OME Lec.

3:
Operation Research 2

Forecasting
Basic Concepts
What is a Forecast?
Prediction
Estimate
Determination
Major Categories of
Forecasting Time Horizons
Short-term Forecast
Intermediate-term
Forecast
Long-term Forecast.
 Forecasting Techniques
• Qualitative Techniques
• Time Series Analysis
• Causal Methods
Factors in Forecasting

• Trend.
• Seasonal Factors
• Cyclical Factors
• Random Factors
 Forecasting Methods
or Time Series Methods
1. Simple Moving Average
2. Weighted Moving Average
3. Simple Exponential
Smoothing
4. Adjusted Exponential
Smoothing
5. Forecast Reliability
 Simple Moving Average
Simple Moving Average is the un-weighted
average of a consecutive number of data points. It is
a forecasting method simply eliminates the effects of
seasonal, cyclical, and erratic fluctuations by getting
historical data. Thus, if seasonality, trend and
cyclical factors are not critical in the variable being
forecast, the moving-average method is an
appropriate tool. It can be used as a forecast
seasonal adjustment of the data. To calculate a
simple moving average, we simply choose the
number of items in the time series data to include in
the average. Then, as each time period changes, add
the new period time and eliminate the oldest time-
period data, and calculate a new average. It is being
computed using the formula…
 or


 Where:



 Example:
 The WSS motorcycle dealer in Quezon
Avenue area wants to accurately
forecast the demand for the WSS hybrid
motorcycle during the next month.
Because the distributor is in Germany, it
is difficult to send back or recorded if
the proper number of motorcycles is not
ordered a month ahead. From sales
records, the dealer has accumulated the
following data for the past 11 months.
Compute the three and five-month
moving average forecast on demand.
Month J F M A M J J A S O N
Motorcycles 60 70 50 90 10 80 150 70 110 150 130

Solution
 Computation for the Forecast on 3 - Month
Moving Average:

 May = 70 September = 100
 June = 50 October = 110
 July = 60 November = 110
 August = 80 December = 130
  
Month J F M A M J J A S O N
Motorcycles 60 70 50 90 10 80 150 70 110 150 130

Solution
 Computation for the Forecast on 5 - Month
Moving Average:

 July = 60 October = 84
 August = 76 November = 112
 September = 80 December = 122
  
 Table for the Three- and Five-Month
Moving Averages
Actual 3-Month Moving 5-Month Moving
Period Month
Demand Average Average
1 January 60 -- --
2 February 70 -- --
3 March 50 -- --
4 April 90 60 --
5 May 10 70 --
6 June 80 50 56
7 July 150 60 60
8 August 70 80 76
9 September 110 100 80
10 October 150 110 84
11 November 130 110 112
12 December -- 130 122
Graph:
160

140

120

100

80

60

40

20

0
ry ry ch il ay e ly t r r r r
u a u a a r pr M un Ju gu
s be o be be be
A J m m m
an br M A
u te ct e e
J Fe e p O ov D
ec
S N

Actual Demand 3-Month Moving Average 5-Month Moving Average

 Weighted Moving Averages
 A weighted moving average is a time series forecasting
method in which the most recent data are weighted
heavier compared to later data. This is desirable to vary
the weights given to historical data forecast future
demand or sales. Smoothing Constant is a weighing
factor used in the exponential smoothing forecast
technique. Mathematically, the weighted moving average
is computed as follows:

 where:


 Example:
The WSS motorcycle dealer in Quezon Avenue area
wants to accurately forecast the demand for the WSS
hybrid motorcycle during the next month. Because the
distributor is in Germany, it is difficult to send
motorcycle back or recorded if the proper number of
motorcycles is not ordered a month ahead. From sales
records, the dealer has accumulated the following data
for the past 11 months. Determine the weighted
moving averages forecast on demand with the
following weights (a) 20%, 30%, and 50% and (b)
15%, 25%, and 60%.

Month Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov

Motorcycles 60 70 50 90 10 80 150 70 110 150 130

Solution:
 Computation of Weighted Moving Average
(20%, 30%, 50%)

 May = 74
 June = 42
 July = 61
 August = 101
 September = 96
 October = 106
 November = 122
 December = 132
 Computation of Weighted Moving Average
(15%, 25%, 60%)

 May = 77
 June = 36.6
 July = 64
 August = 111.5
 September = 91.5
 October = 106
 November = 128
 December = 132
 Demand Forecast
Using Weighted Moving Average
Actual
Period Month WMA (20%, 30%, 50%) WMA (15%, 25%, 60%)
Demand

