2022 IO - Lecture Notes

Introduction to Industrial Organization Course
4th year Mechanical Power
Introduction to Industrial Organization

Definition:
Industrial Organization is the study of the structure of and boundaries between
firms and markets and the strategic interactions of firms.
Examples of topics:
Break-even analysis, forecasting, scheduling, linear programming, sensitivity
analysis, routing, networks, inventory, transportation, depreciation, others.
Forecasting
Definition:
Forecasting is predicting the future demand of the manufactured products and
services based on the past demand information.
Demand
Time
Jan Feb Mar Apr May Jun Jul Aug
IO –Lecture Notes
Page 1
Demand
Season
Importance of forecasting
What are the consequences of bad forecasting?
• Products may be less than the demand; this leads to lost sales
• Products may be more than the demand; this leads to high stock or expired
products
When do we use forecasting?

All the time in our everyday life.
Tools
Scientific tools are used to make the forecasting.
Forecasting classification
Forecasting classification according to the time horizon of forecasting:
Forecasting according to time

horizon
Long term Intermediate term Short term
IO –Lecture Notes
Page 2
Forecasting according to method
Qualitative Quantitative
Market survey Time series
Expert opinion
Regression
Delphi
In time series models, historical data is related to time. In regression models, data
is related to another parameter.
Types of forecasting trends

80
60
Volume
40
(a) Nonlinear trend 20
0
0 5 10 15
Time
IO –Lecture Notes
Page 3
70
60
50
Volume
40
30
(b) Linear trend 20
10
0
0 5 10 15
Time
60
50
Volume
40
30
(c) No trend 20
10
0
0 5 10 15
Time
Time Series Model
For linear trends, the best fitting line is calculated using the least square method.
The equation of the best fitting line is 𝑦 = 𝑎 + 𝑏𝑥
IO –Lecture Notes
Page 4
∑ 𝑥 2 ∑ 𝑦 − ∑ 𝑥 ∑ 𝑥𝑦 𝑛 ∑ 𝑥𝑦 − ∑ 𝑥 ∑ 𝑦
𝑎= 𝑏=
𝑛 ∑ 𝑥 2 − (∑ 𝑥 )2 𝑛 ∑ 𝑥 2 − (∑ 𝑥 )2
where:
y : is the trend value

a : is the intercept of the trend line
b : is the slope of the trend line
x : is the time point
Solved example:
If the sales of a company are as shown in the table, forecast the sales of year 11.
Year (t) 1 2 3 4 5 6 7 8 9 10
Sales (y) 700 724 720 728 740 742 758 750 770 775
Solution:
Year Sales
t2 y2 ty
(t) (y)
1 700 1 490000 700
2 724 4 524176 1448
3 720 9 518400 2160
4 728 16 529984 2912
5 740 25 547600 3700
6 742 36 550564 4452
7 758 49 574564 5306
8 750 64 562500 6000
9 770 81 592900 6930
10 775 100 600625 7750
Sum 55 7407 385 5491313 41358
IO –Lecture Notes
Page 5
385 × 7407 − 55 × 41358

𝑎= = 6.994
10 × 385 − 552
10 × 41358 − 55 × 7407
𝑏= = 7.51
10 × 385 − 552
The trend line is : 𝑦(𝑡) = 6.99.4 + 7.51 𝑡
𝑦(11) = 6.99.4 + 7.51 × 11 = 782.01 𝑢𝑛𝑖𝑡𝑠
Sales of week 11 = 782.01 units
Regression Model
Solved example:
“Healthy Hamburgers” has a chain of 12 stores in northern Illinois. Sales figures
and profits for the stores are given in the following table in million dollars. Check
if a linear model is reasonable. If yes, obtain a regression line for the data and
predict the profit for a store assuming sales of $10 million.
Year 1 2 3 4 5 6 7 8 9 10
Sales 7 2 6 4 14 15 16 12 14 20
Profit 0.15 0.1 0.13 0.15 0.25 0.27 0.24 0.2 0.27 0.44
The correlation coefficient (r) should be calculated twice; once between years and
profit, another between sales and profit.
n ∑ xy − ∑ x ∑ y
rx,y =
√n ∑ x 2 − (∑ x)2 √n ∑ y 2 − (∑ y)2
IO –Lecture Notes
Page 6
Year (t) Sales (x) Profit (y) x2 xy t2 ty y2
1 7 0.15 49 1.1 1 0.2 0.023
2 2 0.10 4 0.2 4 0.2 0.010
3 6 0.13 36 0.8 9 0.4 0.017
4 4 0.15 16 0.6 16 0.6 0.023
5 14 0.25 196 3.5 25 1.3 0.063
6 15 0.27 225 4.1 36 1.6 0.073
7 16 0.24 256 3.8 49 1.7 0.058
8 12 0.20 144 2.4 64 1.6 0.040
9 14 0.27 196 3.8 81 2.4 0.073
10 20 0.44 400 8.8 100 4.4 0.194
55 110 2.20 1522 29 385 14 0.571
𝑟𝑥,𝑦 = 0.919 𝑟𝑡,𝑦 = 0.827
Therefore, the forecasted value should be calculated according to sales (not years)
The best fitting line is: 𝑦(𝑡) = 0.051 + 0.015 𝑡
Profit of $10M = $ 246,000
IO –Lecture Notes
Page 7
Averaging Techniques
Why averaging?
Averaging is used in some cases:
• When historical data has random variations / noise

• When randomness is due to unimportant factors
• To remove randomness and leave “real” variations
• To smooth fluctuations
Moving average (MA)

Solved example:
5-years
2-years
Year moving
Sales moving x2 xy
(x) average
average
(y)
1 45
45.25
2 45.5
47.8
3 50.1 50.64 9 151.92
50.35
4 50.6 53.24 16 212.96
56.3
5 62 57 25 285
60
6 58 61.02 36 366.12
61.15
7 64.3 65.38 49 457.66
67.25
8 70.2 68.1 64 544.8
71.3
9 72.4
74
10 75.6
IO –Lecture Notes
Page 8
80
75
70
65
60
55
50
45
40
0 2 4 6 8 10 12
Original 2-years M.A. 5-years M.A.
Best fitting line for MA5 is: y (x) = 39.15+ 3.649 x
y (9) = 71.991 units

64.3 + 70.2 + 72.4 + 75.6 + 𝑠𝑎𝑙𝑒𝑠(11)
𝑦(9) =
5
𝑆𝑎𝑙𝑒𝑠 (11) = 77.455 𝑢𝑛𝑖𝑡𝑠
Values of (N)
Aspect Lower value of N Higher value of N

Smoothing less smoothing greater smoothing
Responsiveness more responsive to changes less responsive to changes
Uses appropriate for changing pattern appropriate for stable demand

or if it is important to identify
short-term fluctuations
IO –Lecture Notes
Page 9
90
80
70
60
50
40
0 2 4 6 8 10 12
Original 3-years MA 5-years MA Least sq.
Exponential smoothing
- For short term forecasting (daily)
- Easy to use and understand
- The new forecast depends on the previous forecast plus a percentage of the
forecast error.
𝐹𝑡+1 = 𝐹𝑡 + 𝛼 (𝐴𝑡 − 𝐹𝑡 )
𝐴𝑡 : Actual demand of (t)

𝐹𝑡 : Forecasted demand of (t)
𝛼: Percentage
𝐹𝑡+1 = 𝐴𝑡 ∙ 𝛼 + 𝐹𝑡 (1 − 𝛼)
Act. Sales
Year (t) Forecast (Ft)
(yt)
1 A1 F1
2 A2
… …
T-1 AT-1
T AT
T+1
IO –Lecture Notes
Page 10
Solved example:
For F1 = 85, and α = 0.15, forecast the sales of year (9).
Year (t) Sales (At) Forecast (Ft)
1 90 85
2 100 85.75
3 107 87.888
4 113 90.748
5 123 94.09
6 136 98.43
7 144 104.07
8 155 110.06
9 116.8
Values of (α)
Aspect α < 0.5 α > 0.5
Responsiveness Slow responsiveness to adjust Fast responsiveness to adjust

errors errors
Smoothness Higher smoothing Lower smoothing
Potential / Higher potential for earlier Higher potential for later days
weight days
Uses Used for less stable cases Used for more stable cases
IO –Lecture Notes
Page 11
80
75
70
65
60
55
50
45
40
35
0 2 4 6 8 10 12
Original index 0.15 index 0.6
Values of (𝐹1 )
Naive approach 𝐹2 = 𝐴1
𝐴1 + 𝐴2 + ⋯ + 𝐴𝑛
Subjective estimate 𝐹1 =
𝑛
𝐴1 + 𝐴2 + 𝐴3
Practical approach 𝐹4 =
3
IO –Lecture Notes
Page 12
Seasonal decomposition
This is a method to forecast the sales of a product (or service) that faces
seasonality. The basic concept of the method is to smooth and de-seasonalize the
data, so that the data can follow a linear trend. After forecasting, the values
obtained are seasonalized again.
Solved example:
The quarterly unit demand of a product are given below:
Year Winter Spring Autumn Summer
1 190 350 250 200
2 250 400 300 150
3 300 340 260 180
4 280 400 220 200
5 350 420 340 240
Determine the seasonal index, and establish a forecast for each quarter of the next
year.
The steps are as shown in the following table: the fourth column represents the
four quarter moving average. The fifth column is the centred moving average.
When comparing these values to the values of original data, the seasonal index
(seasonal component) can be calculated as in the sixth column. The average
seasonal component for each season is calculated given all its values in column 6.
These average values are added and repeated in column 7. Knowing the original
sales and the average seasonal component for each season, the de-seasonalized
values are calculated and added in column 8.
IO –Lecture Notes
Page 13
4 Quarters Centered Average

Seasonal index De-seasonalized
Year Quarter Sales moving moving Seasonal
St sales
average average index
1 1 190 1.037 183.220829
2 350 1.358 257.731959

247.5
3 250 255 0.98039216 0.949 263.435195
262.5
4 200 268.75 0.744186047 0.655 305.5281236
275
2 1 250 281.25 0.888888889 1.037 241.0471454
287.5
2 400 281.25 1.422222222 1.358 294.6449027
275
3 300 281.25 1.066666667 0.949 316.2595488
287.5
4 150 280 0.535714286 0.655 229.1460927
272.5
3 1 300 267.5 1.121495327 1.037 289.2565745
262.5
2 340 266.25 1.276995305 1.358 250.4481673
270
3 260 267.5 0.971962617 0.949 274.091609
265
4 180 272.5 0.660550459 0.655 274.9753112
280
4 1 280 275 1.018181818 1.037 269.9728028
270
2 400 272.5 1.467889908 1.358 294.6449027
275
3 220 283.75 0.775330396 0.949 231.9236691
292.5
4 200 295 0.677966102 0.655 305.5281236
297.5
5 1 350 312.5 1.12 1.037 337.4660035
327.5
2 420 332.5 1.263157895 1.358 309.3771479
337.5
3 340 0.949 358.4274887
4 240 0.655 366.6337483
IO –Lecture Notes
Page 14
To calculate the values of the 7th column:
0.8889 + 1.1215 + 1.0182 + 1.12

𝑆1 = = 1.037
4
1.4222 + 1.2770 + 1.4679 + 1.2632

𝑆2 = = 1.358
4
0.9804 + 1.0667 + 0.9720 + 0.7753

𝑆3 = = 0.949
4
0.7442 + 0.5357 + 0.6606 + 0.6780

𝑆4 = = 0.655
4
The best fitting line is: y = a + b (t)

