Forecasti
ng
LEAH BAUTISTA-BUADA
Forecasting
Forecasts are a basic input in the decision
processes of operations management because
they provide information on future demand.
Businesses make plans for future operations based
on anticipated future demand. Anticipated demand
is derived from two possible sources: actual
customer orders and forecasts.
Forecasting
Two (2) aspects of forecasts are important:
o expected level of demand – can be a
function of some structural variation, such as a
trend or seasonal variation
o degree of accuracy that can be assigned to a
forecast –a function of the ability of forecasters
to correctly model demand, random variation,
and sometimes unforeseen events
Forecasting
Forecasts are made with reference to a specific time horizon.
The time horizon may be fairly short or somewhat longer.
Short-term forecasts (e.g., an hour, day, week, or month)
pertain to ongoing operations.
Long-term forecasts (e.g., the next six months, the next
year, the next five years, or the life of a product or
service) pertain to new products or services, new
equipment, new facilities, or something else that will
require a somewhat long lead time to develop, construct,
or otherwise implement.
Uses of Forecasts
1. To help managers plan the system
2. To help managers plan the use of
the system
Features Common to All
Forecasts
1. Forecasting techniques generally assume that the
same underlying causal system that existed in the
past will continue to exist in the future.
2. Forecasts are not perfect.
3. Forecasts for groups of items tend to be more
accurate than forecasts for individual items.
4. Forecast accuracy decreases as the time period
covered by the forecast – the time horizon –
increases.
Elements of a Good Forecast
1. Timely
2. Accurate
3. Reliable
4. Meaningful units
5. In writing
6. Simple to understand and use
7. Cost-effective
Steps in the Forecasting Process
Obtain,
Determine Select a
Establish a clean, and Monitor the
the purpose forecast- Make the
time analyse forecast
of the ting tech- forecast.
horizon. appro- errors.
forecast. nique.
priate data.
General Approaches to
Forecasting
Qualitative methods consist mainly of
subjective inputs, which often defy precise
numerical description.
Quantitative methods involve either the
projection of historical data or the
development of associative models that
attempt to utilize causal (explanatory)
variables to make a forecast.
Forecasting Techniques
Judgmental forecasts use subjective
inputs such as opinions from consumer surveys,
sales staff, managers, executives, and experts.
Time-series forecasts project patterns
identified in recent time-series observations.
Associative model uses explanatory
variables to predict future demand.
Qualitative Forecasts
Executive Opinions
Used for long range planning
Salesforce Opinions
Consumer Surveys
Other Approaches
Delphi Method
Qualitative Forecasts
Forecasts Based on Time-Series
Data
A time series is a time-ordered sequence of
observations taken at regular intervals (e.g.,
hourly, daily, weekly, monthly, quarterly, annually).
Forecasting techniques based on time-series data
are made on the assumption that future values of
the series can be estimated from past values.
Analysis of time-series data requires the analyst to
identify the underlying behavior of the series. This
can often be accomplished by merely plotting the
data and visually examining the plot.
Forecasts Based on Time-Series
Data
Trend refers to a long-term upward or downward
movement in the data.
Irregular variation is caused by unusual circumstances,
not reflective of typical behavior.
Random variations are residual variations that remain
after all other behaviors have been accounted for.
Seasonality refers to short-term regular variations related
to the calendar or time of day.
Cycles are wavelike variations of more than one year’s
duration.
Forecasts Based on Time-Series
Data
NAÏVE FORECASTING
¨ Assumes demand in next
period is the same as demand
in most recent period
¨ e.g., If May sales were 48, then June
sales will be 48
¨ Sometimes cost effective &
efficient
© 1995 Corel Corp.
NAÏVE FORECASTING
A naive forecast uses a single previous
value of a time series as the basis of a
forecast
Can be used with a stable series
(variations around an average), with
seasonal variations, or with trend
Techniques for Averaging
Averaging techniques smooth fluctuations
in a time series because the individual
highs and lows in the data offset each other
when they are combined into an average.
