25-Nov-22

Operations and Supply Chain Management

S12: Forecasting

Prof. Vivek Roy

Indian Institute of Management Kashipur

Why Sales and Operations Planning?

1
 25-Nov-22

• Categories (AKA Budget Code) X ___?___ = Permissible Budget Limit

The Hierarchy
Economic,
Corporate competitive, Aggregate
strategies and political demand
and policies conditions forecasts

Macro Business Plan HQ, ∑Units

Meso Production plan ∑Units, Single Unit

Micro Master schedule Single Unit

2
 25-Nov-22

Result of Forecast at Micro

Unit A Product (I) 500 Product (G) 700 Product (M) 1500 Product (D) 100

Result of Forecast at Meso

Unit A Product (I) 500 Product (G) 700 Product (M) 1500 Product (D) 100

Unit B Product (I) 100 Product (G) 400 Product (M) 1000 Product (D) 50

Unit C Product (I) 200 Product (G) 600 Product (M) 1200 Product (D) 0
6

3
 25-Nov-22

Result of Forecast at Macro

Economic,
Corporate competitive, Aggregate
strategies and political demand
and policies conditions forecasts

Macro Business Plan

Meso Production plan

Micro Master schedule

“I know of no way of judging the future but by the

past”
Patrick Henry

4
 25-Nov-22

What is Forecasting?

 Denotation / Dictionary meaning

– Predicting the future
– Based on past information

 Estimating future demand for a product /

service
– Quantity
– Timing
– Quality
10
– Location
10

5
 25-Nov-22

Forecasting Horizon / Levels

Forecast Definition Example
Short-range Time span of up to one Planning purchasing, job
year; generallyless scheduling, work force levels,
than three months job assignments, and
production levels.

Medium-range Time span generally Sales planning, production

from three months to planning and budgeting, and
three years analyzing various operating
plans.

Long-range Time span generally Planning for new products,

five years or more capital expenditures, facility
location or expansion, and
research and development.

11

11

Types of Forecasts

 Demand forecasts
– Predict existing product sales
 Economic forecasts
– Address business cycle e.g., inflation rate,
money supply, governmental policy changes
etc.
 Technological forecasts
– Predict technological change
– Predict new product sales
12

12

6
 25-Nov-22

Demand Management, for example product

A is assembled from B, C, D, E, and F.

FORECAST Independent Demand:

Finished Goods
CALCULATE Dependent Demand:
A
Raw Materials,
Component parts,
B(4) C(2) Sub-assemblies, etc.

D(2) E(1) D(3) F(2)

13

13

Steps in forecasting

“The forecast”

Step 6 Monitor the forecast

Step 5 Prepare the forecast
Step 4 Gather and analyze data
Step 3 Select a forecasting technique
Step 2 Establish a time horizon
Step 1 Determine purpose of forecast 14

14

7
 25-Nov-22

EFFECT OF FORECASTING
EFFORTS

15

BALANCE OF FORECASTING
EFFORTS

16

8
 25-Nov-22

Naive Approach
 Assumes demand in next
period is the same as
demand in most recent
period
– e.g., If May sales were 48,
then June sales will be 48
 Sometimes cost effective
& efficient

© 1995 Corel Corp.

17

17

Forecasting Models
Forecasting Techniques

Qualitative Time-Series Methods Causal

Models Methods

Delphi Moving Regression Analysis

Methods Average

Jury of Executive Exponential Multiple

Opinion Smoothing Regression

Sales Force Trend

Composite Projections

Consumer
Decomposition
Market Survey
18

18

9
 25-Nov-22

Qualitative Models
• Jury of Executive Opinion – collects opinions of a small
group of high-level managers, possibly using statistical
models for analysis
• Sales Force Composite – individual salespersons estimate
the sales in their region and the data is compiled at a
district or national level
• Consumer Market Survey – input is solicited from
customers or potential customers regarding their
purchasing plans
• Delphi Method – an iterative group process where
(possibly geographically dispersed) respondents provide
input to decision makers
19

19

Judgment Forecasts (1)

20

20

10
 25-Nov-22

Judgment Forecast (2)

21

21

Judgment Forecast (3)

22

22

11
 25-Nov-22

Judgment Forecast (4)

23

23

Time-Series Models
• Time-series models attempt to predict the
future based on the past
• Common time-series models are
• Moving average
• Exponential smoothing
• Trend projections
• Decomposition
• Regression analysis is used in trend projections
and one type of decomposition model

24

24

12
 25-Nov-22

Decomposition of a Time-Series
• A time series typically has four components
1. Trend (T) is the gradual upward or downward
movement of the data over time
2. Seasonality (S) is a pattern of demand fluctuations
above or below trend line that repeats at regular
intervals
3. Cycles (C) are patterns in annual data that occur
every several years
4. Random variations (R) are “blips” in the data
caused by chance and unusual situations

25

25

Decomposition of a Time-Series

Trend
Demand for Product or Service

Component

Seasonal Peaks

Actual
Demand
Line
Average Demand
over 4 Years

| | | |
Year Year Year Year
1 2 3 4
Time

26

26

13
 25-Nov-22

Trend Component

• Persistent, overall upward or downward pattern

• Due to population, technology etc.
• Several years duration
Response

Time
27

27

Seasonal Component
• Regular pattern of up & down fluctuations
• Due to weather, customs etc.