𝑎 = 231.42 , b = 4.8676
y = 231.42 + 4.8676 (t)

This equation is used to forecast the de-seasonalized sales of the next four quarters
(t = 21, 22, 23, and 24).
The obtained values are used together with the seasonal component to calculate the
expected sales according to their seasons.
IO –Lecture Notes
Page 15
Deseasonalized
Quarter x2 xy
sales
1 183.220829 1 183.220829
2 257.731959 4 515.463918
3 263.5496 9 790.6489
4 305.5281 16 1222.112
5 241.0471 25 1205.236
6 294.6449 36 1767.869
7 316.2595 49 2213.817
8 229.1461 64 1833.169
9 289.2566 81 2603.309
10 250.4482 100 2504.482
11 274.0916 121 3015.008
12 274.9753 144 3299.704
13 269.9728 169 3509.646
14 294.6449 196 4125.029
15 231.9237 225 3478.855
16 305.5281 256 4888.45
17 337.466 289 5736.922
18 309.3771 324 5568.789
19 358.4275 361 6810.122
20 366.6337 400 7332.675
Sum 210 5653.931 2870 62604.67
The forecasted sales are the de-seasonalized sales multiplied by the index
Forecasted
Year Quarter t Deasonalized sales St
seasonalized sales
1 21 333.6396 1.039 346.65
2 22 338.5072 1.357 459.36
6
3 23 343.3748 0.949 325.86
4 24 348.2424 0.655 228.10
IO –Lecture Notes
Page 16
Inventory Control
Definition
Inventory is any idle goods, materials, or assets carried on hand by a company to
be used in the future
Types
• Raw material and purchased parts
• Components
• Work-in-process
• Finished goods
• Supplies
Control methods
Type Method Application
Loose control e.g. Two bin system Wires, bolts, nuts
Medium control e.g. Pin card Sheets, frames
Tight control e.g. Periodic control Motors
IO –Lecture Notes
Page 17
Determining valuable inventory – ABC inventory analysis

It is a method to categorize the inventory items into three categories according to
their value.
Solved example:
For the following items in the inventory, make an ABC inventory analysis.
Item Projected annual Unit cost

code usage (units) (dollars)
1A 1,500 5.00
2B 6,450 20.00
3C 5,000 45.00
4D 200 12.50
5E 20,000 35.00
6F 84 250.00
7G 800 80.00
8H 300 5.00
9J 10,000 35.00
10K 2,000 65.00
11L 5,000 25.00
12M 3,250 125.00
13N 9,000 0.50
14P 2,900 10.00
15Q 800 15.00
16R 675 200.00
17S 1,470 100.00
18T 8,200 15.00
19U 1,250 0.16
20W 2,500 0.20
IO –Lecture Notes
Page 18
Calculate the annual usage in terms of money for each item
Item Projected annual Unit cost Projected annual

number usage (units) (dollars) dollar usage (dollars)
1A 1,500 5.00 7,500

2B 6,450 20.00 129,000
3C 5,000 45.00 225,000
4D 200 12.50 2,500
5E 20,000 35.00 700,000
6F 84 250.00 21,000
7G 800 80.00 64,000
8H 300 5.00 1,500
9J 10,000 35.00 350,000
10K 2,000 65.00 130,000
11L 5,000 25.00 125,000
12M 3,250 125.00 406,250
13N 9,000 0.50 4,500
14P 2,900 10.00 29,000
15Q 800 15.00 12,000
16R 675 200.00 135,000
17S 1,470 100.00 147,000
18T 8,200 15.00 123,000
19U 1,250 0.16 200
20W 2,500 0.20 500
IO –Lecture Notes
Page 19
1. Arrange in a descending order in ranks
projected annual
Cumulative dollar Cumulative
Rank Item number dollar usage
usage (dollars) percentage
(dollars)
1 5E 700,000 700,000 27.79%
2 12M 406,250 1,106,250 42.34%
3 9J 350,000 1,456,250 55.73%
4 3C 225,000 1,681,250 64.34%
5 17S 147,000 1,828,250 69.97%
6 16R 135,000 1,963,250 75.14%
7 10K 130,000 2,093,250 80.11%
8 2B 129,000 2,222,250 85.05%
9 11L 125,000 2,347,250 89.83%
10 18S 123,000 2,470,250 94.54%
11 7G 64,000 2,534,250 96.99%
12 14P 29,000 2,563,250 98.10%
13 6F 21,000 2,584,250 98.90%
14 15Q 12,000 2,596,250 99.36%
15 1A 7,500 2,603,750 99.65%
16 13N 4,500 2,608,250 99.82%
17 4D 2,500 2,610,750 99.92%
18 8H 1,500 2,612,250 99.97%
19 20W 500 2,612,750 99.99%
20 19U 200 2,612,950 100.00%
IO –Lecture Notes
Page 20
2. Plot Item percentage versus cumulative percentage
Item Cumulative
percentage percentage
5% 27.79%
10% 42.34%
15% 55.73%
20% 64.34%
25% 69.97%
30% 75.14%
35% 80.11%
40% 85.05%
45% 89.83%
50% 94.54%
55% 96.99%
60% 98.10%
65% 98.90%
70% 99.36%
75% 99.65%
80% 99.82%
85% 99.92%
90% 99.97%
95% 99.99%
100% 100.00%
IO –Lecture Notes
Page 21
Graph:
120%
100%
Percent dollar usage
80%
60%
40%
B C
20%
A
0%
0 20 40 60 80 100 120
Percent of items
Another representation:
Value Percentage Category Item Percentage
25%
70% A
35%
B
27% 40%
C
Tight inventory cost calculation

Two types of costs are associated with the inventory:
1. Procurement Cost (Ordering cost)

2. Holding cost (Carrying cost)
IO –Lecture Notes
Page 22
1. Procurement cost (Ordering cost); the cost of placing an order.

Examples:
Shipping and handling; Documents and paper work; Insurance; Vendor selection;
Evaluation; Inspection; tendering (order processing)
Calculation:
For product x:
Annual requirements, R =1,000 units/year
Order size, Q = 200 units/order
Procurement cost per order, p = 20 EGP/order

1000
𝑃𝑟𝑜𝑐𝑢𝑟𝑒𝑚𝑒𝑒𝑛𝑡 𝑐𝑜𝑠𝑡 = 20 ×
200
𝑅
𝑃𝑟𝑜𝑐𝑢𝑟𝑒𝑚𝑒𝑒𝑛𝑡 𝑐𝑜𝑠𝑡 = 𝑝 ×
𝑄
To decrease the procurement cost, it is required to increase the order size (Q).
Chart:
3500
3000
Ordering cost (L.E.)
2500
2000
1500
1000
500
0
0 10 20 30 40 50 60
Order Quantity (units/order)
IO –Lecture Notes
Page 23
2. Holding cost (Carrying cost); the cost of keeping units inside the
inventory.
Examples:
Handling; Storage equipment; Overheads (Electricity, heat, water); Guarding;

Insurance; Taxes; Obsolescence
120
100
Q
Quantity on hold(units)
80
60
Q /2
40
20
0
0 50 100 150 200 250 300 350 400
Time unit
Calculation:
For product x,
Order size Q (200 units/order)
Holding cost h (1 EGP/unit/year)
200
𝐻𝑜𝑙𝑑𝑖𝑛𝑔 𝑐𝑜𝑠𝑡 = 1 ×
2
𝑄
𝐻𝑜𝑙𝑑𝑖𝑛𝑔 𝑐𝑜𝑠𝑡 = ℎ ×
2
To decrease the holding cost, it is required to decrease the order size (Q)
IO –Lecture Notes
Page 24
Chart:
1200
1000
800
Holding cost (L.E.)
600
400
200
0
0 10 20 30 40 50 60
Order quantity (units/order)
Both costs; the procurement cost, and the holding cost, are conflicting. In order to
decrease the total cost, it is required to find the optimum value of Q that minimizes
the total cost. This is obtained through differentiation. It is the value of Q where
both costs are equal.
Total Cost
3500
3000
2500
2000
Cost
1500
1000
500
0
0 10 20 30 40 50 60
Order Quantity
order cost hold cost sum
When procurement cost is equal to the carrying cost, the inventory cost is
minimized. This value of Q is called “Economic Order Quantity”, and it notified as
E.O.Q.
IO –Lecture Notes
Page 25
𝑅 𝑄
∵𝑝 =ℎ
𝑄 2
2𝑝𝑅
∴𝑄=√
ℎ
2𝑝𝑅
∴ 𝐸. 𝑂. 𝑄. = √
ℎ
Total Annual cost
Puchasing cost
Total Inventory cost
(𝑅 × 𝑢)
Holding cost Procurement cost

𝑄 𝑅
(ℎ × ) (𝑝 × )
2 𝑄
Total annual cost

𝑅 𝑄
𝑇=𝑝× +ℎ× +𝑅×𝑢
𝑄 2
𝑅 𝑄
𝑇=𝑝× + 𝐼. 𝑢 × + 𝑅 × 𝑢
𝑄 2
Where:
R : Annual requirements : (units)

Q : Order size or Lot size : (units/order) or (units/lot)
IO –Lecture Notes
Page 26
h : Holding cost : (EGP/unit/year)

p : Procurement cost : (EGP/order)
T : Total annual inventory cost : (EGP)
u : Unit purchasing cost : (EGP/unit)
I : Holding cost percentage of purchasing cost :%
Solved example:
Given Required
R=50,000 (units/ year) E.O.Q,
h = 1.5 (EGP/100 parts/year) Total annual cost,
p = 4 (EGP/order) Order rate for 300 working days /year
u = 0.2 (EGP/unit)
2𝑝𝑅 2 × 4 × 50,000
𝐸. 𝑂. 𝑄. = √ = √ = 5164 𝑢𝑛𝑖𝑡𝑠/𝑜𝑟𝑑𝑒𝑟
ℎ 1.5/100
𝑅 𝑄
𝑇𝑜𝑡𝑎𝑙 𝑎𝑛𝑛𝑢𝑎𝑙 𝑐𝑜𝑠𝑡 = 𝑝 × +ℎ× +𝑅×𝑢
𝑄 2
50,000 1.5 5164
=4× + × + 50,000 × 0.2
5164 100 2
= 10,077.45 𝐿. 𝐸 / 𝑦𝑒𝑎𝑟
𝑅 50,000
𝑁𝑜. 𝑜𝑓 𝑜𝑟𝑑𝑒𝑟𝑠 = = = 9.68 𝑜𝑟𝑑𝑒𝑟𝑠 / 𝑦𝑒𝑎𝑟
𝑄 5164
𝑄 5164
𝑂𝑟𝑑𝑒𝑟 𝑟𝑎𝑡𝑒 = = × 300 = 30.98 𝑑𝑎𝑦𝑠 / 𝑜𝑟𝑑𝑒𝑟
𝑅 50,000
IO –Lecture Notes
Page 27
Re-ordering point (R.O.P)

𝑅. 𝑂. 𝑃. = 𝐶𝑜𝑛𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛 𝑟𝑎𝑡𝑒 × 𝐿𝑒𝑎𝑑 𝑡𝑖𝑚𝑒
FOR THE SAME TIME UNITS
Lead time: The time the order takes from placing it till it is received in the
inventory.
3500
3000
Quantity on hold (units)
2500
2000
1500
1000
500
0
0 5 10 15 20 25 30 35
Time
Regular Higher consumption Safety stock
Safety Stock
• Quantity of safety stock depends on the risk of stock-out.

