Chapter 4

Forecasting Demand
Forecasting is the art and science of predicting
future events.

Forecasting Time Horizons

1. Short-range forecast:
 A time of up to 1 year but generally less
than 3 months.
 Used for planning purchasing, job
scheduling, workforce levels, job assignments
and production levels.
2. Medium-range forecast (intermediate)
 3 months to 3 years.
 Used is sales planning, production
planning and budgeting, cash budgeting and
analysis of various operating plans.

3. Long-range forecast
 Generally 3 years or more.
 Used in planning for new products,
capital expenditures, facility location or
expansion, and research and development.
Medium and long-range forecasts are distinguished from short-
range forecasts by three features:

1. Intermediate and long-run forecast deal with more

comprehensive issues and support management decisions
regarding planning and products, plants and processes.
2. Short-term forecasting usually employs different
methodologies than longer-term forecasting. Mathematical
techniques, such as moving averages, exponential
smoothing and trend extrapolation are common to short-
run projection.
3. Short-range forecasts tend to be more accurate than longer
range forecasts.
 Types of Forecasts
 Economic forecasts
 Address business cycle – inflation rate, money
supply, housing starts, etc.
 Technological forecasts
 Predict rate of technological progress
 Impacts development of new products
 Demand forecasts
 Predict sales of existing products and services
 Strategic Importance of
Forecasting
 Human Resources – Hiring, training, laying
off workers
 Capacity – Capacity shortages can result in
undependable delivery, loss of customers,
loss of market share
 Supply Chain Management – Good supplier
relations and price advantages
 Seven Steps in Forecasting
1. Determine the use of the forecast
2. Select the items to be forecasted
3. Determine the time horizon of the
forecast
4. Select the forecasting model(s)
5. Gather the data
6. Make the forecast
7. Validate and implement results
 The Realities!

 Forecasts are seldom perfect

 Most techniques assume an
underlying stability in the system
 Product family and aggregated
forecasts are more accurate than
individual product forecasts
 Forecasting Approaches
Qualitative Methods
 Used when situation is vague and
little data exist
 New products
 New technology
 Involves intuition, experience
 e.g., forecasting sales on Internet
 Forecasting Approaches
Quantitative Methods
 Used when situation is ‘stable’ and
historical data exist
 Existing products
 Current technology
 Involves mathematical techniques
 e.g., forecasting sales of color televisions
 Jury of Executive Opinion
 Involves small group of high-level experts and
managers
 Group estimates demand by working together
 Combines managerial experience with
statistical models
 Relatively quick
 ‘Group-think’
disadvantage
 Delphi Method
 Iterative group Decision Makers
process, continues (Evaluate responses and
make decisions)
until consensus is
reached
 3 types of Staff
(Administering survey)
participants
 Decision makers
 Staff
Respondents
 Respondents (People who can make
valuable judgments)
 Sales Force Composite

 Each salesperson projects his or her sales

 Combined at district and national levels
 Sales reps know customers’ wants
 Tends to be overly optimistic
 Consumer Market Survey
 Ask customers about purchasing plans
 What consumers say, and what they
actually do are often different
 Sometimes difficult to answer
Overview of Quantitative Methods
1. Naïve approach
2. Moving averages
time-series method
3. Exponential smoothing
4. Trend projection
5. Linear regression  associative method

Time series. A forecasting technique that uses a

series of past data points to make a forecast.
Linear regression analysis. A straight-line
mathematical model to describe the functional
relationships between independent and dependent
variables.
Decomposition of a Time Series

A time series has 4 components:

1. Trend
2. Seasonality
3. Cyclical
4. Random
 Trend Component

 Persistent, overall upward or

downward pattern
 Changes due to population,
technology, age, culture, etc.
 Typically several years duration
 Seasonal Component
 Regular pattern of up and down
fluctuations
 Due to weather, customs, etc.
 Occurs within a single year
Number of
Period Length Seasons
Week Day 7
Month Week 4-4.5
Month Day 28-31
Year Quarter 4
Year Month 12
Year Week 52
 Cyclical Component
 Repeating up and down movements
 Affected by business cycle, political, and
economic factors
 Multiple years duration
 Often causal or
associative
relationships

0 5 10 15 20
 Random Component
 Erratic, unsystematic, ‘residual’ fluctuations
 Due to random variation or unforeseen
events
 Short duration
and nonrepeating

M T W T F
 Components of Demand
Trend
component

Seasonal peaks
Demand for product or service

Actual demand
line

Average demand
over 4 years

Random variation
| | | |
1 2 3 4
Time (years)
Figure 4.1
Naïve Approach

A forecasting techniques which assumes

that demand in the next period is equal to
demand in the most recent period.

forecast
period actual naïve approach
1 99
2 95 99
3 80 95
4 88 80
5 88
 forecast
period actual naïve approach
1 99
2 95 99
3 80 95
Now 4 88 80
5 88
6 ?
7 ?

Forecast of period 6 (now) = 88.

Forecast of period 7 (now) = 88.
Naïve Approach
Seasonal Variation No Trend
Actual Forecast
2015-Q1 116
2015-Q2 92
2015-Q3 84
2015-Q4 125
2016-Q1 116 2017-Q1
2016-Q2 92 2017-Q2
2016-Q3 84 2017-Q3
2016-Q4 125 2017-Q4
Moving Average

A forecasting method that uses an average

of the n most recent periods of data to forecast
the next period.