Summer
Response

Time
28

28

14
 25-Nov-22

Cyclical Component
• Repeating up & down movements
• Due to interactions of factors influencing economy
• Usually 2-10 years duration

Cycle
Response

Time 29

29

30

30

15
 25-Nov-22

Time Series Analysis

• Time series forecasting models try to predict the future based on past
data.
• You can pick models based on:
1. Time horizon to forecast
2. Data availability
3. Accuracy required
4. Size of forecasting budget
5. Availability of qualified personnel

31

31

Forecasting Models
Forecasting Techniques

Qualitative Time-Series Methods Causal

Models Methods

Delphi Moving Regression Analysis

Methods Average

Jury of Executive Exponential Multiple

Opinion Smoothing Regression

Sales Force Trend

Composite Projections

Consumer
Decomposition
Market Survey
32

32

16
 25-Nov-22

Simple Moving Average Formula

• The simple moving average model assumes an average is a
good estimator of future behavior.
• The formula for the simple moving average is:

A t-1 + A t-2 + A t -3 +...+A t - n

Ft =
n
Ft = Forecast for the coming period
n = Number of periods to be averaged
A t-1 = Actual occurrence in the past period for
up to “n” periods
33

33

Simple Moving Average Problem (1)

A t-1 + A t-2 + A t-3 +...+A t- n
Week Demand Ft =
1 650
n
2 678 • Question: What are the 3-week
3 720 and 6-week moving average
4 785 forecasts for demand?
5 859
6 920 • Assume you only have 3 weeks
7 850 and 6 weeks of actual demand
8 758 data for the respective forecasts
9 892
10 920
11 789
12 844
34

34

17
 25-Nov-22

Calculating the moving averages gives us:

Week Demand 3-Week 6-Week
1 650 F4=(650+678+720)/3
2 678 =682.67
3 720 F7=(650+678+720
4 785 682.67 +785+859+920)/6
5 859 727.67 =768.67
6 920 788.00
7 850 854.67 768.67
8 758 876.33 802.00
9 892 842.67 815.33
10 920 833.33 844.00
11 789 856.67 866.50
12 844 867.00 854.83
35

35

Plotting the moving averages and comparing them

shows how the lines smooth out to reveal the overall
upward trend in this example.

ACTUAL TIME SERIES IS RATHER SMOOTH

1000

900
D em and
Demand

800
3 -W e e k
700
6 -W e e k
600

500
1 2 3 4 5 6 7 8 9 10 11 12
W eek
WHICH WOULD YOUR RECOMMEND?
1-WEEK, 3-WEEK, OR 6-WEEK? 36
Irwin/McGraw-Hill

36

18
 25-Nov-22

Simple Moving Average Problem (2) Data

• Question: What is the 3

week moving average
forecast for this data?
Week Demand • Assume you only have 3
1 820 weeks and 5 weeks of
2 775 actual demand data for
3 680 the respective forecasts
4 655
5 620
6 600
7 575

37

37

Simple Moving Average Problem (2)

Solution
Week Demand 3-Week 5-Week
1 820
2 775
3 680
4 655 758.33
5 620 703.33
6 600 651.67 710.00
7 575 625.00 666.00

38

38

19
 25-Nov-22

Weighted Moving Average Formula

While the moving average formula implies an equal
weight being placed on each value that is being
averaged, the weighted moving average permits an
unequal weighting on prior time periods.

The formula for the moving average is:

Ft = w 1 A t -1 + w 2 A t - 2 + w 3 A t -3 + ...+ w n A t- n
wt = weight given to time period “t” n
occurrence. (Weights must add to one.) w
i=1
i =1

39

39

Weighted Moving Average Problem (1)

Data
Question: Given the weekly demand and weights, what is
the forecast for the 4th period or Week 4?

Week Demand Weights:

1 650
2 678 t-1 .5
3 720 t-2 .3
4 t-3 .2

Note that the weights place more emphasis on the

most recent data, that is time period “t-1”.

40

40

20
 25-Nov-22

Weighted Moving Average Problem (1)

Solution

Week Demand Forecast

1 650
2 678
3 720
4 693.4

F4 = 0.5(720)+0.3(678)+0.2(650)=693.4
41

41

Weighted Moving Average Problem (2)

Data

Question: Given the weekly demand information and

weights, what is the weighted moving average
forecast of the 5th period or week?

