Chapter 3

Demand
forecasting

Le Thi Thanh Ngan, MSc.

 Le Thi Thanh Ngan, MSc.

Chapter objectives
After completing this chapter, you should be able to:
 Explain the role of demand forecasting in a supply chain.
 Identify the components of a forecast.
 Compare and contrast qualitative and quantitative
forecasting techniques.

5/3/2024 2
 Le Thi Thanh Ngan, MSc.

Lecture outline
In this chapter, you will learn about:
1. Forecasting Techniques
2. Qualitative Methods
3. Quantitative Methods: Time Series Forecasting Models
4. Software Solutions
5. Summary

5/3/2024 3
 Le Thi Thanh Ngan, MSc.

Opening questions

 Explain the impact of a mismatch in supply and demand. What

strategies can companies adopt to influence demand?

5/3/2024 4
 Le Thi Thanh Ngan, MSc.

How to match supply and demand?

• Hold plenty of stock

Supply
• Flexible pricing
• Managing demand is
Demand
challenging

5/3/2024 5
The Importance of Demand Forecasting
 A forecast is an estimate of future demand & provides the basis
for planning decisions
 The goal is to minimize deviation between actual demand and
forecast
 The factors that influence demand must be considered when
forecasting
 Buyers and sellers should share all relevant information to
generate a single consensus forecast
 Good forecasting provides reduced inventories, costs, &
stockouts, & improved production plans & customer service

6
Forecasting Techniques

 Qualitative forecasting - based on opinion & intuition

 Quantitative forecasting - uses mathematical models &

historical data to make forecasts

7
Forecasting Techniques (Continued)

Qualitative Forecasting Methods

 Used when data are limited, unavailable, or not currently
relevant
 Forecast depends on skill & experience of forecast &
available information

8
 Le Thi Thanh Ngan, MSc.

Forecasting Techniques
Qualitative methods
Jury of Executive Opinion

Delphi Method
Qualitative Methods
Sales Force Composite

Consumer Survey

5/3/2024 9
Qualitative methods (Continued)

Jury of executive opinion

 Group of senior management executives collectively develop the
forecast
 Good for long-range planning and new product introductions
Delphi method
 Internal and external experts are surveyed during several rounds
 summary of responses is sent out to all the experts & they can modify
their responses in next round
 Good for high-risk technology forecasting; large, expensive projects; or
major new product introductions

10
Qualitative methods (Continued)

Sales force composite

 Based on sales force’s knowledge of the market & estimates of
customer needs
 Tendency for sales force to under-forecast
Consumer survey
 Questionnaire uses inputs from customers on future purchasing
needs, new product ideas, & opinions about existing or new products

11
 Le Thi Thanh Ngan, MSc.

Forecasting Techniques
Quantitative methods
Time Series Forecasting Models
Simple Moving Average
Forecast:
uses historical data to
generate a forecast and
works well when the
demand is fairly stable
over time Ft+1 = forecast for period t+1
n = number of periods used to calculate
moving average
Ai = actual demand in period i
5/3/2024 12
 Le Thi Thanh Ngan, MSc.

Forecasting Techniques Period Demand

Quantitative methods 1 1600

2 2200
Simple Moving Average Forecast
3 2000
4 1600 Using the data
provided,
5 2500 calculate the
6 3500 forecast for
7 3300 period 5 using a
8 3200 four-period
Ft+1 = forecast for period t+1 simple moving
9 3900
n = number of periods used to average.
calculate moving average 10 4700
Ai = actual demand in period i 11 4300
12 4400
5/3/2024 13
 Le Thi Thanh Ngan, MSc.

Forecasting Techniques
Quantitative methods
Time Series Forecasting Models
Weighted Moving
Average Forecast:
the weighted average
of the n-period
observations, using
unequal weights Ft+1 = forecast for period t+1;
n = number of periods used in determining
the moving average;
Ai = actual demand in period i;
wi = weight assigned to period i; Σwi =1.
5/3/2024 14
 Le Thi Thanh Ngan, MSc.

