Discussion 7 - 327
wtfbm69
IPPTChap011.pdf
Chapter 11:
Forecasting
Operations Management in the
Supply Chain: Decisions and Cases,
6th edition
Copyright © 2013 by The McGraw-Hill Companies, Inc. All rights reserved.McGraw-Hill/Irwin
11-2
Chapter 11 Outline
• Forecasting for Decision Making
• Qualitative Forecasting Methods
• Time-Series Forecasting
• Moving Average
• Exponential Smoothing
• Forecasting Accuracy
• Advanced Time-Series Forecasting
• Causal Forecasting Methods
• Selecting a Forecasting Method
• Collaborative Planning, Forecasting, and Replenishment
11-3
Forecasting for Decision Making
• Forecasting demand for operations output
• Forecasting: what we think demand will be
• Planning: what we think demand should be
• Demand: may differ from sales
• Forecasts are used in all functional areas
• Forecasts necessary for operations decision areas:
process design, capacity planning, inventory
management, scheduling
11-4
Use of Forecasting: Operations Decisions
Time
Horizon
Accuracy
Required
Number of
Forecasts
Management
Level
Forecasting
Method
Process design Long Medium Single or few Top
Qualitative or
causal
Capacity
planning,
facilities
Long Medium Single or few Top Qualitative and
causal
Aggregate
planning Medium High Few Middle Causal and time
series
Scheduling Short Highest Many Lower Time series
Inventory
management Short Highest Many Lower Time series
11-5
Use of Forecasting: Marketing, Finance, & HR
Time
Horizon
Accuracy
Required
Number of
Forecasts
Management
Level
Forecasting
Method
Long-range
marketing
programs
Long Medium Single or few Top Qualitative
Pricing decisions Short High Many Middle Time series
New product
introduction
Medium Medium Single Top Qualitative and
causal
Cost estimating Short High Many Lower Time series
Capital
budgeting
Medium Highest Few Top Causal and time
series
Labor planning Medium Medium Few Lower Qualitative and
time series
11-6
‘Qualitative’ Forecasting Methods
• Based on managerial judgment when there is a lack of
data. No specific model.
• Major methods:
• Delphi technique
• Market surveys
• Life-cycles analogy
• Informed judgment (naïve models)
11-7
Time-Series Forecasting • Components of data:
• Level - average
• Trend - general direction (up or down)
• Seasonality - short term recurring cycles
• Cycle - long term business cycle
• Error - random or irregular component
11-8
Moving Average • Assumes no trend, seasonality, or cycle
• Simple moving average:
• Weighted moving average:
N
DDD A Nttt
t 11
......
tt AF
1
11211 ......
NtNtttt DWDWDWAF
11-9
Moving Average Example
Period Actual Demand Forecast
1 10
2 18
3 29
4 - 19
Compute three period moving average for Period 4
(number of periods is forecaster’s decision)
F4 = A3 = (29 + 18 + 10) / 3 = 19
F5 will be (Actual demand for period 4 + 29 + 18) / 3
11-10
Time-Series Data (Figure 11.2)
Note: The forecast is smoother as the number of periods
in the moving average increases.
11-11
Exponential Smoothing
1
1
ttt
ADA
• The new average is computed from the old average:
• The value of the smoothing constant () is a choice.
It determines how much the calculation weights
recent demand (smooths random variation). Αlpha
(α) can be between 0 and 1, but is usually 0.1 - 0.2.
11-12
Simple Exponential Smoothing
• The forecast:
F = forecast of demand (both this period and next)
D = actual demand (this period)
t = time period
• Assumes no trend, seasonality, or cycle
• Note: we are adjusting Ft to get Ft+1
tttt
FDFF
1
11-13
Exponential Smoothing Example
• September forecast for sales was 15, but actual sales were
13. Use alpha (α) of 0.2.
• What is the forecast for October?
• October forecast = September forecast + α(September
actual - September forecast)
= 15 + 0.2(13 - 15)
= 15 - 0.4 = 14.6
11-14
Forecast Errors
In addition to the forecast, one should compute an estimate of forecast error. Its uses include:
• To monitor erratic demand observations or “outliers”
• To determine when the forecasting method is no longer tracking actual demand
• To determine the parameter values that provide the forecast with the least error
• To set safety stocks or safety capacity
11-15
Time-Series Data (Figure 11.3)
Note: The forecast is smoother as the value
of alpha (α) is reduced.
11-16
Forecast Accuracy
• Cumulative sum of forecast errors (CFE)
• Mean square error (MSE)
• Mean absolute deviation (MAD)
• Mean absolute percentage errors (MAPE)
• Tracking Signal (TS)
11-17
Forecast Accuracy: Formulas
Cumulative sum of
forecast errors
Mean square error
Mean absolute
deviation
Mean absolute
percentage errors
Tracking signal
Cumulative sum of
forecast errors
Mean square error
Mean absolute
deviation
11-18
Advanced Time-Series Forecasting • Adaptive exponential smoothing
• Smoothing coefficient () is varied
• Mathematical models
• Linear or nonlinear
• Box-Jenkins method
• Requires about 60 periods of past data
11-19
Causal Forecasting Methods • Cause-and-effect model, using other data set to predict
demand (forecast).
• Examples:
• Use population to forecast newspaper sales
• Use supply chain data on inventory levels to forecast flat
screen TV sales
11-20
Causal Forecasting Models
• The general regression model:
• Other forms of causal model:
• Econometric
• Input-output
• Simulation models
xbay ˆ
11-21
Example of Causal Model
I t D t F t
34.6 120 121.15
35.7 124 123.79
36.3 119 125.22
35.2 124 122.59
35.7 125 123.79
36.4 130 125.46
37.6 128.34
Intercept (a) 38.23094
Slope (b) 2.396514
Yt = a + b(t)
F7 = 38.23 + 2.397 (7) = 128.34 = forecast for Period 7
Dt = actual sales in year t
Ft = forecasted sales
It = median family
income (000’s)
11-22
Selecting a Forecasting Method
• User and system sophistication • People are reluctant to use what they don’t understand
• Time and resources available • When is forecast needed?
• Use or decision characteristics • Scheduling decision? Facility expansion?
• Data availability
• Data pattern • Level? Unstable?
11-23
Collaborative Planning, Forecasting, and
Replenishment (CPFR) • Aim is to achieve more accurate forecasts
• Share information in the supply chain with customers and
suppliers
• Compare forecasts
• If discrepancy, look for reason
• Agree on consensus forecast
• Works best in B2B with few customers (e.g., a small
number of large retailers)
11-24
Chapter 11 Summary
• Forecasting for Decision Making
• Qualitative Forecasting Methods
• Time-Series Forecasting
• Moving Average
• Exponential Smoothing
• Forecasting Accuracy
• Advanced Time-Series Forecasting
• Causal Forecasting Methods
• Selecting a Forecasting Method
• Collaborative Planning, Forecasting, and Replenishment