Week Demand Weights:

1 820 t-1 .7
2 775 t-2 .2
3 680
t-3 .1
4 655

42

42

21
 25-Nov-22

Weighted Moving Average Problem (2)

Solution
Week Demand Forecast
1 820
2 775
3 680
4 655
5 672

F5 = (0.1)(755)+(0.2)(680)+(0.7)(655)= 672

43

43

Exponential Smoothing
• Exponential smoothing is easy to use and requires
little record keeping of data
• It is a type of moving average

New forecast = Last period’s forecast

+ (Last period’s actual demand
– Last period’s forecast)

Where  is a weight (or smoothing constant) with a

value between 0 and 1 inclusive

44

44

22
 25-Nov-22

45

45

Exponential Smoothing (Averaging)

Very Popular Very Effective!

Ft = Dt-1 + (1-)Ft-1
(alpha) times last actual value + (1-alpha) times last forecasted value

• D=Actual Value; F=Forecasted Value

• Premise--The most recent observations is normally a better

predict the next observation than are older observations.
• Therefore, we should give more weight to the more recent
time periods when forecasting
46

46

23
 25-Nov-22

Smoothing Constant

• Typical value of α are in the range of 0.1 to 0.5.

• If the time series is very volatile and contains substantial random

variability, a small value of the smoothing constant is preferred.

• For a fairly stable time series with relatively little random variability,
larger values of the smoothing constant have the advantage of quickly
adjusting the forecasts when error occur and therefore allowing the
forecast to react faster to changing conditions. Thus, larger values of α
place more emphasis on recent data.

47

47

Exponential Smoothing Example

• Determine exponential
Week Demand smoothing forecasts for
1 820 periods 2-10 using
2 775 =.10 and =.60.
3 680
4 655
5 750 • Let F1=D1
6 802
7 798
8 689
9 775
10
48

48

24
 25-Nov-22

D F0.1 F0.6

Alpha = 0.1 Week Demand 0.1 0.6

Week 1: F1 = D1 Assumption 1 820 W2 820.00 820.00
Week 2: (0.1) 820 + (0.9) 820 = 820 2 775 W3 820.00 820.00
Week 3: (0.1) 775 + (0.9) 820 = 815.5 3 680 815.50 793.00
Week 4: (0.1) 680 + (0.9) 815.5 = 801.5 4 655 801.95 725.20
5 750 787.26 683.08
6 802 783.53 723.23
7 798 785.38 770.49
8 689 786.64 787.00
9 775 776.88 728.20
10 776.69 756.28
49

49

Effect of  on Forecast

850
800
d 750 Demand
n
a 700 0.1
m650
e 600
D 0.6
550
500
1 2 3 4 5 6 7 8 9 10
Week

50

50

25
 25-Nov-22

Effect of Exponential forecasting on

forecast
Lagging Trend

51

Exponential Smoothing with

Trend Adjustment
• Like all averaging techniques, exponential smoothing
does not respond to trends
• A more complex model can be used that adjusts for
trends
• The basic approach is to develop an exponential
smoothing forecast then adjust it for the trend

Forecast including trend (FITt) = New forecast (Ft)

+ Trend correction (Tt)

52

52

26
 25-Nov-22

Exponential Smoothing with

Trend Adjustment
• The equation for the trend correction uses a new smoothing
constant 
• Tt is computed by
or Tt 1  (1   )Tt   ( Ft 1  Ft )

Tt  (1   )Tt 1   ( Ft  Ft 1 )
where
Tt+1 =smoothed trend for period t + 1
Tt =smoothed trend for preceding period
 = trend smooth constant that we select
Ft+1 =simple exponential smoothed forecast for period t + 1
Ft =forecast for pervious period

53

53

Problem: Portland Mfg.

• We consider a large Portland mfg. that uses exponential smoothing to forecast demand for a pollution
control equipment product. It appears that trend is present.

Month Demand Month Demand

1 12
2 17
3 20
4 19
5 24
6 26
7 31
8 32
9 36

• Smoothing constants are assigned the values of α = 0.2 and β = 0.4. Assume the initial forecast for the
month 1 was 11 units.

