AY2021 BOP Topic 5 Forecasting Demand - Blanks

5.
Demand Forecasting
Copyright ©2016 Pearson Education, Limited. 1-1

Learning Outcome
• Understand the strategic importance of forecasting
• Define qualitative and quantitative forecasting
• Generate forecast using Time-Series quantitative forecast
method
• Determine forecast accuracy using the Mean Average
Deviation and Mean Square Error methods.
• Explore how trend and seasonality affects the data and how to
include these variations into forecasting

Lesson Outline
• Strategic Importance of Forecasting
• Forecasting Approaches
– Qualitative Approaches
– Quantitative Approaches
• Seasonal Variations in Data
• Monitoring and Controlling Forecasts
https://www.youtube.com/watch?v=wM7gul_jfY4

What is Forecasting
• Process of predicting a future event ??

• Underlying basis of all business decisions
• Production
• Inventory
• Personnel
• Facilities

Fill in the blanks
Forecasting Time Horizons
• Short-range forecast
– Generally up to ________ months
– Production level setting, production job and manpower scheduling, raw
material purchasing.
• Medium-range forecast
– 3 months to ______ years
– Sales planning, production capacity planning, annual budget planning,
Make or Buy decision*
• Long-range forecast
– Moe than ______ years
– Facility location planning, capital expenditure planning, new product
planning, research and development planning
Fill in the blanks
Distinguishing Differences between
Short-term forecast Medium/Long-run forecast
-Operational planning & execution -Management planning & decision

within next 3 months. making for longer lead time issues
e.g. production level, scheduling, e.g. for facility location, capital
purchasing level, etc. expenditure etc.
-Employs more Quantitative -Employs broader and more

techniques (e.g. moving average, Qualitative techniques, on top of
exponential smoothing, trend Quantitative ones.
extrapolation)
- __________ accurate - ________ accurate

Distinguishing Differences (additional remark)
• Short term forecasting techniques e.g. moving averages,

exponential smoothing, trend extrapolation
• Long term: Broader, less quantitative methods are used for e.g.
in new product launches, deciding whether a particular
product to be introduced.

Distinguishing Differences (additional remark)
• Short-range forecasts tend to be more accurate than longer-range.

Factors that influence demand tend to change daily.
Thus, as the time horizon lengthens, it is likely that one’s forecast

accuracy will diminish.
• Sales forecast must be updated regularly in order to maintain their

value and integrity. After each sales period, forecasts should be
reviewed and revised.

Types of Forecast
1. Economic forecasts
– Address business cycle – inflation rate, GDP growth rate, etc.
2. Technological forecasts
– Predict rate of technological progress Operations
– Impacts development of new products Manager’s
focus
3. Demand forecasts
– Predict sales of existing products and services
Forecast Types (for self-reading)
• Organizations use the three major types of forecasts identified
in the prvious slide for planning future operations.
• The operations manager typically focuses on demand
forecasts. May use recent point-of-sales (POS) data that will
help drive a company’s production, capacity & scheduling
systems, and serve as inputs to financial, marketing and
personnel planning.

Strategic importance of Forecasting

Strategic Importance of Forecasting
Forecasting and the Product Life Cycle
Introduction – Growth – Maturity – Decline

► Introduction and growth require longer
forecasts than maturity and decline
► As product passes through life cycle,
forecasts are useful in projecting

► Staffing levels
► Inventory levels
► Factory capacity
Forecasting and the Product Life Cycle Self-
Reading

• Supply-Chain Management – Good supplier relations, advantages
in product innovation,
cost and speed to market
• Human Resources – Hiring part-time or full

time, training, laying off workers
• Capacity – Capacity shortages can result in

undependable/unfulfilled delivery, loss of
customers, loss of market share

Additional Info. ( Self reading)
Demand forecasts drive decisions in many areas, including the
three described in this slide. For example, Apple is able to
launch new products very efficiently due to good forecast
already disseminated to the entire supply chain months in
advance.
• Also, firms typically do not hire more workers when demand
is falling, and vice-versa.