• Risk of stock-out depends on the probability of stock-out.
IO –Lecture Notes
Page 28
Description Without safety stock With safety stock

𝐸. 𝑂. 𝑄. = 3000 units 3000 units
𝑄𝑠𝑎𝑓𝑒𝑡𝑦 = 0 units 1000 units
𝑄𝑚𝑎𝑥 = 3000 units 4000 units
𝑄𝑚𝑖𝑛 = 0 units 1000 units
𝑄𝑎𝑣𝑒𝑟𝑎𝑔𝑒 = 1500 units 2500 units
Consumption of stock in inventory

4500
4000
3500
Quantity on hold (units)
3000
2500
2000
1500
1000
500
0
0 20 40 60 80 100 120
Time
IO –Lecture Notes
Page 29
Economic Production Quantity (E.P.Q)

The EPQ model assumes the company will produce its own quantity. Therefore,
the orders are available or received in an incrementally manner while the products
are being produced. It is required to decide the quantities to be ordered for
production at each order.
4000
Cycle time
Qp
3000
Qholdmax
Quantity
2000
1000
0
0 5 10 15 20 25 30 35 40
Tp Tc
Time
■ Production only (P). It is an imaginary curve
● Production & Consumption (P - C)

■ Consumption only (-C)
• During the production, consumption occurs at a slower rate.

• Inventory builds up.
• Inventory continues to build up until production stops.
• When production stops, inventory level is maximum.
• Inventory level decreases due to consumption only.
• When inventory is zero, the cycle repeats.
• Setup cost corresponds ordering cost. This is the cost of production in
batches (e.g. cleaning, adjustment, changing tools and fixtures).
From the graph:

𝑃−𝐶
𝑄ℎ𝑜𝑙𝑑𝑚𝑎𝑥 = 𝑄𝑝 ( )
𝑃
IO –Lecture Notes
Page 30
𝐷 𝑄ℎ𝑜𝑙𝑑𝑚𝑎𝑥
𝑆𝑒𝑡𝑢𝑝 𝑐𝑜𝑠𝑡 = 𝑆 × , 𝐶𝑎𝑟𝑟𝑦𝑖𝑛𝑔 𝑐𝑜𝑠𝑡 = 𝐻 ×
𝑄𝑝 2
𝐷 𝑄ℎ𝑜𝑙𝑑𝑚𝑎𝑥
𝑇𝑜𝑡𝑎𝑙 𝑐𝑜𝑠𝑡 (𝑇) = 𝑆 × +𝐻×
𝑄𝑝 2
𝐷 𝑄𝑝 𝑃 − 𝐶
𝑇=𝑆× +𝐻× ( )
𝑄𝑝 2 𝑃
𝜕𝑇
=0
𝜕𝑄𝑝
2𝐷𝑆 𝑃
∴ 𝑄𝑝 = 𝐸. 𝑃. 𝑄. = √ √
𝐻 𝑃−𝐶
𝑄𝑝 This is also the time taken for

𝑇𝑖𝑚𝑒 𝑡𝑎𝑘𝑒𝑛 𝑓𝑜𝑟 𝑜𝑛𝑒 𝑝𝑟𝑜𝑑𝑢𝑐𝑡𝑖𝑜𝑛 𝑟𝑢𝑛 (𝑇𝑝) =
𝑃 production and consumption
𝑄ℎ𝑜𝑙𝑑𝑚𝑎𝑥
𝑇𝑖𝑚𝑒 𝑡𝑎𝑘𝑒𝑛 𝑓𝑜𝑟 𝑐𝑜𝑛𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛 𝑜𝑛𝑙𝑦 (𝑇𝑐) =
𝐶
𝑄𝑝 𝑄ℎ𝑜𝑙𝑑𝑚𝑎𝑥
𝐶𝑦𝑐𝑙𝑒 𝑡𝑖𝑚𝑒 𝑓𝑜𝑟 𝑖𝑛𝑣𝑒𝑛𝑡𝑜𝑟𝑦 = 𝑇𝑝 + 𝑇𝑐 = +
𝑃 𝐶
𝐷
No. of runs (N) = ( 𝑟𝑢𝑛𝑠/𝑦𝑒𝑎𝑟)
𝑄𝑝
𝑊𝑜𝑟𝑘𝑖𝑛𝑔 𝑑𝑎𝑦𝑠/𝑦𝑒𝑎𝑟 𝑄𝑝 × 𝑊𝑜𝑟𝑘𝑖𝑛𝑔 𝑑𝑎𝑦𝑠/𝑦𝑒𝑎𝑟

Cycle time= =
𝑁 𝐷
Solved example
A toy manufacturer uses 48,000 rubber wheels per year for its popular truck series.
The firm makes its own wheels at a rate of 800 per day. Carrying cost is $1 per
wheel a year. Setup cost for a production run of wheels is $45. The firm operates
240 days per year. Determine:
• Optimal production size

• Minimum total annual holding cost
• Cycle time for inventory
• Production time (Run time)
IO –Lecture Notes
Page 31
Given:
D = 48000 wheels / year P = 800 wheels / day H = $1 / wheel/year
S = $45 / run Working days = 240 days / year
Solution:
𝐷 48000
𝐶= = = 200 𝑤ℎ𝑒𝑒𝑙𝑠 / 𝑑𝑎𝑦
𝑊𝑜𝑟𝑘𝑖𝑛𝑔 𝑑𝑎𝑦𝑠 240
2𝐷𝑆 𝑃
𝑂𝑝𝑡𝑖𝑚𝑎𝑙 𝑝𝑟𝑜𝑑𝑢𝑐𝑡𝑖𝑜𝑛 𝑠𝑖𝑧𝑒 = 𝑄𝑝 = √ √
𝐻 𝑃−𝐶
2 × 48000 × 45 800
√ √ = 2400 𝑤ℎ𝑒𝑒𝑙𝑠 / 𝑟𝑢𝑚
1 800 − 200
48000 2400 800 − 200

𝑀𝑖𝑛𝑖𝑚𝑢𝑚 𝑡𝑜𝑡𝑎𝑙 𝑎𝑛𝑛𝑢𝑎𝑙 𝑖𝑛𝑣𝑒𝑛𝑡𝑜𝑟𝑦 𝑐𝑜𝑠𝑡 = 45 × +1× ( )
2400 2 800
= $1800 / 𝑦𝑒𝑎𝑟
𝑄𝑝 2400
𝑃𝑟𝑜𝑑𝑢𝑐𝑡𝑖𝑜𝑛 𝑡𝑖𝑚𝑒 / 𝑟𝑢𝑛 (𝑇𝑝) = = = 3 𝑑𝑎𝑦𝑠/𝑟𝑢𝑛
𝑃 800
𝑃−𝐶 800 − 200
𝑄ℎ𝑜𝑙𝑑𝑚𝑎𝑥 𝑄𝑝 ( 𝑃 ) 2400 ( 800 )
𝐶𝑜𝑛𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛 𝑡𝑖𝑚𝑒 (𝑇𝑐) = = = = 9 𝑑𝑎𝑦𝑠
𝐶 𝐶 200
𝐶𝑦𝑐𝑙𝑒 𝑡𝑖𝑚𝑒 𝑓𝑜𝑟 𝑖𝑛𝑣𝑒𝑛𝑡𝑜𝑟𝑦 = 𝑇𝑝 + 𝑇𝑐 = 3 + 9 = 12 𝑑𝑎𝑦𝑠
** Cycle time for inventory can be also calculated through the number of runs
per year (N)
𝐷 48000
No. of runs (N) = = = 20 𝑟𝑢𝑛𝑠/𝑦𝑒𝑎𝑟
𝑄𝑝 2400
𝑊𝑜𝑟𝑘𝑖𝑛𝑔 𝑑𝑎𝑦𝑠/𝑦𝑒𝑎𝑟 240

Cycle time= = = 12 𝑑𝑎𝑦𝑠
𝑁 20
IO –Lecture Notes
Page 32
Where: A: Consumption rate (as given in R), B: Production rate

𝐵−𝐴
𝑄ℎ𝑜𝑙𝑑 = 𝑄 ( )
𝐵
𝑅 𝑄 𝐵−𝐴
𝑇 =𝑝× +ℎ× ( )
𝑄 2 𝐵
𝜕𝑇
=0
𝜕𝑄
2𝑝𝑅 𝐵
∴𝑄=√ √
ℎ 𝐵−𝐴
2𝑝𝑅 𝐵
∴ 𝐸. 𝑃. 𝑄. = √ √
ℎ 𝐵−𝐴
2𝑝𝑅 𝐴
𝐸. 𝑃. 𝑄. = √ , 𝜔=
ℎ(1 − 𝜔) 𝐵
Quantity discounts
• Quantity discounts are price reduction for large offers.
• They encourage customers to buy in large quantities.
• They result in large quantities in stock.
IO –Lecture Notes
Page 33
• The objective is to minimize the carrying cost, ordering cost, and purchasing
cost.
Quantity discount is reduction in the value of (u) according to the quantity

purchased. This can result in two cases; either the holding cost (h) is independent
of the unit price (u); or the holding cost (h) depends on the unit price (u). Each case
has its own calculations.
Holding cost (h) is independent of (u)

If there are three offers for purchasing prices (u) for three categories of purchased
quantities, then there are three possible values for the total cost (T).
𝑅 𝑄
𝑇0 = 𝑝 ×
+ℎ×
𝑄 2
𝑅 𝑄
𝑇1 = 𝑝 × + ℎ × + 𝑅 × 𝑢1
𝑄 2
𝑅 𝑄
𝑇2 = 𝑝 × + ℎ × + 𝑅 × 𝑢2
𝑄 2
𝑅 𝑄
𝑇3 = 𝑝 × + ℎ × + 𝑅 × 𝑢3
𝑄 2
For all the previous costs, the economic order quantity is the same according to:
2𝑝𝑅
∴ 𝐸. 𝑂. 𝑄. = √
ℎ
As shown in the following figure, all the costs have the same EOQ, however, each
cost is valid for a certain category of cost only. The dotted lines represent the limits
of the quantities purchased where:
• Highest unit price (u3) is valid for buying up to 90 units.