1 January 60 -- --

2 February 70 -- --

3 March 50 -- --

4 April 90 58 56.5
5 May 10 74 77.0
6 June 80 42 36.0
7 July 150 61 64.0
8 August 70 101 111.5
9 September 110 96 91.5
10 October 150 106 106.0
11 November 130 122 128.0
12 December -- 132 132.0
160

140

120

100

80

60

40

20

0
ry ry ch il ay e ly s t r r r r
u a u a a r pr M un Ju gu be o be be be
A J m m m
an br M A
u te ct e e
J Fe e p O ov D
ec
S N

Actual Demand WMA (20%, 30%,50%) WMA(15%, 25%, 60%)

 Simple Exponential Smoothing
 Exponential Smoothing refers to family of
forecasting models that are very similar to the
weighted moving average that weights the most
recent past data more than distant past data. The
value of is between 0.00 and 1.00. The value of
determines the degree of smoothing that takes place
and how responsive the model is to fluctuations of
the variable being forecast. The setting of is
typically not specific and is usually done by trial and
error. The formula simplest exponential smoothing
model is of the following terms:

 or
 where:
.

 Example:
 The WSS motorcycle dealer in Quezon Avenue area
wants to accurately forecast the demand for the
WSS hybrid motorcycle during the next month.
Because the distributor is in Germany, it is difficult
to send motorcycle back or recorded if the proper
number of motorcycles is not ordered a month
ahead. From sales records, the dealer has
accumulated the following data for the past 11
months. Establish the forecasts using simple
exponential smoothing if and
Month Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov
Motorcycles 60 70 50 90 10 80 150 70 110 150 130

Solution:
If we will use , therefore , then substitute to the exponential
smoothing formula.

→ (take note of this one)

59.90
62.91
57.62
59.86
68.87
68.98
73.08
80.77
85.69
 
Month Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov
Motorcycles 60 70 50 90 10 80 150 70 110 150 130

 When therefore , then substitute to simple

exponential smoothing formula


 59.10
 68.37
 50.86
 59.60
 86.72
 81.70
 90.19
 108.13
 114.69
 Simple Exponential Smoothing
Month Actual Demand

January 60 -- --

February 70 60.00 60.00

March 50 61.00 63.00
April 90 59.90 59.10
May 10 62.91 68.37
June 80 57.62 50.86
July 150 59.86 59.60
August 70 68.87 86.72
September 110 68.98 81.70
October 150 73.08 90.19
November 130 80.77 108.13
December -- 85.69 114.69
Graph
160

140

120

100

80

60

40

20

0
y ry ch ril ay e ly st be
r er er er
ar ru
a a r Ap M J un Ju g u o b b b
n u b M Au t em ct v em cem
Ja Fe e p O
No De
S

Actual Demands α=0.10 α=0.30

 Adjusted Exponential Smoothing
The exponential smoothing forecasting
technique adjusted for trend changes and
seasonal patterns. It consists of simple
exponential smoothing forecast with trend
adjustment factor added to it. The value of is a
value is also between 0.00 to 1.00 similar to .
It reflects the weight given to the recent data.
In addition, both and is often determine
subjectively based on the judgment of the
forecaster. Mathematically, the adjusted
exponential smoothing model can be described
as follows…
Where:
Example:
The WSS motorcycle dealer in Quezon
Avenue area wants to accurately forecast the
demand for the WSS hybrid motorcycle during
the next month. Because the distributor is in
Germany, it is difficult to send motorcycle back
or recorded if the proper number of motorcycles
is not ordered a month ahead. From sales
records, the dealer has accumulated the
following data for the past 11 months.
Establish the forecasts using adjusted
smoothing if
Month Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov
Motorcycles 60 70 50 90 10 80 150 70 110 150 130

Solution:
Using , therefore substitute to adjusted
forecast:
 We will Let

0.20(61 – 60) + 0.80(0)

T3 0.20

61 + 4(0.20)
61.80

0.20( 59.90 – 61) + 0.80(0.2)

T4 – 0.06

59.66
 Month Actual Demand Adjusted Forecast

January 60 --
February 70 60.00
March 50 61.80
April 90 59.66
May 10 65.11
June 80 55.14
July 150 59.66
August 70 75.91
September 110 74.70
October 150 80.42
November 130 93.21
December -- 99.57
Graph:
160

140

120

100

80

60

40

20

0
a ry ar
y
ch p ril ay ne ly u st b er b er ber b er
u u a r A M J u Ju g m o
n br M Au pte ct em cem
Ja Fe O v
Se No De
Actual Adjusted Forecast
Learning is an endless act of
curiosity…
Sir Anre

Thank You
And
Good Day…