7 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 needed to make the forecast
6. Make the forecast
7. Validate and implement results

Additional Info. – Example ( Self reading)
1. What’s the purpose? Eg for headcount planning, equipment
sourcing, etc for mfg. For RWS, use park attendance forecasts
to drive staffing levels, opening hours, food supplies for RWS.
2. Number of item units for manufacturing e.g. phones, no. of
buns, etc. RWS eg attendance
3. Short term or medium term
4. Decide on forecasting models – Moving average.
5. Consolidate data from regional managers
6. Review & constantly update forecasts as it is actualised
The Realities!
• Forecasts are seldom perfect,
unpredictable outside factors may
impact the forecast
• Most techniques assume an underlying

stability in the system
• Product family and aggregated forecasts are more

accurate than individual product forecasts

MyNote

Additional Info. – Example ( Self reading)
Every company must contend with several realities about
forecasting. Forecasts need to be closely monitored to identify,
for example, unusual circumstances or major shifts from
historical trends. Also, when decisions can be based on
aggregated rather than individual product forecasts, the
aggregated forecasts should be used.
• E.g.: when the Gulf War erupted in 1990s, there was an
upsurge in demand for Motorola’s communication equipment
which was unplanned during that period.
Forecasting Approaches

tive
ntita
o n Qua
focQualitative Quantitative
us o nly
To
When existing data is limited When existing data is available

(Example: New Product, New (Example: Existing Product,
Technology) Existing Technology)
Experience is involved Mathematical techniques

Forecasting Approaches - Quantitative
Time series models

Quantitative • Naive approach
• Moving averages
• Exponential
When existing data is available smoothing
(Example: Existing Product,
Existing Technology)
• Trend projection
Mathematical techniques Associative models

• Linear regression

Forecasting Approaches – Time Series Components
Seasonal peaks Cycles – patterns in
the data that occur
every few years Trend component
Pattern that
repeats itself after
Demand for product or service
a period of days,
weeks, months Actual demand
or quarters
line Gradual upward or
downward
movement of data
Average demand over over time
four years
Random
variation Random variations –
“blips” in the data
Year Year Year Year caused by chance or
1 2 3 4 unusual situations

Time Series – Naïve Approach
• Assumes demand in next

period is the same as
demand in most recent period
e.g., If January sales were 1,000 units,
then February sales will be
1,000 units
• Sometimes cost effective and
efficient
• Can be good starting point
Additional Info. – ( Self reading)
• The naive approach is so simple that it could be considered to
be a qualitative forecasting method. Numerous forecasts in
the real world are made this way, especially for infrequent
events. For example, with no additional information
available, it would make sense to forecast attendance at this
year’s Arts Festival to be the same as last year’s.

recap
Time series models
Quantitative • Naive approach
• Moving averages
When existing data is available
(simple and weighted)
(Example: Existing Product,
Existing Technology) • Exponential smoothing
Mathematical techniques
• Exponential smoothing with
Trend projection

Lesson Outline
recap
– Qualitative Approaches (not in scope)
– Quantitative Approaches (Time series model)

Time Series – Moving Averages (MA)
• MA is a series of arithmetic means (or, averages)
• Used if little or no trend
• Used often for smoothing out random fluctuations
– Provides overall impression of data over time

Time Series – Moving Averages (MA)
MONTH ACTUAL SHED SALES 3-MONTH MOVING AVERAGE
January 10
February 12
March 13
April 16 (10 + 12 + 13)/3 = 11 2/3
May 19 (12 + 13 + 16)/3 = 13 2/3
June 23 (13 + 16 + 19)/3 = 16
July 26 (16 + 19 + 23)/3 = 19 1/3
August 30 (19 + 23 + 26)/3 = 22 2/3
September 28 (23 + 26 + 30)/3 = 26 1/3
October 18 (29 + 30 + 28)/3 = 28
November 16 (30 + 28 + 18)/3 = 25 1/3
December 14
(28 + 18 + 16)/3 = 20 2/3

Time Series – Weighted Moving Averages (WMA)
• Used when some trend might be present

• Older data usually less important
• Weights based on experience and intuition
Weighted
moving
average

6

Time Series – Simple and Weighted Moving Averages
reacts quicker
Increasing n smooths the
than simple MA forecast but makes it less
sensitive to changes
Does not forecast trends

well
Requires extensive historical

data

Time Series – Exponential Smoothing Moving Averages (EMA)
• Uses weighted averages of past observations
(i.e. actual demand).
• Weights decay exponentially as observations get older.
(i.e. the more recent observation gets higher weightage)
• Requires smoothing constant ( )
– Ranges from 0 to 1 (for usage in business usually 0.05 to 0.5)
– Subjectively chosen and its accuracy needs to be measured.