• Moderate unit price (u2) is valid for buying more than 90 units and up to
200 units
• Least unit price (u1) is valid for buying more than 200 units.
By logic, the lower cost is offered for larger quantities.
IO –Lecture Notes
Page 34
2500
2000
1500
Cost
1000
500
0
0 100 200 300 400 500 600
Order Quantity
Inventory cost TC3 TC2 TC1
Then, the previous figure should be re-drawn showing the ranges of validity for
each price offer as shown in the following figure. Valid ranges are solid lines.
Dotted lines are the invalid ranges.
2500
2300
2100
1900
1700
Cost
1500
1300
1100
900
700
500
0 100 200 300 400 500 600
Order Quantity
If the problem is to be solved graphically, it is clear that the least cost occurs when
ordering 201 units / order. It might seem that all the time, the least cost occurs for
the lowest unit price. This is not true. The following figure shows the same
previous problem, with the third price category valid for more than 420 units.
IO –Lecture Notes
Page 35
2500
2000
Cost
1500
1000
500
0 100 200 300 400 500 600
Order Quantity
From the previous figure, it is clear that the least cost occurs at ordering the
economic order quantity, within the category of the second unit price offer.
The ranges of the price offers control detecting the ordering size of the least cost.
Holding cost (h) depends on (u)

In this case, the three possible values for the total cost (T) are:
𝑅 𝑄
𝑇0 = 𝑝 × +ℎ×
𝑄 2
𝑅 𝑄
𝑇1 = 𝑝 × + 𝐼𝑢1 × + 𝑅 × 𝑢1
𝑄 2
𝑅 𝑄
𝑇2 = 𝑝 × + 𝐼𝑢2 × + 𝑅 × 𝑢2
𝑄 2
𝑅 𝑄
𝑇3 = 𝑝 × + 𝐼𝑢3 × + 𝑅 × 𝑢3
𝑄 2
In this case, the economic order quantity varies for each price category (u):
2𝑝𝑅
∴ 𝐸. 𝑂. 𝑄. = √
𝐼𝑢
To detect the order quantity of least cost, we need to investigate the minimum cost
for each price category.
IO –Lecture Notes
Page 36
For a certain price category, the point of least cost might either occur at one of the
end values, or at the point of EOQ (in case that the EOQ lies within the validity
range of this price category). This is shown in the next figure.
1400
1300
1200
1100
1000
Cost
900
800
700
600
500
400
0 200 400 600 800 1000 1200 1400
Order Quantity
V1 V1 V2 V2 V2 V3 V3
Solved example:
Surge electric uses approximately 4,000 toggle switches a year. Switches are
priced as given in the table. It costs $30 to place an order. Carrying costs are 40%
of the purchase price per unit on annual basis. Determine the optimal quantity, and
the total annual cost.
Quantity range Unit price (u)

1 – 499 $0.90
500 – 999 $0.85
1,000 or more $0.80
Given:
D = 4,000 (switches/ year) , H = 0.4u , S = $30 (/ order)
IO –Lecture Notes
Page 37
2𝐷𝑆 2 × 4000 × 30 774.6

𝐸𝑂𝑄 = √ =√ =
𝐼𝑢 0.4𝑢 ξ𝑢
Unit price 774.6 (Qo) Point of

Quantity range 𝐸𝑂𝑄 =
(u) ξ𝑢 calculation
1 – 499 $0.90 816.5 499
500 – 999 $0.85 840 840
1,000 or more $0.80 866 1000
@ each point of calculation, the total cost can be calculated through the equation:
𝐷 𝑄 120,000
𝑇 =𝐷∙𝑢+ × 𝑆 + × 𝐼𝑢 = 4000𝑢 + + 0.2𝑄𝑜 𝑢
𝑄 2 𝑄𝑜
Quantity Unit price 774.6 (Qo) Point of

𝐸𝑂𝑄 = Cost
range (u) ξ𝑢 calculation
1 – 499 $0.90 816.5 499 $3,930
500 – 999 $0.85 840 840 $3,686
1,000 or more $0.80 866 1000 $3,480
Optimal order quantity = 1000 units / order
Optimal cost = $3480
Inventory management rules

There are several rules that manage the flow of stock out of the inventory. Two
main rules are discussed here.
IO –Lecture Notes
Page 38
First In First Out (FIFO) rule:

As the name indicates, the units that entered the stock first should leave first. This
is essential in the perishable commodities that have an expiration date.
First Expired First Out (FEFO) rule:

This rule is more accurate in keeping the freshest units in the stock in case of
perishable commodities.
Depreciation
Definition:
Depreciation refers to two very different but related concepts:
1. The decrease in value of assets, and

2. The allocation of the cost of assets to periods in which the assets are used.
The former affects values of businesses and entities. The latter affects net income.
Generally the cost is allocated, as depreciation expense, among the periods in
which the asset is expected to be used.
• If the asset is sold at a sales price exceeding the depreciated value (net book
value), the excess would be considered a gain and would be recognized as
ordinary income by the tax office.
• If the sales price is ever less than the book value, the resulting capital loss is
tax deductible.
• If the sale price were ever more than the original book value, then the gain
above the original book value is recognized as a capital gain.
There are several methods for calculating depreciation, generally based on either
the passage of time or the level of activity (or use) of the asset.
IO –Lecture Notes
Page 39
Straight-line depreciation
Straight-line depreciation is the simplest and most-often-used technique, in which
the company estimates the salvage value of the asset at the end of the period during
which it will be used to generate revenues (useful life) and will expense a portion
of original cost in equal increments over that period. The salvage value is an
estimate of the value of the asset at the time it will be sold or disposed of; it may be
zero or even negative. Salvage value is also known as scrap value or residual
value.
𝐶𝑜𝑠𝑡 𝑜𝑓 𝑎𝑠𝑠𝑒𝑡 − 𝑆𝑐𝑟𝑎𝑝 𝑣𝑎𝑙𝑢𝑒
𝐴𝑛𝑛𝑢𝑎𝑙 𝐷𝑒𝑝𝑟𝑒𝑐𝑖𝑎𝑡𝑖𝑜𝑛 𝐸𝑥𝑝𝑒𝑛𝑠𝑒 =
𝑈𝑠𝑒𝑓𝑢𝑙 𝑙𝑖𝑓𝑒 𝑜𝑓 𝑎𝑠𝑠𝑒𝑡 (𝑦𝑒𝑎𝑟𝑠)
Book value at the beginning of the first year of depreciation is the original cost of
the asset. At any time book value equals original cost minus accumulated
depreciation.
Book value = original cost − accumulated depreciation
Solved example:
A vehicle that depreciates over 5 years, is purchased at a cost of EGP 1,000, and
will have a salvage value of EGP 100. Find the book value of the vehicle at the
end of each year.
1000 − 100
𝐴𝑛𝑛𝑢𝑎𝑙 𝐷𝑒𝑝𝑟𝑒𝑐𝑖𝑎𝑡𝑖𝑜𝑛 𝐸𝑥𝑝𝑒𝑛𝑠𝑒 = = 𝐸𝐺𝑃 180
5
Book value at Depreciation Book value at

Year
beginning of year (EGP) expense (EGP) end of year (EGP)
1,000
1 180 820
(original cost)
2 820 180 640
3 640 180 460
4 460 180 280
IO –Lecture Notes
Page 40
100
5 280 180
(scrap value)
Units-of-production depreciation method

Under the units-of-production method, useful life of the asset is expressed in terms
of the total number of units expected to be produced:
𝐶𝑜𝑠𝑡 𝑜𝑓 𝑎𝑠𝑠𝑒𝑡 − 𝑆𝑐𝑟𝑎𝑝 𝑣𝑎𝑙𝑢𝑒
𝐴𝑛𝑛𝑢𝑎𝑙 𝐷𝑒𝑝𝑟𝑒𝑐𝑖𝑎𝑡𝑖𝑜𝑛 = × 𝐴𝑐𝑡𝑢𝑎𝑙 𝑝𝑟𝑜𝑑𝑢𝑐𝑡𝑖𝑜𝑛
𝐸𝑠𝑡𝑖𝑚𝑎𝑡𝑒𝑑 𝑡𝑜𝑡𝑎𝑙 𝑝𝑟𝑜𝑑𝑢𝑐𝑡𝑖𝑜𝑛
Solved example:
An asset has original cost $70,000, salvage value $10,000, and is expected to
produce 6,000 units. The machine is used to produce 1,000 units in the first year.
Production increases by 100 units each year. Find the book value of the machine
at the end of each year, until it reaches its salvage value.
70,000 − 10,000
𝐷𝑒𝑝𝑟𝑒𝑐𝑖𝑎𝑡𝑖𝑜𝑛 𝑝𝑒𝑟 𝑢𝑛𝑖𝑡 = = $10
6,000
The table below illustrates the units-of-production depreciation schedule of the
asset.
Book value at
Units of Depreciation Depreciation Book value at
Yr beginning of
production cost per unit ($) expense ($) end of year ($)
year ($)
70,000
1 1000 10 10,000 60,000
(original cost)
2 60,000 1100 10 11,000 49,000
3 49,000 1200 10 12,000 37,000
4 37,000 1300 10 13,000 24,000
IO –Lecture Notes
Page 41
$10,000
5 24,000 1400 10 14,000
(Scrap value)
Depreciation stops when book value is equal to the scrap value of the asset. In the
end, the sum of accumulated depreciation and scrap value equals the original cost.
Activity depreciation
Activity depreciation methods are not based on time, but on a level of activity.
This could be miles driven for a vehicle, or a cycle count for a machine. When the
asset is acquired, its life is estimated in terms of this level of activity.
𝐶𝑜𝑠𝑡 𝑜𝑓 𝑓𝑖𝑥𝑒𝑑 𝑎𝑠𝑠𝑒𝑡 − 𝑠𝑐𝑟𝑎𝑝 𝑣𝑎𝑙𝑢𝑒

𝑃𝑒𝑟 𝑎𝑐𝑡𝑖𝑣𝑖𝑡𝑦 𝑑𝑒𝑝𝑟𝑒𝑐𝑖𝑎𝑡𝑖𝑜𝑛 𝑟𝑎𝑡𝑒 =
𝐸𝑠𝑡𝑖𝑚𝑎𝑡𝑒𝑑 𝑡𝑜𝑡𝑎𝑙 𝑎𝑐𝑡𝑖𝑣𝑖𝑡𝑦
Solved example:
A vehicle is estimated to go 150,000 Km in its useful lifetime. Its cost is $17,000.
Its scrap value is $2,000. Find its depreciation rate. Find the depreciation value
of the second year if the usage during that year was 2,500 Km.
The deprivation rate is based on the mileage of the vehicle.
𝟏𝟕, 𝟎𝟎𝟎 − 𝟐, 𝟎𝟎𝟎

𝑷𝒆𝒓 𝒎𝒊𝒍𝒆 𝒅𝒆𝒑𝒓𝒆𝒄𝒊𝒂𝒕𝒊𝒐𝒏 𝒓𝒂𝒕𝒆 =
𝟏𝟓𝟎, 𝟎𝟎𝟎
= $𝟎. 𝟏 𝒑𝒆𝒓 𝑲𝒎
During the second year:

𝑫𝒆𝒑𝒓𝒊𝒕𝒂𝒕𝒊𝒐𝒏 𝒐𝒇 𝒕𝒉𝒆 𝟐𝒏𝒅 𝒚𝒆𝒂𝒓 = 𝟎. 𝟏 × 𝟐𝟓𝟎𝟎
= $𝟐𝟓𝟎
IO –Lecture Notes
Page 42
Units of time depreciation

