𝐼𝑡 : Inventory level in period 𝑡, ∀𝑡 ∈ 𝑇

45/68
Multi-period Decision Problems

46/68
Example: Shoes Manufacturing
 Complex production process involving many pieces:
◼ Demands
◼ Variable workforce size
◼ Overtime possibilities
◼ Inventory requirements
 Goal:
◼ Plan production of shoes for next several months
◼ Meet forecast demands on time
◼ Hire and/or lay off workers
◼ Make overtime decisions
◼ Objective: minimize total cost
47/68
 Example: Shoes Manufacturing
 Planning horizon T = {1, 2, . . . |T|}. (|T| = 4).
 Meet demand 𝑑𝑡 for shoes in period 𝑡 ∈ 𝑇
◼ d = (3000, 5000, 2000, 1000)
 Initial Shoe Inventory:: 𝐼0 = 500
 Have 𝑊0 = 100 workers currently employed.
 Workers paid α = 1500/𝑚𝑜𝑡𝑛ℎ for working H = 160 hours
 They can work overtime (max of O = 20 hours/worker) and get paid $ 𝛽 =
13/hour.
 It take a = 4 hours of labor and 𝛿 = $15 in raw material costs to produce a
shoe
 Hire-Fire costs η = $1600 to hire a worker and ζ = 2000 to fire a
worker.
 ι =$3 holding cost incurred for each pair of shoes held at the end of the
month.
48/68
Mathematical Model
 Objective :Minimize all costs: labor (regular + overtime),
production, inventory, hiring and firing
 Decision Variables:
◼ 𝑥𝑡 : # of shoes to produce during month t
◼ 𝐼𝑡 : Ending inventory in month t, 𝑡 ∈ 𝑇 ∪ 0 .
◼ 𝑤𝑡 :# of workers available in month t, 𝑡 ∈ 𝑇 ∪ 0 .
◼ 𝑜𝑡 :# of overtime hours used in month t
◼ ℎ𝑡 :# workers hired at the beginning of month t
◼ 𝑓𝑡 :# workers fired at the beginning of month t

49/68
Mathematical Model
 Objective Function:
Raw Material Costs: = ෍ 𝛿𝑥𝑡
𝑡∈𝑇

Regular Labor Costs: = ෍ 𝛼𝑤𝑡

𝑡∈𝑇

Ovetime Labor Costs: = ෍ 𝛽𝑜𝑡

𝑡∈𝑇

Hiring Costs: = ෍ ηℎ𝑡

𝑡∈𝑇

Firing Costs: = ෍ ζ𝑥𝑡

𝑡∈𝑇

Inventory Costs: = ෍ ι𝐼𝑡

𝑡∈𝑇

50/68
 Mathematical Model- Constraints
 Const1: Limit on Monthly Production:
◼ Not given explicitly
◼ Determined by number of workers available and overtime decisions
 𝑎𝑥t ≤ 𝐻𝑤t + 𝑜t 𝑓𝑜𝑟 ∀𝑡 ∈ 𝑇
 Const 2: Upper limit on overtime hours/month:
◼ Depends on how many workers you have
 𝑜t ≤ 𝑂𝑤t 𝑓𝑜𝑟 ∀𝑡 ∈ 𝑇
 Const 3: Demand must be met on time
◼ Equivalent to having nonnegative ending inventory each month (no
backlogging)
 𝐼t ≥ 0 ∀𝑡 ∈ 𝑇 𝑖ç𝑖𝑛
◼ This assumes we have balance between production, demand, and
inventory

51/68
Mathematical Model- Constraints
 Const 4: Balance Equations
◼ Shoes
 𝐼t−1 + 𝑥t = 𝑑t + 𝐼t 𝑓𝑜𝑟 ∀𝑡 ∈ 𝑇
◼ People
 𝑤t = 𝑤t−1 + ℎt − 𝑓t 𝑓𝑜𝑟 ∀𝑡 ∈ 𝑇

52/68
Model

53/68
Example: Shoes Manufacturing - Backlogging
 Suppose you don't have to meet forecast demands in every
period.
 Meeting demand is often too stringent a requirement for
the real-world.
 Demand does not have to be met on time, but it must be
met eventually
 There is a shortage cost θ=$20 per unit per month
backlogged
 How should the minimum cost compare with cost of
earlier model?

54/68
 Modelling Backlogging
 Think of inventory being allowed to go negative.
 Picture still makes sense, since if inventory is negative, you need
to " make up" for it during one of the next periods.
 You can set last period demand 𝐼∣𝑇∣ ≥ 0 to ensure that all demand
is eventually met.
 Cost function 𝐹(𝐼𝑡 ):
𝜄𝐼𝑡 𝑖𝑓 𝐼𝑡 ≥ 0
𝐹(𝐼𝑡 ) = ቊ
−𝜃𝐼𝑡 𝑖𝑓 𝐼𝑡 < 0
 Is 𝐹(𝐼𝑡 ) a linear function of 𝐼𝑡 ?

