Supply Chain Management
eMDP on Supply Chain Strategy & Management
Indian Institute of Management Kozhikode
2019-20
FORECASTING TECHNIQUES
2
ROLE OF FORECASTING
Forecasting is a vital function and affects every significant management decision.
o Finance and accounting use forecasts as the basis for budgeting and cost control.
o Marketing relies on forecasts to make key decisions such as new product planning and
personnel compensation.
o Production uses forecasts to select suppliers; determine capacity requirements; and drive
decisions about purchasing, staffing, and inventory.
3
COMPONENTS OF DEMAND
Average Demand (for
a period of time)
Trend
Demand
Seasonal Element
Cyclical Element
Random Variation
Autocorrelation
4
NAÏVE FORECAST
Uses a single previous value of a time series as the basis for
a forecast
The forecast for a time period is equal to the previous
time period’s value
Can be used with
a stable time series
seasonal variations
trend
5
Forecast is the average of a fixed number
SIMPLE MOVING AVERAGE of past periods.
Useful when demand is not growing or
declining rapidly and no seasonality is
present.
Removes some of the random fluctuation
from the data.
Selecting the period length is important.
Longer periods provide more
smoothing.
Shorter periods react to trends more
quickly.
6
SIMPLE MOVING AVERAGE - EXAMPLE
7
SIMPLE MOVING AVERAGE - EXAMPLE
8
WEIGHTED MOVING AVERAGE Selecting Weights
Experience and/or trial-and-error are the
The simple moving average formula implies simplest approaches.
equal weighting for all periods.
The recent past is often the best indicator
A weighted moving average allows unequal of the future, so weights are generally
higher for more recent data.
weighting of prior time periods.
The sum of the weights must be equal to one. If the data are seasonal, weights should
reflect this appropriately.
Often, more recent periods are given higher
weights than periods farther in the past.
𝐹𝑡 = 𝑤1𝐴𝑡 − 1 + 𝑤2𝐴𝑡 − 2 + …+
𝑤𝑛𝐴𝑡 − 𝑛
9
EXPONENTIAL SMOOTHING Week Demand Forecast
1 820 820
2 775 820
3 680 811
4 655 785
5 750 759
6 802 757
7 798 766
8 689 772
9 775 756
10 XXXX
10
EXPONENTIAL SMOOTHING - EXAMPLE
11
EXPONENTIAL SMOOTHING INCLUDING TREND - HOLT’S METHOD
The presence of a trend in the data causes the
exponential smoothing forecast to always lag behind Calculate the new forecast, assuming
the actual data
the following:
This can be corrected by adding a trend adjustment The previous forecast including
The trend smoothing constant is delta (δ) trend (FITt-1) is 110 and the
previous estimate of the trend
(Tt-1) is 10
α = 0.2 and δ = 0.3
Actual demand for period t-1 is
Ft = FITt-1 + α(A t-1 - FITt-1 )
115
12
HOLT’S METHOD - EXAMPLE
Ft = FITt-1 + α(At-1 – FITt-1) = 110 + 0.2(115-110) = 111.0
Tt = Tt-1 + δ(Ft – FITt-1) = 10 + 0.3(111-110) = 10.3
FITt = Ft + Tt = 111.0 + 10.3 = 121.3
Choosing α and δ
Relatively small values for α and δ are common - Usually in the range 0.1 to 0.3
α depends upon how much random variation is present
δ depends upon how steady the trend is
Measurement of forecast error can be used to select values of α and δ to minimize overall
forecast error
13
The least squares
method determines the
LINEAR REGRESSION ANALYSIS parameters a and b such
that the sum of the
Regression is used to identify the functional relationship between two or squared errors is
more correlated variables, usually from observed data. minimized – “least
One variable (the dependent variable) is predicted for given values of the
squares”
other variable (the independent variable).
Linear regression is a special case that assumes the relationship between
the variables can be explained with a straight line.
Quarter Sales Quarter Sales
1 600 7 2,600
2 1,550 8 2,900
3 1,500 9 3,800
4 1,500 10 4,500
5 2,400 11 4,000
6 3,100 12 4,900
14
DETERMINING SEASONAL FACTORS
The seasonal factor (or index) is the ratio of the amount sold during each
season divided by the average for all seasons.
Average Sales
Season Past Sales Seasonal Factor
for Each Season
1000 200
Spring 200
4
= 250 250
= 0.8
1000 350
Summer 350
4
= 250 250
= 1.4
1000 300
Fall 300
4
= 250 250
= 1.2
1000 150
Winter 150
4
= 250 250
= 0.6
Total 1000
Question: Seasonal projections if next year’s forecast is 1100?
15
SEASONAL ALLOCATIONS
Expected Average Next
Demand Sales for Seasonal Year’s
for Each Season Factor Seasonal
Next Year (1,100/4) Forecast
Spring 275 X 0.8 = 220
Summer 275 X 1.4 = 385
Fall 275 X 1.2 = 330
Winter 275 X 0.6 = 165
1100
16
DECOMPOSITION USING REGRESSION
Decompose the time series into its components
Find seasonal component
De-seasonalize the demand
Find trend component
Forecast future values of each component
Project trend component into the future
Multiply trend component by seasonal component
17
DECOMPOSITION USING REGRESSION - EXAMPLE
18
DECOMPOSITION USING REGRESSION – EXAMPLE (2)
Y from Seasonal Forecast
Period Quarter
Regression Factor (Y x Seasonal Factor)
13 I 5,003.5 0.82 4,102.87
14 II 5,345.7 1.10 5,880.27
15 III 5,687.9 0.97 5,517.26
16 IV 6,030.1 1.12 6,753.71
19
FORECAST ERRORS
Forecast error is the difference between the forecast value and what actually
occurred.
All forecasts contain some level of error.
Sources of error
Bias – when a consistent mistake is made
Random – errors that are not explained by the model being used
Measures of error
Mean absolute deviation (MAD)
Mean absolute percent error (MAPE)
Tracking signal
20
FORECAST ERROR MEASUREMENT • MAPE scales the forecast error to the
magnitude of demand.
Ideally, MAD will be zero (no forecasting
error).
Larger values of MAD indicate a less
accurate model.
Tracking signal indicates whether
forecast errors are accumulating over
time (either positive or negative
errors).
21
COMPUTING FORECAST ERROR
22
COMPUTING FORECAST ERROR - EXAMPLE
23
COMPUTING FORECAST ERROR - EXAMPLE
24
METHOD SELECTION GUIDE
Amount of Historical Forecast
Forecasting Method Data Pattern
Data Horizon
Stationary (i.e., no
Simple moving 6 to 12 months; weekly
trend or Short
average data are often used
seasonality)
Weighted moving
5 to 10 observations
average and simple Stationary Short
needed to start
exponential smoothing
5 to 10 observations
Exponential smoothing Stationary and
needed to start Short
with trend trend
Stationary, trend, Short to
Linear regression 10 to 20 observations
and seasonality medium
25
FURTHER READING
Hyndman, R. J., & Athanasopoulos, G. (2014). Forecasting: Principles and Practice. Otexts
Hanke John, E., Wichern Dean, W., & Reitsch Arthur, G. (2005). Business Forecasting. PHI
Publication
Makridakis, S., Wheelwright, S. C., & Hyndman, R. J. (2008). Forecasting methods and
applications. John Wiley & Sons
26
EMDP ON SUPPLY CHAIN STRATEGY & MANAGEMENT