Units of time depreciation is similar to units of production and is used for
depreciation equipment used in mine or natural resource exploration, or cases
where the amount the asset is used is not linear year to year. The depreciation is
related to the operation time and not the absolute time.
A famous example is the data show projector. The projector lamp has a specific
lifetime (for example 2000 hours of operation). Therefore, its depreciation rate can
be related to the operating time.
𝑪𝒐𝒔𝒕 𝒐𝒇 𝒇𝒊𝒙𝒆𝒅 𝒂𝒔𝒔𝒆𝒕 − 𝒔𝒄𝒓𝒂𝒑 𝒗𝒂𝒍𝒖𝒆
𝑷𝒆𝒓 𝒕𝒊𝒎𝒆 𝒅𝒆𝒑𝒓𝒆𝒄𝒊𝒂𝒕𝒊𝒐𝒏 𝒓𝒂𝒕𝒆 =
𝑬𝒔𝒕𝒊𝒎𝒂𝒕𝒆𝒅 𝒕𝒐𝒕𝒂𝒍 𝒐𝒑𝒆𝒓𝒂𝒕𝒊𝒏𝒈 𝒕𝒊𝒎𝒆
Solved example:
A training office bought a new projector for EGP 6000. The lifetime of its lamp
is guaranteed for 2500 hours of operation. Its scrap value then is EGP 300. A
“Matlab basics” course is to be trained using this projector. The Matlab course
needs three days of three hours each. Find the deprivation value of the projector
during the Matlab course.
𝟔𝟎𝟎𝟎 − 𝟑𝟎𝟎 𝟐. 𝟐𝟖 𝑳. 𝑬.
𝑫𝒆𝒑𝒓𝒆𝒄𝒊𝒂𝒕𝒊𝒐𝒏 𝒓𝒂𝒕𝒆 = = ⁄𝒉𝒐𝒖𝒓
𝟐𝟓𝟎𝟎
The course requires 9 operating hours.
∴ 𝑫𝒆𝒑𝒓𝒆𝒄𝒊𝒂𝒕𝒊𝒐𝒏 𝒗𝒂𝒍𝒖𝒆 = 𝟐. 𝟐𝟖 × 𝟗 = 𝑬𝑮𝑷 𝟐𝟎. 𝟓𝟐
Declining-balance method (or Reducing balance method)

It is one of the accelerated depreciation methods, which provides a higher
depreciation charge in the first year of an asset's life and gradually decreasing
charges in subsequent years. This may be a more realistic reflection of an asset's
actual expected benefit from the use of the asset: many assets are most useful when
they are new. Under this method the book value is multiplied by a fixed rate.
Annual Depreciation = Depreciation Rate * Book Value at Beginning of Year

The most common rate used is double the straight-line rate. For this reason, this
technique is referred to as the double-declining-balance method.
IO –Lecture Notes
Page 43
When using the double-declining-balance method, the salvage value is not

considered in determining the annual depreciation, but the book value of the asset
being depreciated is never brought below its salvage value, regardless of the
method used. The process continues until the salvage value or the end of the asset's
useful life, is reached. In the last year of depreciation, a subtraction might be
needed in order to prevent book value from falling below estimated Scrap Value.
Since double-declining-balance depreciation does not always depreciate an asset

fully by its end of life, some methods also compute a straight-line depreciation
each year, and apply the greater of the two. This has the effect of converting from
declining-balance depreciation to straight-line depreciation at a midpoint in the
asset's life.
It is possible to find a rate that would allow for full depreciation by its end of life
with the formula:
𝑵 𝑹𝒆𝒔𝒊𝒅𝒖𝒂𝒍 𝒗𝒂𝒍𝒖𝒆
𝑫𝒆𝒑𝒓𝒆𝒄𝒊𝒂𝒕𝒊𝒐𝒏 𝒓𝒂𝒕𝒆 = 𝟏 − √
𝑪𝒐𝒔𝒕 𝒐𝒇 𝒇𝒊𝒙𝒆𝒅 𝒂𝒔𝒔𝒆𝒕
where N is the estimated life of the asset (for example, in years).
Solved example:
A business has an asset with $1,000 original cost, $100 salvage value, and 5
years useful life. Find the depreciated value each year using the double-
declining-balance method.
180
The straight-line depreciation rate over the 5 years is = 20% per year.
900
For the double-declining-balance method, as the name suggests, double that rate,
or 40 % depreciation rate is used.
Book value at Depreciation Depreciation Book value at

Yr
beginning of year ($) rate (%) expense ($) end of year ($)
IO –Lecture Notes
Page 44
1,000
1 40% 400 600
(Original cost)
2 600 40% 240 360
3 360 40% 144 216
4 216 40% 86.4 129.6
100
5 129.6 ----- 29.6
(Scrap value)
Sum-of-years' digits method
Sum-of-years' digits is a depreciation method that results in a more accelerated
write-off than straight line, but less than declining-balance method. Under this
method annual depreciation is determined by multiplying the Depreciable Cost by
a schedule of fractions.
The formulae used in the calculation of the sum of years’ digits method are:
Annual Depreciation = Depreciation Rate * Total depreciation

𝑵−𝒏+𝟏
𝑫𝒆𝒑𝒓𝒆𝒄𝒊𝒂𝒕𝒊𝒐𝒏 𝒓𝒂𝒕𝒆 𝒐𝒇 𝒚𝒆𝒂𝒓 (𝒏) =
(𝑵𝟐 + 𝑵)
𝟐
N: is the expected useful lifetime (years)
(N2 + N)
𝑆𝑢𝑚 𝑜𝑓 𝑦𝑒𝑎𝑟𝑠 = 1 + 2 + 3 + ⋯ (𝑁 − 2) + (𝑁 − 1) + 𝑁 =
2
Solved Example:
If an asset has original cost of $1000, a useful life of 5 years and a salvage value
of $100, compute its depreciation schedule using the sum of years’ method.
Sum of years = 15, Total depreciation = $900
Book value at
Depreciation Depreciation Book value at
Yr beginning of
rate expense ($) end of year (n)
year
$1,000
1 5/15 (900 * 5/15) = 300 $700
(Original cost)
IO –Lecture Notes
Page 45
2 $700 4/15 (900 * 4/15) = 240 $460

3 $460 3/15 (900 * 3/15) = 180 $280
4 $280 2/15 (900 * 2/15) = 120 $160
$100
5 $160 1/15 (900 * 1/15) = 60
(Scrap value)
Group depreciation method

Group depreciation method is used for depreciating multiple-asset accounts using
straight-line-depreciation method. Assets must be similar in nature and have
approximately the same useful lives.
Scrap Depreciation
Asset Cost Life
value per year
Laptops $20,500 $500 5
Desktops $15,200 $1000 5
Photocopiers $10,000 $200 5
Smart board $20,000 $2,000 5
Group $65,900 $3,700 5 $12,440
Composite depreciation method

The composite method is applied to a collection of assets that are not similar, and
have different service lives. For example, computers and printers are not similar,
but both are part of the office equipment. Depreciation on all assets is determined
by using the straight-line-depreciation method.
Asset Historical Salvage Depreciable Life Depreciation

cost value cost per year
Computers $5,500 $500 $5,000 5 $1,000

Printers $1,000 $100 $ 900 3 $ 300
Composite $ 6,500 $600 $5,900 4.5 $1,300
IO –Lecture Notes
Page 46
Composite life equals the total depreciable cost divided by the total depreciation
per year. $5,900 / $1,300 = 4.5 years.
Composite depreciation rate equals depreciation per year divided by total

historical cost. $1,300 / $6,500 = 0.20 = 20%
Depreciation expense equals the composite depreciation rate times the balance in
the asset account (historical cost). (0.20 * $6,500) = $1,300. Debit depreciation
expense and credit accumulated depreciation.
When an asset is sold, debit cash for the amount received and credit the asset
account for its original cost. Debit the difference between the two to accumulated
depreciation. Under the composite method no gain or loss is recognized on the sale
of an asset. Theoretically, this makes sense because the gains and losses from
assets sold before and after the composite life will average themselves out.
To calculate composite depreciation rate, divide depreciation per year by total

historical cost. To calculate depreciation expense, multiply the result by the same
total historical cost. The result, not surprisingly, will equal to the total depreciation
Per Year again.
Common sense requires depreciation expense to be equal to total depreciation per

year, without first dividing and then multiplying total depreciation per year by the
same number.
IO –Lecture Notes
Page 47
Linear Programming
Definition:
Linear programming is a class of mathematical programming models concerned with
the efficient allocation of limited resources to known activities with the objective of
meeting a desired goal (such as maximizing profit or minimizing cost). The distinct
characteristic of linear programming models is that the functions representing the
objective and the constraints are linear.
Problem formulation:
Objective function:
Maximize (or minimize) x0 = c1  x1 + c2  x2 + c3  x3 +  + cm  xm
Subject to the constraints:
a11  x1 + a12  x2 + a13  x3 +  + a1m  xm  b1

a21  x1 + a22  x2 + a23  x3 +  + a2 m  xm  b2
an1  x1 + an 2  x2 + an3  x3 +  + anm  xm  bn
x1 , x2 , x3 , xm  0
This is a linear programming model of m variables, and n constraints.
Problems can be solved graphically or using the simplex tables depending on the
complexity of the problem.
Solved example 1:
Students at our faculty prefer eating healthy food. During break times, the
available fruits at the cafeteria are oranges and bananas. A kilogram of oranges costs
L.E. 3, and a kilogram of bananas costs L.E. 2. They constitute the following:
IO –Lecture Notes
Page 48
Supplement Iron Vitamin C Vitamin B Zinc

Oranges (mg/kg) 2 1 1 1
Bananas (mg/kg) 0 2 0 9
If the daily requirements of these supplements for an adult human being are:
Supplement Iron Vitamin C Vitamin B Zinc

Daily req. (mg) Minimum 2 Minimum 6 Maximum 9 Maximum 54
Find the amounts (in kilograms) of oranges and bananas that satisfy the daily
requirements and minimize the cost.
The problem needs to be formulated as a linear programming model.
Assuming:
x1 is the amount of oranges bought in Kgs
x2 is the quantity of bananas bought in Kgs
Objective function is to minimize the cost which is denoted as x0,

x0 = 3x1 + 2 x2
The constraints for each supplement are as follows:
For iron 2 x1 + 0 x2  2
For vitamin C 1x1 + 2 x2  6

For vitamin D 1x1 + 0 x2  9
For zinc 1x1 + 9 x2  54
IO –Lecture Notes
Page 49
Hence, the linear programming model is:
Decision variables:
x1 is the amount of oranges bought in Kgs
x2 is the quantity of bananas bought in Kgs
Objective function: Minimize: 𝑍 = 3𝑥1 + 2𝑥2
Subject to: 2𝑥1 ≥2
𝑥1 + 2𝑥2 ≥ 6
𝑥1 ≤9
𝑥1 + 9𝑥2 ≤ 54
𝑥1 , 𝑥2 ≥0
Graphically:
12
10 Objective
8 Const1
6 Const2
4 Const3
2 Const4
Const5
0
0 2 4 6 8 10 12
From the graph, the minimum cost occurs at the intersection of constraints 1 & 2.
Optimum solution occurs at
x1 = 1 kg , x2 = 2.5 kg
 x0 = 8L.E.
IO –Lecture Notes
Page 50
Solved example 2:
A furniture company produces three types of desks. Each desk requires two
resources: labor and material. The following data were collected:
Labor (hr/unit) Material (Kg/unit) Profit (L.E./unit)