Forecasting Techniques Period Demand

Calculate the
Quantitative methods 1 1600
forecast for period
2 2200 5 using a four-
Weighted Moving Average Forecast
3 2000 period weighted
4 1600 moving average.
The weights of 0.4,
5 2500
0.3, 0.2 and 0.1
6 3500 are assigned to the
7 3300 most recent,
Ft+1 = forecast for period t+1 second most
8 3200
n = number of periods used in determining recent, third most
the moving average 9 3900
recent, and fourth
Ai = actual demand in period i 10 4700 most recent
wi = weight assigned to period i; Σwi =1. 11 4300 periods,
12 4400 respectively.
5/3/2024 15
 Le Thi Thanh Ngan, MSc.

Forecasting Techniques
Quantitative methods
Time Series Forecasting Models
Exponential Smoothing
Forecast: Ft+1 = Ft + α (At − Ft)
the forecast for next
period’s demand is the
current period’s Ft+1 = forecast for period t + 1
forecast adjusted by a
fraction of the Ft = forecast for period t
difference between the At = actual demand for period t
current period’s actual
demand and forecast. α = smoothing constant (0 ≤ α ≤ 1)

5/3/2024 16
 Le Thi Thanh Ngan, MSc.

Forecasting Techniques Period Demand

Quantitative methods 1 1600
2 2200
Exponential Smoothing Forecast
3 2000
4 1600 Calculate the
Ft+1 = Ft + α (At − Ft) 5 2500
forecast for period
3 using the
6 3500 exponential
Ft+1 = forecast for period t + 1 7 3300 smoothing method.
Ft = forecast for period t Assume the
8 3200
At = actual demand for period t forecast for period
α = smoothing constant (0 ≤ α ≤ 1) 9 3900
2 is 1600. Use a
10 4700 smoothing
11 4300 constant (α) of 0.3.
12 4400
5/3/2024 17
 Le Thi Thanh Ngan, MSc.

Software Solutions: Forecasting Software

 Business Forecast Systems, Inc. (www.forecastpro.com/)
 John Galt (www.johngalt.com/)
 JustEnough (www.justenough.com/)
 SAS (www.sas.com/)

5/3/2024 18
 Le Thi Thanh Ngan, MSc.

Practise 1/p.163

5/3/2024 19
 Le Thi Thanh Ngan, MSc.

Practise 2/p.164

5/3/2024 20
 Le Thi Thanh Ngan, MSc.

The Toro Cutlery Company has collected monthly sales information below:

Month Sales Month Sales Month Sales

January 20,000 May 92,000 September 30,000
February 16,000 June 30,000 October 90,000
March 42,000 July 90,000 November 80,000
April 100,000 August 50,000 December 90,000

The company is examining two forecasting methods, moving average and exponential
smoothing for forecasting sales.
a. What will the forecast be for January the following year using a 3-, 4-, and 5-
month moving average?
b. What will the forecast be for January the following year using exponential
smoothing with α values of 0.5? Assume the forecast for February this year is
20,000.

5/3/2024 21
 Forecasting Techniques (Continued)

Time Series Forecasting Models

Linear Trend Forecast – trend can be estimated using simple linear regression
to fit a line to a time series

Ŷ = b0 + b1x

Where Ŷ = forecast or dependent variable

x = time variable
b0 = intercept of the line
b1 = slope of the line

22
Forecasting Techniques (Continued)

Linear Regression

(Fig. 5.4)

23
 Forecasting Techniques (Continued)

Cause & Effect Models

One or several external variables are identified & related to demand
Simple regression – Only one explanatory variable is used & is similar to the previous
linear trend model.
 x variable is no longer time but an explanatory variable

Ŷ = b0 + b1x

Where Ŷ = forecast or dependent variable

x = explanatory or independent variable
b0 = intercept of the line
b1 = slope of the line

24
 Forecasting Techniques (Continued)

Cause & Effect Models

Multiple regression – several explanatory variables are
used to predict the dependent variable
Ŷ = b0 + b1x1 + b2x2 + . . . Bkxk
Where
Ŷ = forecast or dependent variable
xk = kth explanatory or independent
variable
b0 = intercept of the line
bk = regression coefficient of the
independent variable xk

