Supply Chain Management

Forecasting

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-1

 What is a Forecast?

Forecast

A prediction of
future events used
for planning,
procurement,
inventory, logistics
purposes.

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-2

 Forecasting across organization
• HR
• Finance
• Planning & Procurement
• Materials

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-3

 Demand Patterns
•A time series is the repeated observations of
demand for a service or product in their order of
occurrence
• There are five basic time series patterns
– Horizontal
– Trend
– Seasonal
– Cyclical
– Random

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-4

 Demand Patterns
Quantity

Time
Figure 8.1

(a) Horizontal: Data cluster about a horizontal line

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-5
 Demand Patterns
Quantity

Time
Figure 8.1

(b) Trend: Data consistently increase or decrease

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-6
 Demand Patterns

Year 1
Quantity

Year 2

| | | | | | | | | | | |
J F M A M J J A S O N D
Figure 8.1 Months
(c) Seasonal: Data consistently show peaks and valleys
Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-7
 Demand Patterns
Quantity

| | | | | |
1 2 3 4 5 6
Figure 8.1 Years
(d) Cyclical: Data reveal gradual increases and decreases over
extended periods
Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-8
 Demand Management Options
• Demand Management
– The process of changing demand patterns using
one or more demand options

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-9

 Demand Management Options
• Complementary Products
 Products or services that have similar resource requirements but different
demand cycles

 Promotional Pricing(More/ New car launches)

• Prescheduled Appointments
 Service providers can schedule customers for definite periods of order
fulfillment
• Reservations
 Like appointment systems, but used when the customer actually occupies or
uses facilities associated with the service
• Revenue Management
 Varying price at the right time for different customer segments to maximise
revenues yielded by existing supply capacity (Yield management)
Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-10
 Demand Management Options
• Backlogs
– An accumulation of customer orders that a
manufacturer has promised for delivery at some
future date
• Backorders
– A customer order that cannot be filled when
promised or demanded but is filled later
• Stockouts(Sale)
– An order that cannot be satisfied, resulting in a loss
of the sale.

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-11

 Key Decisions on Making Forecasts
• Deciding What to Forecast
– Level of aggregation(2 tier forecasting)
– Units of measurement(SKU’s)

• Choosing the Type of Forecasting

Technique
– Qualitative methods

– Quantitative methods

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-12

 Forecast Error
• For any forecasting method, it is important to measure
the accuracy of its forecasts.

• Forecast error is simply the difference found by

subtracting the forecast from actual demand for a given
period, or
Et = Dt – Ft
where
Et = forecast error for period t
Dt = actual demand in period t
Ft = forecast for period t
Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-13
 Qualitative/Judgmental methods
• Salesforce estimate
• Executive Opinion
• Market research
• Delphi Technique

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-14

 Quantitative Methods
• Causal Methods
– Regression
• Time Series Methods

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-15

 Causal Methods: Linear Regression
• A dependent variable is related to one or more
independent variables by a linear equation
• The independent variables are assumed to “cause”
the results observed in the past
• Simple linear regression model is a straight line
Y = a + bX
where
Y = dependent variable
X = independent variable
a = Y-intercept of the line
b = slope of the line

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-16

 Linear Regression
Y
Deviation, Regression
or error equation:
Estimate of Y = a + bX
Y from
Dependent variable

regression
equation
Actual
value
of Y

Value of X used
to estimate Y

X
Figure 8.3
Independent variable

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-17

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-18
 Linear Regression
Regression equation:
Y = a + bX
n ∑ (XY) - ∑ X ∑ Y
b =  Y = Dependent variable
∑ 2 ∑ 2
n X - ( X)  X = independent variable

 a = Y intercept of line

 b = slope of the line

∑ X2 ∑ Y - ∑ (XY) ∑(X)
a =
2 2
n ∑ X - ( ∑ X)

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-19

9/17/2024
 Quantitative Methods
• Causal Methods
– Regression
• Time Series Methods

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-20

 Demand Forecasting –
Time Series Methods

Year Demand

2016 25

2017 32

2018 24

2019 28

2020 26 What is the demand

2021 27
forecast for 2022?
2022 ?

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-21

 Demand Forecasting
Year Demand

2016 25

2017 32
27
2018 24

2019 28

2020 26
Average of all the past
2021 27 demands!
2022 ?

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-22

 Demand Forecasting
Year Demand

2016 25

2017 32
26.25
2018 24

2019 28

2020 26
I am only interested in the
2021 27 last four years demand!
2022 ?

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-23

 Demand Forecasting
Year Demand

2016 25

2017 32
28
2018 24

2019 28

2020 26
In the past two years,
2021 27 demand is showing an
2022 ? increasing trend!