Model A 7 4 4
Model B 3 4 2
Model C 6 5 3
The supply of raw material is restricted to 200Kg/day. The daily availability of

labor is 150 hrs/day.
Formulate a linear programming model to determine the daily production of each

model to maximize the profit.
The linear programming model is:
Decision variables:
x1 : Number of units of model A
x2 : Number of units of model B
x3 : Number of units of model C
Objective function: Maximize: 𝑍 = 4𝑥1 + 2𝑥2 + 3𝑥3
Subject to: 7x1+3x2+6x3 ≤ 150
4x1+4x2+5x3 ≤ 200
x1 , x2 , x 3 ≥ 0
IO –Lecture Notes
Page 51
Solved example 3:
A company has two grades of inspectors; “1” and “2”, who are to be assigned for a
quality control inspection. The company requires that at least 1800 pcs/ 8hrs are
inspected daily.
Grade 1 inspectors can check pieces at the rate of 25 pcs/ hr with an accuracy of
98%. Grade 2 inspectors can check at the rate of 15 pcs/ hr with an accuracy of
95%.
The average cost rate of grade 1 inspectors is L.E.4/hr, while that of grade 2 is
L.E.3/hr.
Each time an error is made by an inspector costs the company L.E. 2. The available
inspectors at the company are 8- grade 1 inspectors, and 10- grade 2 inspectors.
The company wants to determine the optimal assignment of inspectors which will
minimize the total cost of inspection.
Inspection Cost Available

Accuracy
Rate (pcs/hr) (L.E./hr) inspectors
Grade 1 25 98% 4 8
Grade 2 15 95% 3 10
Required Error cost

1800 2
pcs/8 hrs (L.E./pc)
In order to formulate the linear programming model:
Decision variables:
x1 : number of grade-1 inspectors
IO –Lecture Notes
Page 52
Objective function: Minimize cost:
Z = Wages + Total error cost
Wages= 4 × 8 𝑥1
+3 × 8 𝑥2
Total error cost =25 × 8 × 0.02 × 2 𝑥1
+15 × 8 × 0.05 × 2𝑥2
𝑍 = 4 × 8 𝑥1 + 3 × 8 𝑥2 + 25 × 8 × 0.02 × 2 𝑥1 + 15 × 8 × 0.05 × 2𝑥2
Subject to:
Total items inspected: 25 × 8 𝑥1 + 15 × 8 𝑥2 ≥ 1800
Maximum number of inspectors G1: 𝑥1 ≤ 8
Maximum number of inspectors G2: 𝑥2 ≤ 10
---------------------------
Hence, the linear programming model is:
Decision variables: x1 : number of grade-1 inspectors
Objective function: Minimize cost: 𝑍 = 40 𝑥1 + 36 𝑥2
Subject to: 5𝑥1 + 3𝑥2 ≥ 45

x1 ≤8
x2 ≤ 10
x1 ,x2 ≥ 0
IO –Lecture Notes
Page 53
Solving linear programming models graphically

Solved example 1:
The Jones Company manufactures two products, J1 and J2. Each J1 sold
contributes $6, and each J2 contributes $5. In addition, each product must pass
through two assembly points, A1 and A2. J1 requires 4 hours in A1 and 4 hours in
A2. J2 requires 3 hours in A1 and 5 hours in A2. There are 40 hours available in
A1 and 30 hours available in A2. Determine the optimum product mix.
GIVEN DATA J1 J2
Contribution ($) 6 5
A1 (40 hrs) 4 3
A2 (30 hrs) 4 5
The linear programming model is:
Decision variables: x1 : Units produced of product J1

x2 : Units produced of product J2
Objective function: Maximize contribution: 𝑍 = 6 𝑥1 + 5 𝑥2
Subject to: 4𝑥1 + 3𝑥2 ≤ 40

4𝑥1 + 5𝑥2 ≤ 30
𝑥1 , 𝑥2 ≤ 0
14
12
Const 1
10
Const 2
8
X2
6
4
2
0
0 2 4 6 8 10 12
X1
𝑥1 = 7.5 𝑢𝑛𝑖𝑡𝑠 𝑜𝑓 𝐽1 , 𝑥2 = 0 𝑢𝑛𝑖𝑡𝑠 𝑜𝑓 𝐽2 , 𝑍 = $45
IO –Lecture Notes
Page 54
Solved example 2:
Find the solution of the following linear programming model graphically:
Objective function: Maximize: 𝑍 = 5 𝑥1 + 4 𝑥2
Subject to: 6𝑥1 + 34𝑥2 ≤ 24

𝑥1 + 2 ≤ 6
−𝑥1 + 𝑥2 ≤ 1
𝑥2 ≤ 2
𝑥1 , 𝑥2 ≤ 0
Plot all the constraints on the graph, as well as the slope of the objective function.
X2 8
0
-2 0 2 4 6 8
-1
X1
-2
N-N1 N-N2 Const 1 Const 2 Const 3 Const 4 obj
The maximized objective function occurs at 𝑥1 = 3 , 𝑥2 = 1.5 , 𝑍 = 21
IO –Lecture Notes
Page 55
Transportation
Definition:
The transportation problem is concerned with shipping a commodity between a set
of sources (e.g. manufacturers) and a set of destinations (e.g. retailers).
Transportation problem parameters and assumptions:

• Each source has a capacity dictating the amount it supplies.
• Each destination has a demand dictating the amount it receives.
• For now, we assume that the system is balanced; i.e. the total amount
demanded at all destinations is equal to the total supply available at all
sources.
• The objective is to determine the amounts to be shipped between each
source and destination in a way as to minimize shipping cost while meeting
demand and supply constraints.
• The shipping cost is proportional to the amount shipped.
• The transportation problem can be modeled and solved as a linear program.
Transportation problem formulation as Linear Programming Model

Suppose a company has m warehouses and n retail outlets. A single product is to
be shipped from the warehouses to the outlets. Each warehouse has a given level of
supply, and each outlet has a given level of demand. We are also given the
transportation costs between every pair of warehouse and outlet, and these costs
are assumed to be linear. More explicitly, the assumptions are:
• The total supply of the product from warehouse i is ai , where i = 1, 2, . . ., m.

• The total demand for the product at outlet j is bj , where j = 1, 2, . . ., n.
• The cost of sending one unit of the product from warehouse i to outlet j is
equal to cij , where i = 1, 2, . . ., m and j = 1, 2, . . ., n. The total cost of a
shipment is linear in the size of the shipment.
The problem of interest is to determine an optimal transportation scheme between

the warehouses and the outlets, subject to the specified supply and demand
constraints. Graphically, a transportation problem is often visualized as a network
IO –Lecture Notes
Page 56
with m source nodes, n destination nodes, and a set of m × n “directed arcs.” This
is depicted in the following figure.
Source Destinations
s
a 1 1 b
2 b
a 2
3 b
a 3 4 b
i
j
The Decision Variables
A transportation scheme is a complete specification of how many units of the
product should be shipped from each warehouse to each outlet. Therefore, the
decision variables are:
xij = the size of the shipment from warehouse i to outlet j,
where i = 1, 2, . . ., m and j = 1, 2, . . ., n.
This is a set of m × n variables.
The Objective Function

Consider the shipment from warehouse i to outlet j. For any i and any j, the
transportation cost per unit is cij ; and the size of the shipment is xij . Since we
assume that the cost function is linear, the total cost of this shipment is given by
cijxij . Summing over all i and all j now yields the overall transportation cost for all
warehouse-outlet combinations. That is our objective function is:
Minimize 𝑚 𝑛
𝑍 = ∑ ∑ 𝑐𝑖𝑗 ∙ 𝑥𝑖𝑗
𝑖=1 𝑗=1
IO –Lecture Notes
Page 57
The Constraints
Consider warehouse i. The total outgoing shipment from this warehouse is the sum
xi1 + xi2 + · · · + xin. In summation notation, this is written as∑𝑛𝑗=1 𝑥𝑖𝑗 . Since the
total supply from warehouse i is ai , the total outgoing shipment cannot exceed ai .
That is, we must require
𝑛
∑ 𝑥𝑖𝑗 = 𝑎𝑖 ∀𝑖
𝑗=1
Consider outlet j. The total incoming shipment at this outlet is the sum of
x1j+x2j+· · ·+xmj . In summation notation, this is written as ∑𝑚

𝑖=1 𝑥𝑖𝑗 . Since the
demand at outlet j is bj , the total incoming shipment should not be less than bj .
That is, we must require
𝑚
∑ 𝑥𝑖𝑗 = 𝑏𝑗 ∀𝑗
𝑖=1
This results in a set of m + n functional constraints. Of course, as physical
shipments, the xij ’s should be nonnegative.
LP Formulation
In summary, we have arrived at the following formulation:
Minimize:
𝑚 𝑛
𝑍 = ∑ ∑ 𝑐𝑖𝑗 ∙ 𝑥𝑖𝑗
𝑖=1 𝑗=1
Subject to:
𝑛
∑ 𝑥𝑖𝑗 = 𝑎𝑖 ∀𝑖
𝑗=1
𝑚
∑ 𝑥𝑖𝑗 = 𝑏𝑗 ∀𝑗
𝑖=1
𝑥𝑖𝑗 ≥ 0 ∀ 𝑖, 𝑗
IO –Lecture Notes
Page 58
This is a linear program with m × n decision variables, m + n functional

constraints, and m × n non-negativity constraints.
Transportation problem form

The data of the transportation problem is given in a table as the following. The
rightmost column is the capacities of the three given sources. The lowermost row
is the demands of the four destination. The intermediate cells are the transportation
cost from each source to each destination.
D1 D2 D3 D4
S1 10 0 20 11 15
S2 12 7 9 20 25
S3 0 14 16 18 5
5 15 15 10
IO –Lecture Notes
Page 59
The given data is re-filled in a table that shows all the data while giving space for
filling the transported quantities. This is shown in Table (1). Assume that the unit
cost in this problem is EGP.
Table (1)
Destination
Supply
1 2 3 4
10 0 20 11
1 15
Source
12 7 9 20
2 25
0 14 16 18
3 5
Demand 5 15 15 10 45
A standard transportation problem must have:

Sum (Demand) = Sum (Supply)
The solution comes in two steps; finding a start solution, then improving this
solution.
Step One: Find a start solution
There are three methods to find the start solution. You can choose any of them
unless the problem specifies one.
• Northwest corner method

• Least cost method
• Vogel’s approximation method
Northwest corner method

Table (2). Start by filling the cell in the northwest corner of the table (i.e. the top
left corner; from Source1 to Destination1). Then cross out the row/column that is
satisfied. Proceed with the remaining cells until all the quantities are delivered.
IO –Lecture Notes
Page 60
Table (2)
Destination
Supply
1 2 3 4
10 0 20 11
1 15
5 10
Source
12 7 9 20
2 25
5 15 5
0 14 16 18
3 5
5
Demand 5 15 15 10 45
Cost = EGP 410.