54

54

27
 25-Nov-22

Steps
Step 1: Forecast for month 2 (F2) = Forecast for month 1(F1) +
α(Month 1 demand – Forecast for month 1)
F2= 11 + 0.2(12-11) = 11.2 units
Step 2: Compute the trend present. Assume an initial trend adjustment of zero, that is T1
=0
T2= (1-β)T1 + β (F2-F1)
= 0 + 0.4 (11.2-11) = 0.08
Step 3: Compute the forecast including trend (FIT)
FIT2 = F2 + T2
= 11.2 + 0.08 = 11.28 units
We will do the same calculation for the third month also
Step 1. F3= F2+ α(Demand in month 2 – F2) = 11.2 + .2(17-11.2) = 12.36
Step 2. T3= (1-β)T2 + β (F3-F2) = (1-.4).08 + .4(12.36-11.2) = .51
Step 3. FIT3 = F3 + T3 = 12.36 + .51 = 12.87

55

55

Month Actual Forecast Ft Trend Adjusted FITt

Demand (without
trend)
1 12 11 0 11
2 17 11.20 .08 11.28
3 20 12.36 .51 12.87
4 19 13.89 .92 14.81
5 24 14.91 .96 15.87
6 26 16.73 1.30 18.03
7 31 18.58 1.52 20.10
8 32 21.07 1.91 22.98
9 36 23.25 2.02 25.27

56

56

28
 25-Nov-22

Selecting a Smoothing Constant

 As with exponential smoothing, a high value of  makes the forecast more
responsive to changes in trend
 A low value of  gives less weight to the recent trend and tends to smooth
out the trend
 Values are generally selected using a trial-and-error approach based on the
value of the MAD for different values of 
 Simple exponential smoothing is often referred to as first-order smoothing, or
Winter’s Method
 Trend-adjusted smoothing is called second-order, double smoothing, or Holt’s
method

57

Seasonal Variations
• Recurring variations over time may indicate the
need for seasonal adjustments in the trend line
• A seasonal index indicates how a particular
season compares with an average season
• When no trend is present, the seasonal index
can be found by dividing the average value for a
particular season by the average of all the data

58

58

29
 25-Nov-22

Seasonal Variation

Quarter Year 1 Year 2 Year 3 Year 4

1 45 70 100 100
2 335 370 585 725
3 520 590 830 1160
4 100 170 285 215
Total 1000 1200 1800 2200
Average 250 300 450 550

59

59

Seasonal Variation

Quarter Year 1 Year 2 Year 3 Year 4

1 45 70 100 100
2 335 370 585 725
3 520 590 830 1160
4 100 170 285 215
Total 1000 1200 1800 2200
Average 250 300 450 550

Actual Demand 45
Seasonal Index = = = 0.18
Average Demand 250
Forecast for Year 5 = 2600 = 650 for each quarter
60

60

30
 25-Nov-22

Seasonal Variation

Quarter Year 1 Year 2 Year 3 Year 4

1 45/250 = 0.18 70/300 = 0.23 100/450 = 0.22 100/550 = 0.18
2 335/250 = 1.34 370/300 = 1.23 585/450 = 1.30 725/550 = 1.32
3 520/250 = 2.08 590/300 = 1.97 830/450 = 1.84 1160/550 = 2.11
4 100/250 = 0.40 170/300 = 0.57 285/450 = 0.63 215/550 = 0.39

Quarter Average Seasonal Index Forecast

1 (0.18 + 0.23 + 0.22 + 0.18)/4 = 0.20 650(0.20) = 130
2 (1.34 + 1.23 + 1.30 + 1.32)/4 = 1.30 650(1.30) = 845
3 (2.08 + 1.97 + 1.84 + 2.11)/4 = 2.00 650(2.00) = 1300
4 (0.40 + 0.57 + 0.63 + 0.39)/4 = 0.50 650(0.50) = 325

61

61

Eg. Seasonal Variations

• Eicher Supplies sells telephone answering machines

• Data has been collected for the past two years sales of one particular model

• They want to create a forecast that includes seasonality

• The expected annual demand for answering machines in the third year is 1200
unit, Eicher want to forecast the monthly demand for third period.

62

62

31
 25-Nov-22

Eg. Seasonal Variations

AVERAGE
SALES DEMAND
AVERAGE TWO- MONTHLY SEASONAL
MONTH YEAR 1 YEAR 2 YEAR DEMAND DEMAND INDEX
January 80 100 90 94 0.957
February 85 75 80 94 0.851
March 80 90 85 94 0.904
April 110 90 100 94 1.064
May 115 131 123 94 1.309
June 120 110 115 94 1.223
July 100 110 105 94 1.117
August 110 90 100 94 1.064
September 85 95 90 94 0.957
October 75 85 80 94 0.851
November 85 75 80 94 0.851
December 80 80 80 94 0.851
Total average demand = 1,128
1,128 Average two-year demand
Average monthly demand = = 94 Seasonal index = Average monthly demand
12 months
Table 63