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-24

 Demand Forecasting
Year Demand

2016 25

2017 32
26
2018 24

2019 28

2020 26
In alternate years demand is
2021 27 increasing/decreasing!
2022 ?

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-25

 Demand Forecasting
Year Demand

2016 25

2017 32
26
2018 24

2019 28

2020 26 Average of all past demands,

2021 27 except the demand for the
2022 ? year 2017 which seems to
be an outlier.

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-26

 Demand Forecasting
Year Demand

2016 25

2017 32
26.33
2018 24

2019 28

2020 26 I am interested in recent three demands,

2021 27 but I want to give different weightages –
2022 ? 3 for 2021, 2 for 2020 and 1 for 2019

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-27

 Demand Forecasting
Year Demand

2016 25

2017 32
21
2018 24

2019 28

2020 26 You cannot forecast 2020 demand based

2021 27 on past data; sales are going to be hit due
2022 ? to COVID-19!

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-28

 Quantitative Methods
• Time Series Methods
– Naïve Forecast
– Simple Moving Average
– Weighted Moving Average
– Exponential Smoothing
– Regression
– Holts Method (Trend adjusted Exponential
Smoothing)
– Seasonal Forecasting

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-29

 Naïve Forecast
• Ft+1 = Dt

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-30

 Simple Moving Averages
• Specifically, the forecast for period t + 1 can be
calculated at the end of period t (after the actual
demand for period t is known) as

Sum of last n demands Dt + Dt-1 + Dt-2 + … + Dt-n+1

Ft+1 = =
n n

where
Dt = actual demand in period t
n = total number of periods in the average
Ft+1 = forecast for period t + 1

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-31

 Ex. 4
a. Compute a three-week moving average forecast for the
arrival of medical clinic patients in week 4. The numbers
of arrivals for the past three weeks were as follows:
Week Patient Arrivals
1 400
2 380
3 411

b. If the actual number of patient arrivals in week 4 is

415, what is the forecast error for week 4?
c. What is the forecast for week 5?

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-32

 Ex. 4
Week Patient Arrivals
a. The moving average forecast 1 400
at the end of week 3 is: 2 380
3 411
411 + 380 + 400
F4 = = 397.0
3
b. The forecast error for week 4 is
E4 = D4 – F4 = 415 – 397 = 18

c. The forecast for week 5 requires the actual arrivals from

weeks 2 through 4, the three most recent weeks of data
415 + 411 + 380
F5 = = 402.0
3
Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-33
 Ex. 5
Estimating with Simple Moving Average using the
following customer-arrival data:
Month Customer arrival
1 800
2 740
3 810
4 790
Use a three-month moving average to forecast customer
arrivals for month 5
D4 + D3 + D2 790 + 810 + 740
F5 = = = 780
3 3

Forecast for month 5 is 780 customer arrivals

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-34
 Ex. 5
If the actual number of arrivals in month 5 is 805,
what is the forecast for month 6?

Month Customer arrival

1 800
2 740
3 810
4 790
D5 + D4 + D3 805 + 790 + 810
F6 = = = 801.667
3 3

Forecast for month 6 is 802 customer arrivals

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-35
 Ex. 5
Forecast error is simply the difference found by subtracting
the forecast from actual demand for a given period, or
Et = Dt – Ft

Given the three-month moving average forecast for

month 5, and the number of patients that actually
arrived (805), what is the forecast error?

E5 = 805 – 780 = 25

Forecast error for month 5 is 25

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-36

 Weighted Moving Averages
In the weighted moving average method, each historical
demand in the average can have its own weight,
provided that the sum of the weights equals 1.0.

The average is obtained by multiplying the weight of

each period by the actual demand for that period, and
then adding the products together

Ft+1 = W1D1 + W2D2 + … + WnDt-n+1

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-37

 Customer
Ex.6
Month
arrival
1 800
2 740
3 810
Using the customer arrival data in Ex.5, let 4 790
W1 = 0.50, W2 = 0.30, and W3 = 0.20. Use the weighted
moving average method to forecast arrivals for month 5.
F5 = W1D4 + W2D3 + W3D2
= 0.50(790) + 0.30(810) + 0.20(740) = 786
Forecast for month 5 is 786 customer arrivals.
Given the number of customers that actually arrived (805),
what is the forecast error?
E5 = 805 – 786 = 19
Forecast error for month 5 is 19.
Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-38
 Customer
Month
arrival
Ex.6 1
2
800
740
3 810
If the actual number of arrivals
4 790
in month 5 is 805, compute
the forecast for month 6:
F6 = W1D5 + W2D4 + W3D3

= 0.50(805) + 0.30(790) + 0.20(810)

= 801.5

Forecast for month 6 is 802 customer arrivals.

