Forecasting

Forecasting is a process of predicting a

future event. It is the underlying basis of all
business decisions such as: (1) production,
(2) inventory, (3) personnel, (4) facilities, etc.
NEED OF FORECASTING

• When there is a lag between awareness of

an impending event or need and
occurrence of that event.
• Forecasting forms the basis of planning.
• It is essential for the organization to know
for what level of activities one is planning
before investments in input.
 TYPES OF
FORECASTING
• Short Term Forecasting is the
forecasting that made for short term
objectives covering less than one year. Ex.
Material Requirement Planning (MRP),
scheduling, sequencing, budgeting, etc.
• Long Term Forecasting is the forecasting
that made for long term objectives
covering more than five years. Ex. Product
diversification, sales and advertisement.
Forecasting methods can be divided in to
three main categories:

•Extrapolative or Time-series Methods

•Casual or explanatory methods
•Qualitative or judgmental methods
A. Extrapolative or Time-
series Methods
1. Naïve forecast
2. Simple moving average
3. Weighted moving average
4. Exponential smoothing
 Naïve forecast

Assumes demand in the next period is the

same as demand in most recent period.
 Examples:

1. If May sales were 48 cakes, then June sales will be 48

cakes also.
2. Forecast the order for the month of May by Naïve
approach.
MONTH ORDER PER FORECAST
MONTH
January 120 -
February 90 120
March 100 90
April 75 100
May - 75
Solution: Forecast order for the month of May, (F)May = 75 units
Simple moving average
-Uses average demand for a fixed sequence of
periods.
-Techniques that averages a number of recent
actual values, updated as new values become
available
-A moving average forecast uses a number of the
most recent actual data values in generating a
forecast.
-It can be computed using the following equation:
 Ft = forecast for period t
Man = n period moving average
At-1 = actual value in period t-i
n = number of periods (data points) in the moving
average
For example, MA3 would refer to a three-period moving average
forecast, and MA5 would refer to a five-period moving average.
 Examples:

1. Compute a three-period moving average

forecast given demand for shopping carts
for the last five periods.
PERIOD DEMAND FORECAST

1 42
2 40
3 43
4 40
5 41
6
7
 Solution:

F6 = 43 + 40 + 41
3
F6 = 41.33
If actual demand in period 6 turns out to be 38, the
moving average forecast for period 7 would be:
F7 = 40 + 41 + 38
3
F7 = 39.67
 Weighted moving
average
More recent values in a series are given more weight in
computing a forecast.
A weighted average is similar to a moving average, except
that it assigns more weight to the most recent values in a
times series.
For instance, the most recent value might be assigned a
weight of .40, the next most recent value a weight of .30,
the next after that a weight of .20, and the next after that a
weight of .10.
Note that weights must sum to 1.00, and that the heaviest
weights are assigned to the most recent values.
 Where:

wt = weight for the period t, wt-1= Weight for

period t-1, etc.
At = actual value in period t, At-1= Actual
value for period t-1, etc
 Examples:
Compute a weighted average forecast using a
weight of .40 for the most recent period, .30 for the
next most recent, .20 for the next, and .10 for the
next. Or the problem will be: Compute the for 4-
month weighted moving average for period 6

PERIOD DEMAND FORECAST

1 42
2 40
3 43
4 40
5 41
6
7
 Solution:
F6 = .10(40) + .20(43) + .30(40) + .40(41)
F6 =41
Or
F6 = (41 x 4) + (40 x 3) + (43 x 2) + (40 x 1)
(4+3+2+1) – for weight
F6 = 164 + 120 + 86 + 40
10
F6 = 410
10
F6= 41
If the actual demand for period 6 is 39, forecast
demand for period 7 using the same weights as in
part a.
F7 = .10(43) + .20(40) +.30(41) + .40(39)
Or = (39 x 4) + (41 x 3) + (40 x 2) + (43 x 1)
10
F7 = 40.2
Exponential smoothing