63

Eg. Seasonal Variations

• The calculations for the seasonal indices are
1,200 1,200
Jan.  0.957  96 July  1.117  112
12 12
1,200 1,200
Feb.  0.851  85 Aug.  1.064  106
12 12
1,200 1,200
Mar.  0.904  90 Sept.  0.957  96
12 12
1,200 1,200
Apr.  1.064  106 Oct.  0.851  85
12 12
1,200 1,200
May  1.309  131 Nov.  0.851  85
12 12
1,200 1,200
June  1.223  122 Dec.  0.851  85
12 12 64

64

32
 25-Nov-22

Seasonal Variations with Trend

• When both trend and seasonal components are present,
the forecasting task is more complex
• Seasonal indices should be computed using a centered
moving average (CMA) approach
• There are four steps in computing CMAs
1. Compute the CMA for each observation (where
possible)
2. Compute the seasonal ratio = Observation/CMA for
that observation
3. Average seasonal ratios to get seasonal indices
4. If seasonal indices do not add to the number of
seasons, multiply each index by (Number of
seasons)/(Sum of indices)

65

65

ABC Industries Example

 The following are ABC Industries’ sales figures for
the past three years

QUARTER YEAR 1 YEAR 2 YEAR 3 AVERAGE

1 108 116 123 115.67
2 125 134 142 133.67
3 150 159 168 159.00
4 141 152 165 152.67
Average 131.00 140.25 149.50 140.25

Seasonal
Definite trend pattern

66

33
 25-Nov-22

ABC Industries Example

 To calculate the CMA for quarter 3 of year 1 we
compare the actual sales with an average quarter
centered on that time period
 We will use 1.5 quarters before quarter 3 and 1.5
quarters after quarter 3 – that is – we take quarters
2, 3, and 4 and one half of quarters 1, year 1 and
quarter 1, year 2

0.5(108) + 125 + 150 + 141 + 0.5(116)

CMA(q3, y1) = = 132.00
4

67

ABC Industries Example

 We compare the actual sales in quarter 3 to the
CMA to find the seasonal ratio

Sales in quarter 3 150

Seasonal ratio    1.136
CMA 132

QUARTER YEAR 1 YEAR 2 YEAR 3 AVERAGE

1 108 116 123 115.67
2 125 134 142 133.67
3 150 159 168 159.00
4 141 152 165 152.67
Average 131.00 140.25 149.50 140.25

68

34
 25-Nov-22

ABC Industries Example

YEAR QUARTER SALES CMA SEASONAL RATIO
1 1 108
2 125
3 150 132.000 1.136
4 141 134.125 1.051
2 1 116 136.375 0.851
2 134 138.875 0.965
3 159 141.125 1.127
4 152 143.000 1.063
3 1 123 145.125 0.848
2 142 147.875 0.960
3 168
4 165

Table

69

ABC Industries Example

 There are two seasonal ratios for each quarter so
these are averaged to get the seasonal index

Index for quarter 1 = I1 = (0.851 + 0.848)/2 = 0.85

Index for quarter 2 = I2 = (0.965 + 0.960)/2 = 0.96
Index for quarter 3 = I3 = (1.136 + 1.127)/2 = 1.13
Index for quarter 4 = I4 = (1.051 + 1.063)/2 = 1.06

70

35
 25-Nov-22

ABC Industries Example

 Scatter plot of ABC Industries data and CMAs

200 – CMA

  
150 –  
  
  
Sales

100 – 

50 –
Original Sales Figures
0– | | | | | | | | | | | |

1 2 3 4 5 6 7 8 9 10 11 12
Time Period
Figure 6

71

The Decomposition Method of Forecasting

• Decomposition is the process of isolating linear trend and

seasonal factors to develop more accurate forecasts
• There are five steps to decomposition
1. Compute seasonal indices using CMAs
2. Deseasonalize the data by dividing each number by its
seasonal index
3. Find the equation of a trend line using the
deseasonalized data
4. Forecast for future periods using the trend line
5. Multiply the trend line forecast by the appropriate
seasonal index

72

72

36
 25-Nov-22

ABC Industries –
Decomposition Method
SALES SEASONAL DESEASONALIZED
($1,000,000s) INDEX SALES ($1,000,000s)
108 0.85 127.059
125 0.96 130.208
150 1.13 132.743
141 1.06 133.019
116 0.85 136.471
134 0.96 139.583
159 1.13 140.708
152 1.06 143.396
123 0.85 144.706
142 0.96 147.917
168 1.13 148.673
165 1.06 155.660
Table 11 73

73

ABC Industries –
Decomposition Method

 Find a trend line using the deseasonalized data

b1 = 2.34 b0 = 124.78

 Develop a forecast using this trend and multiply

the forecast by the appropriate seasonal index
Ŷ = 124.78 + 2.34X
= 124.78 + 2.34(13)
= 155.2 (forecast before adjustment for
seasonality)

Ŷ x I1 = 155.2 x 0.85 = 131.92

74

74

37
 25-Nov-22

SunHospital Example
 A Sun hospital used 66 months of adult inpatient
days to develop the following seasonal indices