A 4-month moving average is found by

simply summing the demand during the past 4
month and dividing by 4. With each passing
month, the most recent month’s data are added
to the sum and the earliest month is dropped.
This practice tends to smooth out short-term
irregularities in the data series.
When a detectable trend or pattern is present,
weights can be used to place more emphasis
on recent values.
 Potential Problems With
Moving Average
 Increasing n smooths the forecast but
makes it less sensitive to changes
 Do not forecast trends well
 Require extensive historical data
 Moving Average And
Weighted Moving Average
Weighted
30 – moving
average
25 –
Sales demand

20 – Actual
sales
15 –
Moving
10 – average

5 –
| | | | | | | | | | | |
J F M A M J J A S O N D
Figure 4.2
 Exponential Smoothing

Is a sophisticated weighted-moving forecasting

method in which data points are weighted by an
exponential function.

New forecast = Last period’s forecast

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

α is a weight, or smoothing constant, chosen by the

forecaster, that has a value between 0 and 1.
Where Ft = new forecast
Ft-1 = previous period’s forecast
α = smoothing constant (0  α  1)
At-1 = previous period’s actual demand

The concept of exponential smoothing is that the

latest estimate of demand is equal to the estimate
adjusted by a fraction of the difference between the
last period’s actual demand and the old estimate.
Example.
In January, a car dealer predicted February
demand for 142 Ford Mustang. Actual February
demand was 153 autos. Using a smoothing
constant chosen by management of α = 0.2, the
dealer wants to forecast March demand using
the exponential smoothing model.

March forecast = 142 + 0.2(153 – 142)

= 144.2
The smoothing constant, α, is generally in the range
from 0.05 to 0.5 for business applications. It can be
changed to give more weight to recent data (when α is
high) or more weight to past data (when α is low).

When α reaches the extreme value of 1, then

Ft = At-1. The forecast becomes identical to the naïve
model.

High values of α are chosen when the underlying

average are likely to change.
Low values of α are used when the underlying average
is fairly stable.
 2nd Most 3rd Most 4th Most 5th Most
Most
Smoothing Recent Recent Recent Recent
Recent
Constant Period Period Period Period
Period (a)
a(1-a) a(1-a)2 a(1-a)3 a(1-a)4

a = 0.1 0.1 0.09 0.081 0.073 0.066

a = 0.5 0.5 0.25 0.125 0.063 0.031

Measuring Forecast Error

Forecast error (or deviation) is defined as:

Forecast error = Actual demand – Forecast value

= At - Ft
Mean Absolute Deviation (MAD) is the average of
the absolute differences between the forecasted
and observed values.
Example:

During the past 8 quarters, the Port of Baltimore has

unloaded large quantities of grain from ships. The port’s
operations manager wants to test the use of exponential
smoothing to see how all the technique works in
predicting tonnage unloaded. He guesses that the forecast
of grain unloaded in the first quarter was 175 tons.
Two values of α are to be examined: α = 0.10 and α = 0.50.
Mean Squared Error (MSE) is the average of the
squared differences between the forecasted and
observed values.
Example:
The operation manager for the Port of
Baltimore now wants to compute the MSE for α
= 0.10.
 ∑ (forecast errors)2
MSE =
n
For a = .10
= 1,526.54/8 = 190.82
For a = .50
= 1,561.91/8 = 195.24
Mean Absolute Percent Error (MAPE) is the average of
the absolute differences between the forecast and
actual values, expressed as a percent of actual values.
Example:
The Port of Baltimore wants to now calculate the
MAPE when α = 0.1.
 n
∑100|deviationi|/actuali
MAPE = i=1
n
For a = .10
= 44.75/8 = 5.59%
For a = .50
= 54.05/8 = 6.76%
Exponential Smoothing with Trend Adjustment

Forecast including trend (FITt)

= Exponentially smoothed forecast (Ft)
+ Exponentially smoothed trend (Tt)
Where
Ft = exponentially smoothed forecast of data series
in period t
Tt = exponentially smoothed trend in period t
At = actual demand in period t
α = smoothing constant for the average (0  α  1)
β = smoothing constant for the trend (0  β  1)
Exponential Smoothing with Trend
Adjustment Example
35 –

30 – Actual demand (At)

25 –
Product demand

20 –

15 –

10 – Forecast including trend (FITt)

with a = .2 and  = .4
5 –

0 – | | | | | | | | |
1 2 3 4 5 6 7 8 9
Figure 4.3
Time (month)
Trend Projections

A time-series forecasting method that fits a

trend line to a series of historical data points and
then projects the line into the future for
forecast.

Least square method

 Predicted value
y-axis intercept
b slope of the regression line

x known values of the independent variable

y known values of the dependent variable
average of the x-values
average of the y-values
n number of data points or observations
References:

Operations Management by Jay Heizer and Barry

Render, 10th edition, 2011

Operations Management by Russell and Taylor

Operations Management for Competitive

Advantage by Chase, Aquilano and Jacobs