55/68
 Nonlinear => Linear Transformation
 To model the case where we are minimizing a convex piecewise
linear function (like F(.) or . ), we can introduce a variable for
each piece:
 Write constraints 𝐼𝑡 = 𝐿𝑡 − 𝑆𝑡 ∀𝑡 ∈ 𝑇
◼ Think of this as (Leftover - Shortage)
 Objective gets terms:
𝐹(𝐼𝑡 ) = ෍ ι𝐿𝑡 + 𝜃𝑆𝑡
𝑡∈𝑇
 This trick only works if we are minimizing costs. Then at most one
of 𝐿𝑡 𝑎𝑛𝑑 𝑆𝑡 will ever be positive in an optimal solution.

56/68
Modeling Absolute Value - Objective
 Is 𝑓(𝑥) =∣ 𝑥 ∣ linear?
 Is 𝑓(𝑥) =∣ 𝑥 ∣ convex?
 Trick: Define auxiliary variables (𝑥+ 𝑣𝑒𝑥− ):
◼ Add constraint 𝑥 = 𝑥+ − 𝑥−
◼ Replace ∣ 𝑥 ∣ with 𝑥+ + 𝑥− in objective.
 When will this trick work?
◼ When we are minimizing the absolute value of an
expression.

57/68
Modeling Absolute Value - Constraints
 We would like to model ∣ 𝑎𝑇 𝑥 ∣≤ 𝑏 :
 The Trick:
◼ Replace ∣ 𝑎 𝑇 𝑥 ∣≤ 𝑏 with with two inequalities
◼ 𝑎 𝑇 𝑥 ≤ 𝑏 and −𝑎 𝑇 𝑥 ≤ 𝑏

58/68
Shoes Manufacturing

59/68
Shoes Manufacturing - Backlogging

60/68
 Multi-period Work Scheduling Problem
 CSL is a chain of computer service stores. The number of hours of skilled repair
time that CSL requires during the next five months is as follows:
t Jan Feb March April May

𝒅𝒕 6000 7000 8000 9500 11000

 At the beginning of January, 50 skilled technicians work for CSL. Each skilled
technician can work up to 160 hours per month. To meet future demands, new
technicians must be trained. It takes one month to train a new technician. During
the month of training, a trainee must be supervised for 50 hours by an
experienced technician. Each experienced technician is paid $2,000 a month
(even if he or she does not work the full 160 hours).
 During the month of training, a trainee is paid $1,000 a month. At the end of
each month, 5% of CSL’s experienced technicians quit to join Plum Computers.
Formulate an LP whose solution will enable CSL to minimize the labor cost
incurred in meeting the service requirements for the next five months.

61/68
 Multi-period Work Scheduling Problem
 Decisions to make?
◼ The number of technicians who should be trained during month t (t = 1, 2, 3,
4, 5).
◼ 𝑥𝑡 = number of technicians trained during month t
 Objective Function:
◼ Minimizing total labor cost during the next five months
◼ Total labor cost = cost of paying trainees + cost of paying experienced
technicians
◼ We need number of experienced technicians at the beginning of month t
◼ 𝑦𝑡 = number of experienced technicians at the beginning of month t

min 𝑍 = 1000 𝑥1 + 𝑥2 + 𝑥3 + 𝑥4 + 𝑥5 + 2000( 𝑦1 + 𝑦2 + 𝑦3 + 𝑦4 + 𝑦5 )

62/68
Multi-period Work Scheduling Problem
 Constraints:
◼ Number of available technician hours during month t ≥ Number of technician
hours required during month t
◼ Because each trainee requires 50 hours of experienced technician time, and
each skilled technician is available for 160 hours per month:
Number of available technician hours during month t = 160𝑦𝑡 − 50𝑥𝑡
160𝑦𝑡 − 50𝑥𝑡 ≥ 𝑑𝑡 𝑓𝑜𝑟 𝑡 = 1,2,3,4,5
◼ The number of skilled technicians available at the beginning of any month is
determined by the number of skilled technicians available during the previous
month and the number of technicians trained during the previous month:
◼ Experienced technicians available at beginning of month t = Experienced
technicians available at beginning of month (t -1) + technicians trained during
month (t -1) - experienced technicians who quit during month (t -1)
◼ 𝑦1 = 50
◼ 𝑦𝑡 = 𝑦𝑡−1 + 𝑥𝑡−1 − 0.05𝑦𝑡−1 = 0.95𝑦𝑡−1 + 𝑥𝑡−1 𝑓𝑜𝑟 𝑡 = 2,3,4,5

63/68