A weighted averaging method based on previous

forecast plus a percentage of the forecast error.
Exponential smoothing is a sophisticated weighted
averaging methods that is still relatively easy to
use and understand.
Each new forecast is based on the previous
forecast plus a percentage of the difference
between that forecast and the actual value of the
series at that point.
That is,
Next forecast = Previous forecast + α (Actual demand for the
previous period – Previous Forecast)
- or-
Ft = (1 – α ) Ft-1 + α At-1
Where:
Ft = forecast for period t
Ft-1 = forecast for the previous period t
α = smoothing constant (percentage) At-1 = actual demand or
sales for the previous period
The smoothing constant α represents a percentage of the
forecast error.
 Examples

1. Suppose the previous forecast was 42

units, actual demand was 40 units, and α
= .40. The new forecast would be
computed as follows:

Solution:
Ft = 42 + .40 (40 – 42)
Ft = 41.20
Alternative solution:
Ft = (1 – .40) (42) + .40 (40)
= .60 (42) + .40 (40)
Ft = 41.20
If the actual demand turns out to be 43, then
the next forecast would be:
Ft = 41.2 + .40 (43 – 41.2)
Ft = 41.92
Alternative solution:
Ft = (1 – .40) (41.2) + .40 (43)
= .60 (41.2) + .40 (43)
Ft= 41.92
B. Casual or Explanatory
Methods
Simple Linear Regression Model
– The object in linear regression is to obtain an equation
of a straight line that minimizes the sum of squared
vertical deviations of data points from the line (i.e., the
least squares criterion). This least squares linehas the
equation:
y = a+bx
Where:
y=Predicted (dependent) variable
x = Predictor (independent) variable
b = Slope of the line
a = Value of y c when x = 0 (i.e., the height of the line at the y intercept)
The coefficients a andb of the line are based
on the following two equations:
 Examples

The sales of a company (in million dollars) for each year

are shown in the table below.
x (year) 2005 2006 2007 2008 2009
y (sales) 12 19 29 37 45

Find the least square regression line y = ax + b

Use the least square regression as a model to forecast the
sales of the company in 2012?
a.We first change the variable x to t such that t = x – 2005 and
therefore t represents the number of years after 2005. Using t instead
of x makes the numbers smaller and therefore manageable. The table
of values becomes:

t (years after 2005) 0 1 2 3 4

y (sales, in million dollars) 12 19 29 37 45
We now use the table to calculate a and b
included in the least regression line formula.

t y ty t2
0 12 0 0
1 19 19 1
2 29 58 4
3 37 111 9
4 45 180 16
∑t = 10 ∑y = 142 ∑ty = 368 ∑t2 = 30
 We now calculate a and b using the least
square regression formula for a and b.

b = [(5)(368) – (10)(142)] / [(5)(30) – (10)2]

b = 8.4

a = [(142) – 8.4(10)] / 5
a = 11.6
In 2012, t = 2012 – 2005 = 7

The forecasted sales in 2012 are:

y = a + bx
= 11.6 + 8.4(7)
= 70.4 million dollars
 Qualitative or
Judgmental Methods
Delphi Method
•The Delphi method is a process of gaining
consensus from a group of experts while
maintaining their anonymity. It is forecasting
techniques applied to subjective nature
demand values. It is useful when there is no
historical data from which to develop
statistical models and when managers
inside the firm have no experience.
Market Research

– It is systematic approach to determine

external consumer interest in a service or
product by creating and testing hypothesis
through data-gathering surveys.
FORECAST ACCURACY

Forecast Error is the difference between the

value that occurs and the value that was
predicted for a given time period. Hence,
Error = Actual – Forecast:

Where:
t = any given time period
Three commonly used measures for
summarizing historical errors:

• Mean Absolute Deviation (MAD)

• Mean Squared Error (MSE)
• Mean Absolute Percent Error (MAPE)
 Example

Compute MAD, MSE and MAPE for the following data,

showing actual and forecasted numbers of accounts
served:
Error (e) [/Error/÷Actual]
Period Actual Forecast (A – F) /Error/ e2 x 100
1 217 215
2 213 216
3 216 215
4 210 214
5 213 211
6 219 214
7 216 217
8 212 216

Total