EMA method – ( Self reading)
• It is a form of weighted moving average, the special recursive formula means
that less record keeping is necessary than for a regular weighted moving
average forecast. If the weights are plotted on a graph, the curve drawn over
them will be exponential in shape (hence the name). All old data remain part
of the forecast, but the weights applied to very old data are extremely small.
• The only data needed are last period’s actual and forecasted demands. The
method adjusts the forecast each period by a certain percentage (α) of the
error in the previous period’s forecast. If the previous forecast was 40 units
too high and α = 20%, then the new forecast will be 8 units (.20 × 40) lower.

Predicted demand = 142 Ford Mustangs

Actual demand = 153
Smoothing constant a = .20
Ft = Ft – 1 + a(At – 1 - Ft – 1)


Actual demand = 153
Ft = Ft – 1 + a(At – 1 - Ft – 1)
New forecast = 142 + .2(153 – 142)

Actual demand = 153
Ft = Ft – 1 + a(At – 1 - Ft – 1)
New forecast = 142 + .2(153 – 142)

= 142 + 2.2
= 144.2 ≈ 144 cars

• Smoothing constant generally .05 ≤ a ≤ .50
• As a increases, older forecast values become less significant
Ft = Ft – 1 + a(At – 1 - Ft – 1)
Re-arranging RHS of above equation
Ft = Ft – 1 (1-a) + a(At – 1)
As a increases, Ft – 1 becomes less siginificant
while At – 1 becomes more significant
Chose high values of 

when underlying average
is likely to change
Choose low values of 

is stable
A high value of α places much more weight on the very recent periods, so the forecast can react
much quicker to trends
Choosing 
• The objective is to obtain the most accurate forecast no

matter the technique
• We generally do this by selecting the model that gives us the
lowest forecast error
Forecast error = Actual demand - Forecast value

= At – F t

Measuring Forecast Error with Mean Absolute Deviation (MAD)
• Taking the sum of absolute values of individual forecast

errors and dividing them by the number of data periods (n)
Most popular
method to
determine forecast
accuracy

Ft = Ft – 1 + a(At – 1 - Ft – 1)

MAD for  = 0.1

is 10.31
MAD for  = 0.5

is 12.33
Therefore  = 0.1 is a
better smoothing
constant

Chose high values of 

is likely to change
Choose low values of 

is stable
A high value of α places much more weight on the very recent periods, so the forecast can react
much quicker to trends
Measuring Forecast Error with Mean Squared Error (MSE)
• Because positive and negative differences may eventually offset the
difference, MSE is used to capture the ‘actual’ differences
• Mean Squared Error (MSE) refers to the average of squared
differences between the forecasted and observed values
MSE penalises
large differences
(i.e. errors)


Actual average tonnage

unloaded is 179.88
Both MAD and MSE

suggested that  = 0.1
has the lowest forecast
error
Therefore  = 0.1 is a
better smoothing
constant

Forecasting Approaches - Quantitative Exponential
smoothing
forecast severely
lags actual
Exponential Smoothing with Trend Adjustments demand
• When a trend is present, exponential smoothing must be modified

MONTH ACTUAL DEMAND FORECAST (Ft) FOR MONTHS 1 – 5
1 100 Ft = 100 (given)
2 200 Ft = F1 + a(A1 – F1) = 100 + .4(100 – 100) =

100
3 300 Ft = F2 + a(A2 – F2) = 100 + .4(200 – 100) =
140
4 400 Ft = F3 + a(A3 – F3) = 140 + .4(300 – 140) =
204
5 500 Ft = F4 + a(A4 – F4) = 204 + .4(400 – 204) =
Copyright ©2016 Pearson Education, Limited. 282 1-54
Exponential Smoothing with Trend Adjustments