MONTH SEASONALITY INDEX MONTH SEASONALITY INDEX

January 1.0436 July 1.0302
February 0.9669 August 1.0405
March 1.0203 September 0.9653
April 1.0087 October 1.0048
May 0.9935 November 0.9598
June 0.9906 December 0.9805

75

75

Sun Hospital Example

 Using this data they developed the following
equation
Ŷ = 8,091 + 21.5X
where
Ŷ = forecast patient days
X = time in months

 Based on this model, the forecast for patient days

for the next period (67) is

Patient days = 8,091 + (21.5)(67) = 9,532 (trend only)

Patient days = (9,532)(1.0436)
= 9,948 (trend and seasonal) 76

76

38
 25-Nov-22

Forecasting Models
Forecasting Techniques

Qualitative Time-Series Methods Causal

Models Methods

Delphi Moving Regression Analysis

Methods Average

Jury of Executive Exponential Multiple

Opinion Smoothing Regression

Sales Force Trend

Composite Projections

Consumer
Decomposition
Market Survey
77

77

Causal Models
• Causal models use variables or factors that
might influence the quantity being forecasted
• The objective is to build a model with the best
statistical relationship between the variable
being forecast and the independent variables
• Regression analysis is the most common
technique used in causal modeling

78

78

39
 25-Nov-22

Simple Linear Regression Model

Yt = a + bx Y

0 1 2 3 4 5 x (weeks)

• b is similar to the slope. However, since it is

calculated with the variability of the data in mind, its
formulation is not as straight-forward as our usual
notion of slope
79

79

Calculating a and b
a = y - bx

 xy - n(y)(x)
b= 2 2
 x - n(x )
OR

b
 ( X  X )(Y  Y )
(X  X ) 2
80

80

40
 25-Nov-22

Regression Equation Example

Week Sales
1 150
2 157
3 162
4 166
5 177

Develop a regression equation to predict sales

based on these five points.
81

81

x x*x y x*y
1 1 150 150
2 4 157 314
3 9 162 486
4 16 166 664
5 25 177 885
Average Sum Average Sum
3 55 162.4 2499

b=
 xy - n(y)(x) = 2499 - 5(162.4)(3)  63 = 6.3
 x - n(x )
2 2
55  5(9 ) 10

a = y - bx = 162.4 - (6.3)(3) = 143.5

82

82

41
 25-Nov-22

y = 143.5 + 6.3t
180
175
170
165
160 Sales
Sales

155 Forecast
150
145
140
135
1 2 3 4 5

Period
83

83

Regression with Trend and Seasonal

Components
• Multiple regression can be used to forecast both trend and
seasonal components in a time series
• One independent variable is time
• Dummy independent variables are used to represent the seasons
• The model is an additive decomposition model

Yˆ  a  b1 X 1  b2 X 2  b3 X 3  b4 X 4
where
X1 = time period
X2 = 1 if quarter 2, 0 otherwise
X3 = 1 if quarter 3, 0 otherwise
X4 = 1 if quarter 4, 0 otherwise

84

84

42
 25-Nov-22

Regression with Trend and Seasonal

Components
• The resulting regression equation is
Yˆ  104.1  2.3 X 1  15.7 X 2  38.7 X 3  30.1X 4

 Using the model to forecast sales for the first two

quarters of next year

Ŷ  104.1  2.3(13)  15.7(0)  38.7(0)  30.1(0)  134

Ŷ  104.1  2.3(14 )  15.7(1)  38.7(0)  30.1(0)  152

 Use MAD and MSE to determine the best model

85

85

Measures of Forecast Accuracy

• We compare forecasted values with actual values to see
how well one model works or to compare models

Forecast error = Actual value – Forecast value

 One measure of accuracy is the mean absolute deviation (MAD)

MAD 
 forecast error
n

86

86

43
 25-Nov-22

Measures of Forecast Accuracy

• There are other popular measures of forecast accuracy
• The mean squared error

MSE 
 (error) 2

n
 The mean absolute percent error
error
 actual
MAPE  100% OR
n
MAPE | error| / | actual| *100
 And bias is the average error 87

87

Monitoring and Controlling Forecasts

 Tracking signals can be used to monitor the

performance of a forecast
 Tacking signals are computed using the following
equation
RSFE
Tracking signal 
MAD
where

MAD 
 forecast error
n

Running sum of the forecast errors (RSFE)

88

88

44
 25-Nov-22

Monitoring and Controlling Forecasts

Signal Tripped
Upper Control Limit Tracking Signal
+

Acceptable
0 MADs Range

–
Lower Control Limit

Time

89

89

Aggregate Forecasts

• Aggregate forecasts are more accurate

• Forecast at the most aggregate/generic level possible
• Similarly, forecast at the most upstream of the supply chain (if
possible)
• If possible, never use forecast information at the lower levels. At the
lower levels, decisions should be based on actual demand

90

90

45
 25-Nov-22

Case: HP desktop (Aggregate Forecast)

• Is it always possible to use it?