25
Forecast Accuracy

Forecast error - the difference between actual quantity &

the forecast

Forecast error, et = At - Ft

Where et = forecast error for Period t

At = actual demand for Period t
Ft = forecast for Period t

26
Forecast Accuracy (Continued)

Mean absolute deviation (MAD)- MAD indicates the

forecast exactly predicted demand

Where et = forecast error for period t

At = actual demand for period t
n = number of periods of evaluation

27
Forecast Accuracy (Continued)

Mean absolute percentage error (MAPE) – provides a

perspective of the true magnitude of the forecast error

Where et = forecast error for period t

At = actual demand for period t
n = number of periods of evaluation

28
Forecast Accuracy (Continued)

Mean squared error (MSE) – analogous to variance,

large forecast errors are heavily penalized

Where et = forecast error for period t

n = number of periods of evaluation

29
Forecast Accuracy (Continued)

Running Sum of Forecast Errors (RSFE) –

indicates bias in the forecasts or the tendency of a
forecast to be consistently higher or lower than
actual demand.
n
Running Sum of Forecast Errors, RSFE = ∑e
t =1
t

Where et = forecast error for period t

30
Forecast Accuracy (Continued)

 Tracking signal determines if forecast is within acceptable control

limits
 If tracking signal falls outside pre-set control limits
 there is bias problem with the forecasting method
 evaluation of forecast generation is warranted

RSFE
Tracking Signal =
MAD

31
Forecast Accuracy (Continued)

 Biased forecast will lead to excessive inventories or stockouts

 Key to accurate forecasts is collaboration with different partners inside

outside the firm working together to eliminate forecasting error

 Collaboration can lead to significant improvements in forecasting accuracy

32
Collaborative Planning, Forecasting, &
Replenishment (CPFR)

What is CPFR?
It is a business practice that combines the intelligence of
multiple trading partners in the planning & fulfillment of
customer demands.
It links sales & marketing best practices, such as category
management, to supply chain planning processes to increase
availability while reducing inventory, transportation & logistics
costs.
33
 Collaborative Planning, Forecasting, &
Replenishment Continued ( )

 Real value of CPFR comes from sharing of forecasts among firms

 Eliminates shifting of inventories among trading partners that suboptimizes
the supply chain
 Provides the supply chain with a plethora of benefits but requires a
fundamental change in the way that buyers & sellers work together.

34
Collaborative Planning, Forecasting, & Replenishment
(Continued)

CPFR Benefits
 Strengthens partner relationships
 Provides analysis of sales & order forecasts
 Uses point-of-sale data, seasonal activity, promotions,
to improve forecast accuracy
 Manages demand chain & eliminates problems before
they appear
 Allows collaboration on future requirements & plans

35
Collaborative Planning, Forecasting, &
Replenishment Continued
( )

CPFR Benefits (continued)

 Uses joint planning & promotions management

 Integrates planning, forecasting and logistics activities
 Provides efficient category management &
understanding of consumer purchasing patterns

36
Collaborative Planning, Forecasting, &
Replenishment Continued ( )

CPFR Benefits (continued)

 Provides analysis of key performance metrics to:

 reduce supply chain inefficiencies
 improve customer service
 increase revenues and profitability

37
Collaborative Planning, Forecasting, & Replenishment
(Continued)

Top three challenges for CPFR implementation

1. Difficulty of making internal changes

2. Cost
3. Trust

38
 Le Thi Thanh Ngan, MSc.

Summary
Forecasting is an integral part of demand management since it provides an
estimate of future demand and the basis for planning and making sound
business decisions. A mismatch in supply and demand could result in
excessive inventories and stockouts and loss of profits and goodwill. Proper
demand forecasting enables better planning and utilization of resources for
businesses to be competitive. Both qualitative and quantitative methods are
available to help companies forecast demand better. The qualitative methods
are based on judgment and intuition, whereas the quantitative methods use
mathematical techniques and historical data to predict future demand. The
quantitative forecasting methods can be divided into time series and cause-
and-effect models. Since forecasts are seldom completely accurate,
management must monitor forecast errors and make the necessary
improvements to the forecasting process.
5/3/2024 39