Step 1: Compute Ft
Step 2: Compute Tt
Step 3: Calculate the forecast FITt = Ft + Tt

Extra slide
Actual Demand (At)
?
40
35
30
25
Quantity
20
15
10
0
1 2 3 4 5 6 7 8 9 10
month
Given a = .2 b = .4

TABLE 4.1 Forecast with a - .2 and b = .4
SMOOTHED FORECAST
FORECAST SMOOTHED INCLUDING TREND,
MONTH ACTUAL DEMAND AVERAGE, Ft TREND, Tt FITt
1 12 11 2 13.00 Given
2 17 12.80
3 20
4 19
Step 1: Average for Month 2
5 24
6 21
F2 = aA1 + (1 – a)(F1 + T1)
7 31
8 28 F2 = (.2)(12) + (1 – .2)(11 + 2)
9 36
10 — = 2.4 + (.8)(13) = 2.4 + 10.4
= 12.8 units
SMOOTHED FORECAST
1 12 11 2 13.00
2 17 12.80 1.92
3 20
4 19
5 24 Step 2: Trend for Month 2
6 21
7 31 T2 = b(F2 - F1) + (1 - b)T1
8 28
9 36 T2 = (.4)(12.8 - 11) + (1 - .4)(2)
10 —
= .72 + 1.2 = 1.92 units

SMOOTHED FORECAST
1 12 11 2 13.00
2 17 12.80 1.92 14.72
3 20
4 19
5 24 Step 3: Calculate FIT for Month 2
6 21
7 31 FIT2 = F2 + T2
8 28
9 36 FIT2 = 12.8 + 1.92
10 —
= 14.72 units

Answer


Extra slide

Exponential Smoothing with and without Trend Adjustments
Actual Demand
40
35
Forecast With Trend Adjustment
30
25
Forecast Without Trend Adjustment
Product Demand
20
15
Actual Demand (At)
10
FITt
5 Ft
0
1 2 3 4 5 6 7 8 9 10
month

Actual Demand (At)
40
35
30
25
Product Demand
20 Actual Demand (At)
Ft
15
10
0
1 2 3 4 5 6 7 8 9 10
month
Given a = .2 b = .4

Associative Model

Trend Projections – Using Least Square Method
• Fitting a trend line to historical data points to project into

the medium to long-range
• Linear trends can be found using the least squares
technique

Least squares
method
minimizes the
sum of the
squared errors
(deviations)

Coefficient of
Independent variable (x)
Baseline: 56.7 Mega watts

MyNote

Seasonal Variations in Data

• Fluctuations in data is common due to effects called
seasonality
• How can we adjust our trend data for seasonal variations in
demand?

Steps involved
1. Find average historical demand for each month
2. Compute the average demand over all months
3. Compute a seasonal index for each month
4. Estimate next year’s total demand
5. Divide this estimate of total demand by the
number of months, then multiply it by the
seasonal index for that month

Adjusting for Seasonal Variation in Data Step 1: find
historical average
for each month

Step 2: compute
Adjusting for Seasonal Variation in Data average demand
over all months
Average Monthly
Demand
= 1,128 / 12 mths
= 94

Step 3: compute
Adjusting for Seasonal Variation in Data seasonal index
Seasonal Index
= Average
monthly demand
for the past 3
years/ Average
monthly demand

Step 3: compute
Adjusting for Seasonal Variation in Data seasonal index
Seasonal Index
= Average
monthly demand
for the past 3
years/ Average
monthly demand

Seasonal Variations in Data Step 4: Estimate
next year’s total
Adjusting for Seasonal Variation in Data annual demand
Estimated Total
Demand = 1,200
units
Step 5: divide
annual demand by
12 & multiply by
the month’s
seasonal index

Example of Seasonal Index
140 –
Year 4 Forecast
130 – Year 3 Demand
Demand
100 –
90 –
80 –
70 –
| | | | | | | | | | | |
J F M A M J J A S O N D
Time
MyNote

Monitoring and Controlling Forecast

Using Tracking Signals
• Measures how well the forecast is predicting actual values
• Ratio of cumulative forecast errors to mean absolute
deviation (MAD)
– Good tracking signal has low values
– If forecasts are continually high or low, the forecast has a bias
error


