Chapter 5

Demand forecasting
Lecturer: Pham Thi Trang, Ph.D.
 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.
2
 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
3
How to match supply and demand?

• Hold plenty of stock

Supply
• Flexible pricing
• Managing demand is
Demand
challenging

4
 Demand Forecasting
New Orders
Production
Customers’ Inventories
Price

Backlog of Orders
Inventories

Manufacturing Employment Supplier Deliveries

New Export Orders and Imports 5

Forecasting Techniques-Qualitative methods
Jury of Executive
Opinion

Delphi Method
Qualitative
Methods
Sales Force Composite

Consumer Survey
6
 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 Ft+1 = forecast for period t+1
stable over time
n = number of periods used to calculate
moving average
Ai = actual demand in period i 7
 Forecasting Techniques-Quantitative methods
Simple Moving Average Forecast Period Demand

1 1600
2 2200 Using the data
3 2000 provided,
4 1600 calculate the
5 2500 forecast for
6 3500 period 5 using
7 3300 a four-period
Ft+1 = forecast for period t+1 8 3200 simple moving
n = number of periods used to 9 3900 average.
calculate moving average 10 4700
11 4300
Ai = actual demand in period i
12 4400 8
Forecasting Techniques-Quantitative methods

Time Series Forecasting Models

Weighted Moving
Average Forecast:
the weighted
average of the
n-period Ft+1 = forecast for period t+1;
observations, using n = number of periods used in
determining the moving average;
unequal weights Ai = actual demand in period i;
wi = weight assigned to period i; Σwi =1.
9
 Period Demand Calculate the
Forecasting Techniques 1 1600 forecast for period
Quantitative methods 2 2200 5 using a four-
3 2000
period weighted
Weighted Moving Average Forecast moving average.
4 1600
The weights of
5 2500 0.4, 0.3, 0.2 and
6 3500 0.1 are assigned to
7 3300 the most recent,
Ft+1 = forecast for period t+1 8 3200
second most
n = number of periods used in recent, third most
9 3900
determining the moving average recent, and fourth
10 4700 most recent
Ai = actual demand in period i
wi = weight assigned to period i; Σwi 11 4300 periods,
=1. 12 4400 respectively.
10
 Forecasting Techniques
Quantitative methods
Exponential
Smoothing Forecast: Time Series Forecasting Models
the forecast for next
period’s demand is
the current period’s
Ft+1 = Ft + α (At − Ft)
forecast adjusted by a Ft+1 = forecast for period t + 1
fraction of the
difference between Ft = forecast for period t
the current period’s At = actual demand for period t
actual demand and α = smoothing constant (0 ≤ α ≤ 1)
forecast.
11
 Period Demand
Forecasting Techniques
1 1600
Quantitative methods 2 2200
Calculate the
forecast for
3 2000 period 3 using
Exponential Smoothing Forecast 4 1600 the exponential
5 2500 smoothing
Ft+1 = Ft + α (At − Ft) 6 3500 method.
7 3300 Assume the
Ft+1 = forecast for period t + 1 forecast for
Ft = forecast for period t 8 3200
period 2 is 1600.
At = actual demand for period t 9 3900
Use a
α = smoothing constant (0 ≤ α 10 4700 smoothing
≤ 1) 11 4300 constant (α) of
12 4400 0.3. 12
Forecasting Techniques-Quantitative methods

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
13
Forecasting Techniques-Quantitative methods

Cause & Effect Models

One or several external variables are identified that are related to demand
Simple regression – Only one explanatory variable is used & is similar to the
previous trend model. The difference is that the x variable is no longer time
but an explanatory variable.
Ŷ = b0 + b1 x
Where Ŷ = forecast or dependent variable
x = explanatory or independent variable
b0 = intercept of the line
b1 = slope of the line
14
 Forecasting Techniques-Quantitative methods
Cause & Effect Models

Multiple regression – several explanatory variables are used to make the forecast
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
15
 Forecast Accuracy
The formula for forecast error, defined as 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

16
 Forecast Accuracy (Continued)

Several measures of forecasting accuracy follow –

• Mean absolute deviation (MAD)- a MAD of 0
indicates the forecast exactly predicted demand
• Mean absolute percentage error (MAPE)- provides a
perspective of the true magnitude of the forecast error
• Mean squared error (MSE)- analogous to variance,
large forecast errors are heavily penalized

17
 Forecast Accuracy (Continued)

Mean absolute deviation (MAD)-

MAD of 0 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
18
 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
19
 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
20
 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

21
RSFE
MAD

Forecast Accuracy (Continued)

Tracking signal –
determines if forecast is within acceptable control limits. If
the tracking signal falls outside the pre-set control limits,
there is a bias problem with the forecasting method and an
evaluation of the way forecasts are generated is warranted.

RSFE
Tracking Signal =
MAD
22
 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

23
 Collaborative Planning, Forecasting,
& Replenishment Continued ( )

Real value of CPFR comes from sharing of forecasts among firms

rather than sophisticated algorithms from only one firm.

Does away with the shifting of inventories among trading partners that
suboptimizes the supply chain.

CPFR provides the supply chain with a plethora of benefits but requires
a fundamental change in the way that buyers & sellers work together.

24
 Collaborative Planning, Forecasting,
& Replenishment Continued ( )

CPFR Benefits
• Strengthens partner relationships
• Provides analysis of sales and order forecasts
• Uses point-of-sale data, seasonal activity, promotions, to improve forecast
accuracy
• Manages the demand chain and proactively eliminates problems before they
appear
• Allows collaboration on future requirements and plans
25
 Collaborative Planning, Forecasting,
& Replenishment Continued ( )

CPFR Benefits (continued)

• Allows collaboration on future requirements and plans

• Uses joint planning and promotions management
• Integrates planning, forecasting and logistics activities
• Provides efficient category management and understanding of
consumer purchasing patterns
26
 Collaborative Planning, Forecasting,
& Replenishment Continued ( )

CPFR Benefits (continued)

• Provides analysis of key performance metrics (e.g., forecast accuracy,

forecast exceptions, product lead times, inventory turnover, percentage
stockouts) to reduce supply chain inefficiencies, improve customer
service, and increase revenues and profitability.

27
 Software Solutions: Forecasting
Software
• Business Forecast Systems, Inc. (www.forecastpro.com/)
• John Galt (www.johngalt.com/)
• JustEnough (www.justenough.com/)
• SAS (www.sas.com/)

28
Practise 1/p.163

29
Practise 2/p.164

30
 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.
31