• Only if the power supply can be assembled in small lead time
• Power supply assembly should be at the end of the
manufacturing process
Power
supply Testing
110 V
Board Hard disk
assembly Assembly Power
supply Testing
220 V
Delayed product
differentiation

Product
Power postponement
supply
110 V
Board Hard disk
Testing
assembly assembly
Power
supply 91
220 V

91

Euro Plugs
• No standardized
power supplies for
Europe
• Different power
supply for every
country.

92

92

46
 25-Nov-22

Case: HP Desktop

Produc
t
110 V 220 V
Power Months PC PC
supply 1 10000 8000
110 V
2 14000 4000
Board Hard disk 3 16000 2500
assembly assembly
Testing
4 12000 6500
Power 5 18000 2000
supply
6 15000 4000
220 V
7 14000 3000
8 11000 7000
9 13000 5000
10 11000 6000

93

Forecast accuracy at different levels

110 V 220 V Total

Months Demand MA(4) Error Demand MA(4) Error Demand MA(4) Error
1 10000 8000 18000
2 14000 (10000+14000+16000+12000)/4) 4000 18000
3 16000 18000-13000
2500 18500
4 12000 6500 18500
5 18000 13000 5000 2000 5250 -3250 20000 18250 1750
6 15000 15000 0 4000 3750 250 19000 18750 250
7 14000 15250 -1250 3000 3750 -750 17000 19000 -2000
8 11000 14750 -3750 7000 3875 3125 18000 18625 -625
9 13000 14500 -1500 5000 4000 1000 18000 18500 -500
10 11000 13250 -2250 6000 4750 1250 17000 18000 -1000
MAD 2291.67 1604.17 1020.83

Forecast
Accuracy 83.23% 64.35% 94.38%

100-[(5+1.25+3.75+1.5+2.25)/(18+15+14+11+13+11)]100
(5000+1250+3750+1500+2250) / 6

94

47
 25-Nov-22

Aggregate VS. Item Forecasts

• AGGREGATE • ITEM FORECASTS

• CORP. OR PRODUCT LINE • INDIVIDUAL ITEMS

SALES
• NEW PRODUCTS
• STRATEGIC
• CONSEQUENCES GREAT
• MAY BE EXPENSIVE
• 100’S OF FORECASTS
• AUTOMATED SYSTEMS FORECAST
• TACTICAL
• CONSEQUENCES NOT GREAT
• COST/ITEM IS LOW
• MILLIONS OF FORECASTS

95

95

Monitoring and Controlling Forecasts

 Positive tracking signals indicate demand is greater than

forecast
 Negative tracking signals indicate demand is less than
forecast
 Some variation is expected, but a good forecast will have
about as much positive error as negative error
 Problems are indicated when the signal trips either the
upper or lower predetermined limits
 This indicates there has been an unacceptable amount of
variation
 Limits should be reasonable and may vary from item to
item

96

96

48
 25-Nov-22

Kim’s Bakery Example

 Tracking signal for quarterly sales of cakes

TIME FORECAST ACTUAL |FORECAST | CUMULATIVE TRACKING

PERIOD DEMAND DEMAND ERROR RSFE | ERROR | ERROR MAD SIGNAL
1 100 90
2 100 95
3 100 115
4 110 100
5 110 125
6 110 140

97

97

The Bottom Line

Forecasting methods are based upon past behavior.

If future behavior is significantly different from that of the past,

forecasting methods will not work well.

98

98

49
 25-Nov-22

Realities of Forecasting
• Forecasts are seldom perfect
• Most forecasting methods assume that there is
some underlying stability in the system
• Both product family and aggregated product
forecasts are more accurate than individual
product forecasts

99

99

Kim’s Bakery Example

 Tracking signal for quarterly sales of cakes

TIME FORECAST ACTUAL |FORECAST | CUMULATIVE TRACKING

PERIOD DEMAND DEMAND ERROR RSFE | ERROR | ERROR MAD SIGNAL
1 100 90 –10 –10 10 10 10.0 –1
2 100 95 –5 –15 5 15 7.5 –2
3 100 115 +15 0 15 30 10.0 0
4 110 100 –10 –10 10 40 10.0 –1
5 110 125 +15 +5 15 55 11.0 +0.5
6 110 140 +30 +35 30 85 14.2 +2.5

MAD 
 forecast error  85  14.2
n 6
RSFE 35
Tracking signal    2.5MADs
MAD 14.2
100