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-39
 Demand Forecasting – Exponential
Smoothing
A sophisticated weighted moving average that calculates the average of

No. Year Demand - Dt a time series by implicitly giving recent demands more weight than
earlier demands
1 2016 25
For finding next period’s forecast (Ft+1), requires only three items of
2 2017 32 data
3 2018 24 • The last period’s forecast (Ft)
• The demand for this period (Dt)
4 2019 28
• A smoothing parameter, alpha (α), where 0 ≤ α ≤ 1.0
5 2020 26
The equation for the forecast is
6 2021 27 Ft+1 = α(Demand this period) + (1 – α)(Forecast calculated
last period)
7 2022
= αDt + (1 – α)Ft

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-40

 Demand Forecasting – Exponential
Smoothing
No. Year Demand Forecast Ft+1 = αDt + (1 – α)Ft F2 = αD1 + (1 – α)F1

F7 = αD6 + (1 – α)F6 = (0.2) 25 + (1 – 0.2) 27

(Dt) (Ft)
F6 = αD5 + (1 – α)F5 = 26.6
1 2016 25
F5 = αD4 + (1 – α)F4 F3 = αD2 + (1 – α)F2
2 2017 32
F4 = αD3 + (1 – α)F3 = (0.2) 32 + (0.8) 26.6
3 2018 24
F3 = αD2 + (1 – α)F2 = 27.68
4 2019 28
F2 = αD1 + (1 – α)F1 Similarly we can calculate
5 2020 26 the rest of the values
Let us assume

6 2021 27 α = 0.2 and F4 = 26.944, F5 = 27.1552,

F1 = 27 (Simple Average)
7 2022 F6 = 26.92416, F7 = 26.94

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-41

 Demand Forecasting – Exponential
F = αD + (1 – α) F
Smoothing
7 6 6

= αD6 + (1 – α) [αD5 + (1 – α) F5]

= αD6 + α(1 – α) D5 + (1 – α)2 F5

= αD6 + α(1 – α) D5 + (1 – α)2 [αD4 + (1 – α) F4]

= αD6 + α(1 – α) D5 + α(1 – α)2 D4 + (1 – α)3 F4

.
.
.
.

= αD6 + α(1 – α) D5 + α(1 – α)2 D4 + α(1 – α)3 D4 + ………………... + α(1 – α)5 D1 + (1 – α)6 F1

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-42

 Demand Forecasting – Exponential
Smoothing
F7 = αD6 + α(1 – α) D5 + α(1 – α)2 D4 + α(1 – α)3 D4 + ………………... + α(1 – α)5 D1 + (1 – α)6 F1

α values
0.9
If, α = 0.2 If, α = 0.8
0.8
α(1 – α) = 0.16 α(1 – α) = 0.16 0.7

α(1 – α)2 = 0.128 α(1 – α)2 = 0.032 0.6

α(1 – α)3 = 0.1024 α(1 – α)3 = 0.0064 0.5

α(1 – α)4 = 0.08192 α(1 – α)4 = 0.00128 0.4

0.3
α(1 – α)5 = 0.065536 α(1 – α)5 = 0.000256
0.2
α(1 – α)6 = 0.0524288 α(1 – α)6 = 0.0000511999
0.1

0
1 2 3 4 5 6 7

α = 0.8 α = 0.6 α = 0.4 α = 0.2

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-43

 Demand Forecasting – Exponential
Smoothing
• The emphasis given to the most recent demand levels can be
adjusted by changing the smoothing parameter.

• Larger α values emphasize recent levels of demand and result

in forecasts more responsive to changes in the underlying
average.

• Smaller α values treat past demand more uniformly and result

in more stable forecasts.

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-44

 Patient
Week
Arrivals
Example 7 1 400
2 380
3 411
a. Reconsider the patient arrival data in Example 4.
It is now the end of week 3 so the actual arrivals
is known to be 411 patients. Using α = 0.10,
calculate the exponential smoothing forecast for
week 4. Assume, F3 = 390

b. What was the forecast error for week 4 if the

actual demand turned out to be 415?

c. What is the forecast for week 5?

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-45

 Patient
Week
Example 7 1
Arrivals
400
a. To obtain the forecast for week 4, using 2 380
exponential smoothing with and the 3 411
initial forecast of 390*, we calculate the
average at the end of week 3 as:
F4 = 0.10(411) + 0.90(390) = 392.1
Thus, the forecast for week 4 would be 392 patients.

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-46

 Example 7

b. The forecast error for week 4 is

E4 = 415 – 392 = 23
c. The new forecast for week 5 would be

F5 = 0.10(415) + 0.90(392.1) = 394.4

or 394 patients.

Copyright ©2016 Pearson Education, Inc. All rights reserved. 8-47