Forecast Error


Management is using control limits of 4 MADs
Variation is
deemed
acceptable if
company
controls within
4 MADs

Summary
1. Forecasting is a critical part of a Operations manager’s function as it
has an overall impact on the company’s operations.
2. There are a variety of quantitative and qualitative forecasting
techniques.
3. Qualitative approaches uses judgement, experience, intuition and
other factors to quantify.
4. Quantitative forecasting uses historical data to project future
demands.
5. Forecasting can be a challenging part of management

MyNote

Lesson Outline
recap
– Qualitative Approaches (not in scope)
– Quantitative Approaches (Time series model)

Reference
1. Chapters 4, Heizer, Render & Munson, “Principles of

Operations Management, Sustainability & Supply Chain
Management”, 10E, Pearson

AY2021 BOP Topic 5 Forecasting Demand - Blanks

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

AY2021 BOP Topic 5 Forecasting Demand - Blanks

Uploaded by

Copyright:

Available Formats

5.

Copyright ©2016 Pearson Education, Limited. 1-1

Copyright ©2016 Pearson Education, Limited. 1-2

Copyright ©2016 Pearson Education, Limited. 1-3

• Process of predicting a future event ??

Copyright ©2016 Pearson Education, Limited. 1-4

-Operational planning & execution -Management planning & decision

-Employs more Quantitative -Employs broader and more

Copyright ©2016 Pearson Education, Limited. 1-6

• Short term forecasting techniques e.g. moving averages,

Copyright ©2016 Pearson Education, Limited. 1-7

• Short-range forecasts tend to be more accurate than longer-range.

Thus, as the time horizon lengthens, it is likely that one’s forecast

• Sales forecast must be updated regularly in order to maintain their

Copyright ©2016 Pearson Education, Limited. 1-8

Copyright ©2016 Pearson Education, Limited. 1-10

Copyright ©2016 Pearson Education, Limited. 1-11

Introduction – Growth – Maturity – Decline

forecasts are useful in projecting

Copyright ©2016 Pearson Education, Limited. 1-13

• Human Resources – Hiring part-time or full

• Capacity – Capacity shortages can result in

Copyright ©2016 Pearson Education, Limited. 1-14

Copyright ©2016 Pearson Education, Limited. 1-15

Copyright ©2016 Pearson Education, Limited. 1-16

• Most techniques assume an underlying

• Product family and aggregated forecasts are more

Copyright ©2016 Pearson Education, Limited. 1-18

Copyright ©2016 Pearson Education, Limited. 1-19

Copyright ©2016 Pearson Education, Limited. 1-21

When existing data is limited When existing data is available

Experience is involved Mathematical techniques

Copyright ©2016 Pearson Education, Limited. 1-22

Time series models

Mathematical techniques Associative models

Copyright ©2016 Pearson Education, Limited. 1-23

Copyright ©2016 Pearson Education, Limited. 1-24

• Assumes demand in next

Copyright ©2016 Pearson Education, Limited. 1-26

Copyright ©2016 Pearson Education, Limited. 1-27

Copyright ©2016 Pearson Education, Limited. 1-28

Copyright ©2016 Pearson Education, Limited. 1-29

Copyright ©2016 Pearson Education, Limited. 1-30

• Used when some trend might be present

Copyright ©2016 Pearson Education, Limited. 1-31

Copyright ©2016 Pearson Education, Limited. 1-34

Does not forecast trends

Requires extensive historical

Copyright ©2016 Pearson Education, Limited. 1-35

Copyright ©2016 Pearson Education, Limited. 1-37

Copyright ©2016 Pearson Education, Limited. 1-38

Copyright ©2016 Pearson Education, Limited. 1-39

Predicted demand = 142 Ford Mustangs

Copyright ©2016 Pearson Education, Limited. 1-40

Predicted demand = 142 Ford Mustangs

New forecast = 142 + .2(153 – 142)

Copyright ©2016 Pearson Education, Limited. 1-41

New forecast = 142 + .2(153 – 142)

Copyright ©2016 Pearson Education, Limited. 1-42

Chose high values of 

Choose low values of 

• The objective is to obtain the most accurate forecast no

Forecast error = Actual demand - Forecast value