Number of variables in the table = m + n – 1 = 6
Least cost method
Table (3). Start by filling the cell of the least cost (from Source3 to Destination1).
If two values are equal, choose any of them arbitrary. Then cross out the
row/column that is satisfied. If both the row and column are satisfied, cross out
ONE of them ONLY. Proceed with the remaining cells until all the quantities are
delivered.
Table (3)
Destination Supply
1 2 3 4
10 0 20 11
1 15
0 15 __ 0
Source
12 7 9 20
2 25
__ __ 15 10
0 14 16 18
3 5
5 __ __ __
Demand 5 15 15 10 45
Cost = EGP 335.
IO –Lecture Notes
Page 61
Vogel’s approximation method

Tables (4) to (9). Vogel’s method comes in iterative steps that are explained in the
following.
1. Evaluate a penalty for each row and each column (Lowest cost – next lowest
cost)
2. Choose the row/ column of highest penalty and allocate as much as possible
in the least cost of this row/column.
3. Eliminate the row/column.
4. Repeat and build new penalties for the remaining rows and columns.
Table (4)
Destination
Supply Penalty
1 2 3 4
10 0 20 11 10
1 15
__
Source
12 7 9 20 2
2 25
__
0 14 16 18 14
3 5
5
Demand 5 15 15 10 45
Penalty 10 7 7 7
Table (4). Highest penalty is 14 (in row 3 for S3). Allocate as much quantities as
possible in the least cost of S3 (from S3 to D1). Remove column1. BUT remove
costs from row 3. They are not assignable any more.
IO –Lecture Notes
Page 62
Table (5)
Destination Supply Penalty
2 3 4
0 20 11
1 15 11
Source
7 9 20
2 25 2
15
3 0 --
Demand 15 15 10 45
Penalty 7 11 9
Table (5). The highest penalty has a value of 11. It appears twice. Choose arbitrary
between the penalties of Source1, and Destination3. Choosing Destination3,
allocate as much quantities as possible in the least cost of D3 (from S2 to D3).
Remove column of D3. (Try removing row 1).
Table (6)

2 4
0 11
1 15 11
Source
7 20
2 10 13
10
3 0 --
Demand 15 10 35
Penalty 7 9
Table (6). Highest penalty is 13of S2. Allocate quantities from S2 to D2. Remove
row of S2.
IO –Lecture Notes
Page 63
Table (7)

2 4
0 11
1 15 11
Supply
5 10
3 0 --
Demand 5 10 15
Penalty -- --
Table (7). Highest penalty occurs at S1. Allocate quantities from S1 to D2.
Remove column of D2 as its demand is satisfied.
Table (8)
Destination Supply
4
11
1 10
Supply
10
3 0
0
Demand 10 10
Table (8). The only remaining quantities go from S1 to D4. Solution ends.
If we collect all the variables obtained from all the previous tables in one table,
table (9) is obtained.
IO –Lecture Notes
Page 64
Table (9)
Destination
Supply
1 2 3 4
10 0 20 11
1 15
5 10
Source
12 7 9 20
2 25
10 15
0 14 16 18
3 5
5 0
Demand 5 15 15 10 45
Starting solution has 6 variables.

Cost = EGP 315.
(This is the optimal solution. When choosing row 1 instead of column 3 in Table
(5), cost = EGP 335)
IO –Lecture Notes
Page 65
Step Two: Improve the start solution

While improving an existing solution, there is no need to have the capacities of the
sources, nor the demands of the destinations. These constraints are already satisfied
in the initial solution. We need only the quantities obtained from the initial
solution. Table (10) shows the initial solution obtained by the Northwest method.
Table (10)
D1 D2 D3 D4
10 0 20 11
S1
5 10
12 7 9 20
S2
5 15 5
0 14 16 18
S3
5
The solution is improved by steps. In each step, there is an entering variable, and a
leaving variable.
There are two methods for improving the current solution. Both will be explained.
Stepping stone method:
Each non-basic variable (a variable of no value) is examined. What will be the

effect on the cost if the variable is increased by one unit? Will the total cost
increase or decrease? By how much?
The examination is achieved by creating a loop of increasing, and decreasing the

surrounding basic variables by a value of one unit. For each non-basic variable,
there is only one possible loop. The clockwise loop is identical to the anticlockwise
one.
Table (11) shows the examination of the non-basic variable from S2 to D1. The
dotted line represents the loop along the surrounding basic variables. A step in the
loop must be either vertical or horizontal. Diagonal steps are not allowed. The
IO –Lecture Notes
Page 66
vertical step should be followed by a horizontal step, and vice versa. Two
successive horizontal (or vertical) steps are not allowed.
Adding one unit to the cell of S2-D1 requires removing one unit from S1-D1 in
order to keep the demand of D1 unchanged. Similarly, we need to add one unit to
S1-D2 in order to keep the capacity of S1 unchanged. Consequently, we need to
remove one unit from S2-D2 in order to keep the demand of D2 unchanged.
Returning back to S2-D1 closes the loop.
Table (11)
D1 D2 D3 D4
S1 - 10 + 0 20 11
5 10
S2
+ 12 - 7 9 20
5 15 5
0 14 16 18
S3
5
This move of one unit causes the cost to increase twice, and decrease twice. The
net change in cost can be calculated as follows:
∆𝐶𝑜𝑠𝑡 = +12 − 10 + 0 − 7 = −5 EGP
Hence the cost will be decreased by 5 units.
Table (12). The move of one unit in the loop of S2-D1 causes a decrease in the cost
by EGP 5. Hence, this cell will be denoted by (-5). Similarly, all the non-basic
variables cells are examined, and their effect is registered in the table. If a cell
causes an increase in the cost, there is no need to have the exact value of the
increase. It is enough to denote the cell by (+ve). This means that this cell will
cause an increase in the cost, and not a decrease.
IO –Lecture Notes
Page 67
Table (12)
D1 D2 D3 D4
10 0 +18 20 -2 11
S1
5 10
-5 12 7 9 20
S2
5 15 5
-15 0 +9 14 +9 16 18
S3
5
This indicates that the best improvement is moving units in the cell of S3-D1. Each
unit will decrease the cost by 15 units. The maximum units that can be moved is 5
units. The loop of increasing and decreasing units can be summarized as:
Cell Description Quantity moved Was Became

S3-D1 Increased +5 units (Non-basic) 5
S1-D1 Decreased -5 units 5 0
S1-D2 Increased +5 units 10 15
S3-D4 Decreased -5 units 5 (Non-basic)
Notice that in the previous table, only one non-basic variable becomes basic, and
only one basic variable becomes non-basic. Cells S1-D1, and S2-D2 should be
filled by zeros in order to keep the total number of basic variables the same.
Table (13) shows the quantities transported after making that movement of the 5
units around the loop. Mind that
IO –Lecture Notes
Page 68
Table (13)
D1 D2 D3 D4
10 0 20 11
S1
0 15
12 7 9 20
S2
0 15 10
0 14 16 18
S3
5
Table (14) is the same as Table (13) after making the calculations of ΔCost for
each cell of the non-basic variables.
Table (14)
D1 D2 D3 D4
10 0 +ve 20 -2 11
S1
0 15
-5 12 7 9 20
S2
0 15 10
0 +ve 14 +ve 16 +ve 18
S3
5
Table (14) indicates that quantities should now be added to cell S2-D1 because of
its high negative value according to the following:

Theoretically, values will stay the same. The only change is in the basic, and non-
basic variables.
Table (15) shows the new transported quantities, and the calculations of ΔCost for
each cell of the non-basic variables.
IO –Lecture Notes
Page 69
Table (15)
D1 D2 D3 D4
+ve 10 0 +ve 20 -2 11
S1
15
12 7 9 20
S2
0 0 15 10
0 +ve 14 +ve 16 +ve 18
S3
5
The values of (Δ Cost) indicate that quantities should now be added to cell S1-D4
because of its negative value according to the following:

Table (16) shows the updated values of the quantities transported. All the Δ cost
values are positive which indicates that there is no more improvement, and an
optimum solution is reached.
Notice that the optimum solution reached after three steps of improvement is the
same solution that was reached from the first time using Vogel’s method. In this
case the total transportation cost is EGP 315.
Table (16)
D1 D2 D3 D4
+ve 10 0 +ve 20 11
S1
5 10
12 7 9 +ve 20
S2
0 10 15
0 +ve 14 +ve 16 +ve 18
S3
5
IO –Lecture Notes
Page 70
U-V method:
The U-V method will be used to improve the start solution obtained by the north-
west corner method. U values will be assigned to the rows, while V values will be
assigned to the columns. Start by assuming u1 = 0, then calculate the rest u and v
using the rule:
cij = ui + v j , for all the cells with basic variables.
Table (17)
v1 = 10 v2 = 0 v3 = 2 v4 = 13
10 0 20 11
u1 = 0
5 10
12 7 9 20
u2 = 7
5 15 5
0 14 16 18
u3 = 5
5
Current Cost = 410 units
After calculating all the U and V values, calculate the change in cost (𝑐̅) for all the
empty cells (non-basic variables) using the equation:
cij = cij − ui − v j
The 𝑐̅ values are given in the following table:
Table (18)
IO –Lecture Notes
Page 71
v1 = 10 v2 = 0 v3 = 2 v4 = 13
10 0 +18 20 -2 11
u1 = 0
5 10
-5 12 7 9 20
u2 = 7
5 15 5
-15 0 +2 14 +9 16 18
u3 = 5
5
Transport units (as much as possible) through the cell whose 𝑐̅ is of highest
negative value; from Supplier 3 to Destination 1.
Table (19): transported quantities are decided according to the stepping stone
method. Choose a suitable path for removing and adding units to keep the
constraints satisfied. The path is illustrated in the table. The (+) and (-) signs show
the cells where the units will be added (+), and where the units will be removed (-).
Table (19)
v1 = 10 v2 = 0 v3 = 2 v4 = 13
10 0 +18 20 -2 11
u1 = 0
- 5 + 10
-5 12 7 9 20
u2 = 7 15
- 5 + 5
-15 0 +2 14 +9 16 18
u3 = 5
+ - 5
The cells with -ve values are the cells from which units will be removed. The
maximum possible quantity to be removed is 5. Add and remove 5 units from the
cells with +ve and –ve signs. The following table is obtained.
Then, the new U and V values are calculated. The new 𝑐̅ values are calculated as
shown in table (20).
IO –Lecture Notes
Page 72
Table (20)
v1 = 10 v2 = 0 v3 = 2 v4 = 13
10 0 +18 20 -2 11
u1 = 0
- 0 + 15
-5 12 7 9 20
u2 = 7 + - 0 15 10
0 +24 14 +24 16 +15 18
u3 = -10
5
Cost = 335 units.