A forecast based on an average thus tends
to exhibit less variability than the original
data.
Simple Moving Average
Technique that averages a number of
recent actual values, updated as new
values become available.
MA Demand in Previous n Periods
n
Moving Average Example
• The table below shows the historical supply of
palm civet coffee for the past 5 years:
Year Supply Determine the
projected supply for
2014 3,532 the next five (5) years
2015 4,356 using
2016 4,972 a. 5-year simple
moving average
2017 5,561 b. 3-year simple
2018 6,238 moving average
Moving Average Solution
a. Using 5-year SMA
Projected
Year
Supply
(3,532+4,356+4,972+5,561+6,23
2019 4,932
8)/5
2020
2021
2022
2023
Moving Average Solution
a. Using 5-year SMA
Year Supply
2014 3,532
2015 4,356
2016 4,972
2017 5,561
2018 6,238
2019 4,932
Moving Average Solution
Projected
Year
Supply
(3,532+4,356+4,972+5,561+6,23
2019 4,932
8)/5
(4,356+4,972+5,561+6,238+4,93
2020 5,212
2)/5
2021
2022
2023
Simple Moving Average
a. Using 5-year SMA
Projected
Year
Supply
(3,532+4,356+4,972+5,561+6,238
2019 4,932
)/5
(4,356+4,972+5,561+6,238+4,932
2020 5,212
)/5
(4,972+5,561+6,238+4,932+5,212
2021 5,383
)/5
(5,561+6,238+4,932+5,212+5,383
2022 5,465
)/5
Simple Moving Average
a. Using 3-year SMA
Projected
Year
Supply
2019 (4,972+5,561+6,238)/3 5,590
2020 (5,561+6,238+5,590)/3 5,796
2021 (6,238+5,590+5,796)/3 5,875
2022 (5,590+5,796+5,875)/3 5,754
2023 (5,796+5,875+5,754)/3 5,808
Weighted Moving Average
Similar to a moving average, except that it assigns more weight to the
most recent values in a time series.
Weighted Moving Average
Compute a weighted moving average for F6 using a
weight of .40 for the most recent period, .30 for the
next most recent, .20 for the next, and .10 for the
next.
Weighted Moving Average
Forecast for period 5 using a 3
Period Demand year weighted moving average
1 60 Forecast for period 6 using a 3
2 65
year weighted moving average if
2 55
period 5 is 64
4 58
5
Weights: 50%, 30%, 20%
6
F5 = 65(.20)+55(.30)+58(.50) =
58.5
F6 = 55(.20)+58(.30)+64(.50) =
60.4
Exponential Smoothing Model
is a forecasting model that uses a sophisticated weighted
average procedure to obtain a forecast
Considerations when using Exponential Smoothing Model
The current period’s forecast,
The current period’s actual value
The value of a smoothing coefficient, which varies
between 0 and 1
Exponential Smoothing Model
The Hot Tamale Mexican restaurant uses
exponential smoothing to forecast monthly
usage of tabasco sauce. Its forecast for
September was 200 bottles, whereas actual
usage in September was 300 bottles. If the
restaurant’s managers use an of 0.70, what
is their forecast for October?
F=
F=
F = (.70)(300)+(.30)(200)
F = 270 bottles
Exponential Smoothing Model
Period Actual Forecast Compute of Period 5
1 60 using exponential
2 65
3 55
smoothing.
4 58
5
F3= .40(65)+(1-.40)(60) = 62
F4= .40(55)+(1-.40)(62) = 59. 2
F5= .40(58)+(1-40)(59.2) = 58.72
Forecast Accuracy
Forecast Accuracy
Mean Absolute Deviation (MAD)
Measure of forecast error that computes
error as the average of the sum of the
absolute errors.
Mean Squared Error (MSE)
Measure of forecast error that computes
error as the average of the squared error.
Mean Absolute Percent Error (MAPE
The average absolute percent error.