100

50
 25-Nov-22

Some Features
u Depending upon the situation, judgmental or qualitative
forecasts may be best.
u “Group” forecasts are usually more accurate than individual
forecasts (e.g., product line versus individual products).
u Forecast accuracy decreases as the time horizon increases

101

101

Judgmental Forecasting Applications

Small and Large Firms

Low High
Sales Sales
Technique < $100M > $500M
Manager’s opinion 40.7% 39.6%
Jury of executive opinion 40.7% 41.6%
Sales force composite 29.6% 35.4%
Number of Firms 27 48

Source: Nada Sanders and Karl Mandrodt (1994) “Practitioners Continue to Rely on Judgmental Forecasting
Methods Instead of Quantitative Methods,” Interfaces, vol. 24, no. 2, pp. 92-100.

102

102

51
 25-Nov-22

Quantitative Forecasting Applications

Small and Large Firms

Low High
Sales Sales
Technique < $100M > $500M
Moving average 29.6% 29.2%
Straight line projection 14.8% 14.6%
Naive 18.5% 14.6%
Exponential smoothing 14.8% 20.8%
Regression 22.2% 27.1%
Simulation 3.7% 10.4%
Classical decomposition 3.7% 8.3%
OTHER 3.7% 6.3%
Number of Firms 27 48
Source: Nada Sanders and Karl Mandrodt (1994) “Practitioners Continue to Rely on Judgmental Forecasting
Methods Instead of Quantitative Methods,” Interfaces, vol. 24, no. 2, pp. 92-100.

103

103

Adaptive Smoothing

• Adaptive smoothing is the computer monitoring of tracking signals

and self-adjustment if a limit is tripped
• In exponential smoothing, the values of  and  are adjusted when
the computer detects an excessive amount of variation

104

104

52
 25-Nov-22

Using The Computer to Forecast

• Spreadsheets can be used by small and medium-sized forecasting

problems
• More advanced programs (SAS, SPSS, Minitab) handle time-series and
causal models
• May automatically select best model parameters
• Dedicated forecasting packages may be fully automatic
• May be integrated with inventory planning and control
• Web-based forecasting (CPFR)

105

105

Do it yourself: Problem Set

106

106

53
 25-Nov-22

Problem 1
• Sales of industrial vacuum cleaners at Ram corp. over the
past 13 months are shown below.
Sales Month Sales Month
(thousands) (thousands)
11 January 14 August
14 February 17 September
16 March 12 October
10 April 14 November
15 May 16 December
17 June 11 January
11 July

(a) Using a moving average with three periods, determine the demand for
vacuum cleaners for next February.
107

107

Problem 1 (Contd.)
(b) Using a weighted moving average with three periods, determine the demand for vacuum
cleaners for February. Use 3,2, and 1 for the weights of the most recent, second most recent, and
third most recent periods, respectively.
(c) Evaluate the accuracy of each of these methods
(d) What other factors might Ram consider in forecasting sales?

108

108

54
 25-Nov-22

Problem 2
• Sales of Cool-Man air conditioner have grown steadily
during the past five years.
Years Sales
1 450
2 495
3 518
4 563
5 584
• The6sales manager had predicted? before the business started,
that year 1’s sales would be 410 air conditioner. Using
exponential smoothing with an alpha weight of 0.30, develop
forecasts for years 2 through 6.

109

109

Problem 3
• Passenger miles flown on Indigo Airlines, are shown below for the past
12 weeks.
Week Actual Passenger Miles (1,000s)
1 17
2 21
3 19
4 23
5 18
6 16
7 20
8 18
9 22
10 20
11 15
12 22 110

110

55
 25-Nov-22

Problem 3 (Contd.)

(a) Assuming a initial forecast for week 1 of 17,000 miles, use exponential smoothing to compute
miles for weeks 2 through 12. Use alpha =0.2
(b) What is the MAD for this model?

111

111

Problem 4

• Room registration in the Chandra Plaza Hotel have been recorded for the past nine years.
Management would like to determine the mathematical trend of guest registration in order to
project future occupancy. This estimate would help the hotel determine whether a future
expansion will be needed. Given the following time series data, develop a regression equation
relating registration to time. Then forecast next year’s registration. Room registration are in
thousands:
Year 1: 17, Year 2: 16, Year 3:16, Year 4: 21, Year 5: 20
Year 6: 20, Year 7: 23, Year 8: 25, Year 9: 24

112

112

56
 25-Nov-22

113

113

“I think there is a world market for maybe five

computers”
Thomas Watson, Chairman of IBM, 1943

“640,000 bytes of memory ought to be enough for

anybody”
Bill Gates, 1981

“The Internet will catastrophically collapse in 1996”

Robert Metcalfe, Inventor of the Internet

114

57
 25-Nov-22

THANK YOU

115

115

58