As shown in table (20), the next non-basic variable chosen is the one with the
highest negative (-5), from supplier 2 to destination 1. The (+) and (-) signs are as
shown. The maximum amount to be transported is 0.
A new table (21) is established to show the transportation of the 0 units, and the
calculation of the new U and V.
Table (21)
v1 = 5 v2 = 0 v3 = 2 v4 = 13
+5 10 0 +18 20 -2 11
u1 = 0
- 15 +
-5 12 7 9 20
u2 = 7
0 + 0 15 - 10
0 +19 14 +19 16 +10 18
u3 = -5
5
IO –Lecture Notes
Page 73
Cost = 335 units
From table (21), the cell with the highest negative 𝑐̅ value is from Supplier1 to
Destnation4. By drawing the suitable path, and writing the (+) and (-) signs, the
maximum amount to be transported through this cell is 10.
Table (22) shows the transportation of 10 units and the calculation of the new U
and V values. The new 𝑐̅ values are also calculated.
Table (22)
v1 = 5 v2 = 0 v3 = 2 v4 = 11
+5 10 0 +18 20 11
u1 = 0
5 10
-5 12 7 9 +2 20
u2 = 7
0 10 15
0 +19 14 +19 16 +12 18

u3 = -5
5
Cost = 315 units
There are no negative values for 𝑐̅, which means that the optimal solution is reached.
IO –Lecture Notes
Page 74
Special cases of problems

Unbalanced model
In case that: Sum (Supply) ≠ Sum (Demand), a dummy supply (or destination) is
added to balance the sum.
1 2 3 4
1 1 4 6 3 15
2 9 7 10 9 25
3 4 5 11 7 15
5 15 15 10
Sum (supplies) = 55 units. Sum (Demand) = 45 units
Add a dummy destination to increase the Sum of Demand to 55. The demand of the
dummy destination is 10 units. (55-45 = 10 units).
The costs of transportation associated with the dummy destination are zeros.
1 2 3 4 D
1 1 4 6 3 0 15
2 9 7 10 9 0 25
3 4 5 11 7 0 15
5 15 15 10 10
Then the problem is solved like any standard problem.
Destination
Supply
1 2 3 4 D
10 0 20 11 0
1 15
5 10
Source
12 7 9 20 0
2 25
5 15 5
0 14 16 18 0
3 15
5
Demand 5 15 15 10 10 55
IO –Lecture Notes
Page 75
Prohibited routes
Sometimes one or more of the routes in the transportation problem are prohibited
(units cannot be transported from a certain source to a certain destination). In order
to ensure that no units will be transported from this source to this destination, the
prohibited route is assigned a Big M value.
In the following table, the route between S2 and D4 is temporarily closed. The
route is assigned a big M cost.
Destination
Supply
1 2 3 4
10 0 20 11
1 15
Source
12 7 9 M
2 25
0 14 16 18
3 5
Demand 5 15 15 10 45
Maximizing the profit

Sometimes, the transportation table is filled by the profit instead of the cost. This is
the profit of transporting one unit.
In this case, the least cost method is replaced by the maximum profit.
In Vogel’s method, the penalty is calculated as the difference between the highest
two profits. Quantities are transported for the maximum profit cells.
In the improvement method, the cells of the highest positive values of 𝑐̅ have
priority.
IO –Lecture Notes
Page 76
Scheduling
Definition:
It is determining the order in which jobs at a work center will be processed.
Priority rules
• First Come First Served (FCFS) / First In First Out (FIFO)
• Shortest Processing Time (SPT)
• Earliest Due Date (EDD)
• Critical Ratio (CR)
• Slack per Operation (S/O)
• Rush
Performance measures
• Job Flow time: It is the length of time the job spends in a work station
(including waiting)
• Job tardiness: It is the length of time the job completion date exceeds its due
date.
IO –Lecture Notes
Page 77
• Makespan: It is the total time to complete a group of jobs; from the start of
the first job, till the end of the last job.
• Average number of jobs: It is the average work-in-process inventory for a

group of jobs.
Sequencing jobs on one machine / one work center
Solved example:
In a workshop, six jobs are to be processed on one Work Centre. Determine the
sequence and the performance measures (Average flow time, Average tardiness,
Makespan, Average number of jobs) for the different priority rules (FCFS, SPT,
EDD)
Due
Arrival Processing
Job date
order time (days)
(days)
A 1 2 7
B 2 8 16
C 3 4 4
D 4 10 17
E 5 5 15
F 6 12 18
FCFS
Sequence:
Arrival Processing Due date Flow time Lateness

Job
order time (days) (days) (days) (days)
A 1 2 7
B 2 8 16
C 3 4 4
D 4 10 17
E 5 5 15
F 6 12 18
IO –Lecture Notes
Page 78
SPT
Sequence:

Job
A 1 2 7
B 2 8 16
C 3 4 4
D 4 10 17
E 5 5 15
F 6 12 18
EDD
Sequence:

Job
A 1 2 7
B 2 8 16
C 3 4 4
D 4 10 17
E 5 5 15
F 6 12 18
Comparing the three rules:
Av. Flow time Av. Tardiness Av. No. of

Rule
(days) (days) jobs
FCFS 20.00 9.00 2.93
SPT 18.00 6.67 2.63
EDD 18.33 6.33 2.68
IO –Lecture Notes
Page 79
Sequencing jobs through two work centers: Johnson’s rule
It is a technique to minimize makespan for a group of jobs to be processed on two

machines (work centers).
Steps of Johnson’s rule:

• List the jobs and their times at each work center
• Select the job with the shortest time. Of the shortest time is at the first work
center, schedule that work first; if the time is at the second work center,
schedule the job last. Break ties arbitrarily.
• Eliminate the job and its time from further consideration.
• Repeat the previous steps, working towards the center of the sequence, until
all the jobs have been scheduled.
Solved example:
For six jobs to be processed on a two-machine flow shop, find the sequence that
minimizes the total completion time of all jobs.
Processing time (hrs)
Job
Work centre 1 Work centre 2
A 5 5
B 4 3
C 8 9
D 2 7
E 6 8
F 12 15
Sequence: D-E-C-F-A-B
10 20 30 40 50
C1
C2
Makespan =
IO –Lecture Notes
Page 80
Shortest Path Problems

Problem definition:
Given n nodes in a network that are denoted by (1, 2, 3…n), it is required to find
the shortest path from the source node to the sink node.
Solved example
- Given the roads of the city, go from (A) to (H) by the shortest route. Determine
the path and its distance.
B 5 G
E
5 4 10
A 1 C 9 6
3
9
7 2
3 F 8 H
D
Sink
Solving steps:
Step L (0):
• Start at source node (A). Its distance is zero.

• Distances to neighboring nodes are written.
• Distances to non-neighboring nodes are infinity.
• Give the node of zero value a permanent label (*).
• Give the node of minimum value a temporary label (.)
Step L (1):
• Calculate the new distances from the temporary label (as explained in the
last table).
• Temporary label is permanent.
• Choose a new node to have the temporary label at L (2).
IO –Lecture Notes
Page 81
Repeat the previous steps till all the nodes are labeled as permanent nodes.
A B C D E F G H
L(0)= [0, 5, 1, 7, ∞, ∞, ∞, ∞]
* .
L(1)= [0, 5, 1, 3, 11, 10, ∞, ∞]

* * .
L(2)= [0, 5, 1, 3, 11, 6, ∞, ∞]

* . * *
L(3)= [0, 5, 1, 3, 11, 6, ∞, ∞]

* * * * .
L(4)= [0, 5, 1, 3, 9, 6, 15, 14]

* * * * . *
L(5)= [0, 5, 1, 3, 9, 6, 14, 14]

* * * * * * .
L(6)= [0, 5, 1, 3, 9, 6, 14, 14]

* * * * * * * *
Follows is the table of comparing distances. L (1) will be explained. Its table goes
as follows:
• The first column is the starting node. This is the node with the temporary
label in L (0). “Node C”
• The second column is the list of the nodes that have no labels in L (0).
Nodes “B, D, E, F, G, H”
• The third column lists the current values of the distances as in L (0).
• The fourth column calculates the new distance to each node when reached
through the current temporary node (C). This is the addition of the value of
IO –Lecture Notes
Page 82
the current temporary node (1) and the distance between it and the required
node.
• When comparing the Current and New distances, choose the minimum value
of both and record in the new trial of L (1).
L(1) L(3)
From To Current New From To Current New
C B 5 1+4 B E 11 5+∞
C D 7 1+2 B F 6 5+∞
C E ∞ 1+10 B G ∞ 5+∞
C F ∞ 1+9 B H ∞ 5+∞
C G ∞ 1+∞
C H ∞ 1+∞ L (4)
F E 11 6+3
L(2) F G ∞ 6+9
D B 5 3+∞ F H ∞ 6+8
D E 11 3+∞
D F 10 3+3 L(5)
D G ∞ 3+∞ E G 15 9+5
D H ∞ 3+∞ E H 14 9+∞
L(6)
G H 14 14+6
The shortest path is 14 units length. This is the last value obtained in L (6).
The shortest path is detected first in a reverse order.
• The last node is the sink node (H). Its value is 14. The first appearance of the
“14” was in L (4). That was because node (F) was temporary in L (3).
Therefore, node (F) becomes before the sink node (H).
• For node (F) with value “6”, the first appearance of “6” was in L (2). That
was because (D) was temporary in L (1). Therefore, node (D) becomes
before node (F).
IO –Lecture Notes
Page 83
• For node (D) of value of value “3”, the first appearance of “3” was in L (1).
That was because (C) was temporary in L (0). Therefore, node C becomes
before node (D).
• For node (C) of value of value “1”, the first appearance of “1” was in L (0).
Therefore, the source node (A) comes before node (C).
The path in a reverse order is:

H, F, D, C, A.
The path is: A, C, D, F, H
IO –Lecture Notes
Page 84

2022 IO - Lecture Notes

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2022 IO - Lecture Notes

Uploaded by

Copyright:

Available Formats

Introduction to Industrial Organization Course

4th year Mechanical Power

Introduction to Industrial Organization

When do we use forecasting?

Forecasting according to time

Long term Intermediate term Short term

Forecasting according to method

Market survey Time series

Types of forecasting trends

Time Series Model

The equation of the best fitting line is 𝑦 = 𝑎 + 𝑏𝑥

y : is the trend value

385 × 7407 − 55 × 41358

𝑦(11) = 6.99.4 + 7.51 × 11 = 782.01 𝑢𝑛𝑖𝑡𝑠

Sales of week 11 = 782.01 units

Year (t) Sales (x) Profit (y) x2 xy t2 ty y2

1 7 0.15 49 1.1 1 0.2 0.023

2 2 0.10 4 0.2 4 0.2 0.010

3 6 0.13 36 0.8 9 0.4 0.017

4 4 0.15 16 0.6 16 0.6 0.023

5 14 0.25 196 3.5 25 1.3 0.063

6 15 0.27 225 4.1 36 1.6 0.073

7 16 0.24 256 3.8 49 1.7 0.058

8 12 0.20 144 2.4 64 1.6 0.040

9 14 0.27 196 3.8 81 2.4 0.073

10 20 0.44 400 8.8 100 4.4 0.194

55 110 2.20 1522 29 385 14 0.571

𝑟𝑥,𝑦 = 0.919 𝑟𝑡,𝑦 = 0.827

The best fitting line is: 𝑦(𝑡) = 0.051 + 0.015 𝑡

Profit of $10M = $ 246,000

• When historical data has random variations / noise

Moving average (MA)

Original 2-years M.A. 5-years M.A.

Best fitting line for MA5 is: y (x) = 39.15+ 3.649 x

y (9) = 71.991 units

Aspect Lower value of N Higher value of N

Responsiveness more responsive to changes less responsive to changes

Uses appropriate for changing pattern appropriate for stable demand

Original 3-years MA 5-years MA Least sq.

𝐴𝑡 : Actual demand of (t)

Year (t) Sales (At) Forecast (Ft)

Aspect α < 0.5 α > 0.5

Responsiveness Slow responsiveness to adjust Fast responsiveness to adjust

Smoothness Higher smoothing Lower smoothing

Original index 0.15 index 0.6

Year Winter Spring Autumn Summer

1 190 350 250 200

2 250 400 300 150

3 300 340 260 180

4 280 400 220 200

5 350 420 340 240

4 Quarters Centered Average

1 1 190 1.037 183.220829

2 350 1.358 257.731959

4 240 0.655 366.6337483

To calculate the values of the 7th column:

0.8889 + 1.1215 + 1.0182 + 1.12

1.4222 + 1.2770 + 1.4679 + 1.2632

0.9804 + 1.0667 + 0.9720 + 0.7753

0.7442 + 0.5357 + 0.6606 + 0.6780

The best fitting line is: y = a + b (t)

y = 231.42 + 4.8676 (t)

Loose control e.g. Two bin system Wires, bolts, nuts