Forecasting CPIM
Post on 11-Oct-2015
156 Views
Preview:
DESCRIPTION
Transcript
5/21/2018 Forecasting CPIM
1/100
McGraw-Hill/Irwin Copyright 2008 by The McGraw-Hill Companies, Inc. All rights reserved.
Demand Management
Processing,Influencing, &Anticipating
Demand
BuySell BuyStore
MoveMake Sell Store
MoveSell
MakeMake
MoveMoveBuy
5/21/2018 Forecasting CPIM
2/100
4-2
Managing the sell side of a business
Plant
Plant
Plant
Warehouse
Suppliers
Customer
s
Supply-Demand Management
"Make, Move, Store"
SupplierRelationship
Management
"Buy"
CustomerRelationshipManagement
"Sell"
5/21/2018 Forecasting CPIM
3/100
4-3
Key questions1. What is the scope of demand management?
2. What does order processing involve; why is it an important area for management
attention?3. What is customer profit potential, & how is it relevant for influencing demand?
4. What are 5 alternatives for improving forecast accuracy, what do they mean, & howcan they be applied?
5. How do the tactics ofpart standardization&postponement of form or placehelpimprove forecast accuracy?
6. What is the difference between long term & short term forecasting?
7. What are 4 long term forecasting methods; what are the risks ofsalesperson/customer input?
8. What are the components of demand, & which component is not forecasted?
9. How do the moving average, Winters, & focus forecasting methods work?
10. What is the role of the number of periods in the moving average method, & thesmoothing parameters in the Winters method?
11. What is the purpose of filtering, & why is it important for computer-basedforecasting?
12. What do the following principles of nature mean & how are they relevant fordemand management? (1) law of large numbers, (2) trumpet of doom, (3) recencyeffect, (4) hockey stick effect, (5) Pareto phenomenon
13. What are the managerial insights from the chapter?
5/21/2018 Forecasting CPIM
4/100
4-4
Road map
Processing Demand
Influencing Demand
How to Improve Forecast Accuracy
Long Term Forecasting
Short Term Forecasting
Summary
5/21/2018 Forecasting CPIM
5/100
4-5
Scope of demand management
So what is demand management?
Concerned with processing, influencing,and anticipating demand
Well begin with processing demand or,in more common terms, order
processingor order fulfillment
5/21/2018 Forecasting CPIM
6/100
4-6
Order processing
Order processing is usually viewed to spanorder booking to order shipment
Example steps?
Customer validation, order entry, credit checking,pricing, design changes, availability checks, deliverytime estimation, notification of shipment, notificationof delays
Processing Demand
5/21/2018 Forecasting CPIM
7/1004-7
Processing Demand
CUSTOMER ORDER ENTRY AND
CHECKING
Customer Validation
Credit Control Operations
ER
P
INVOICING
SHIPPING
CUSTOMER SERVICE
ORDER
INTERRUPTION
ORDERPICKING AND
ASSEMBLY
RETURNS
5/21/2018 Forecasting CPIM
8/1004-8
Characteristics Can be a complex & time consuming process
dealing largely with information flow
Susceptible to ad hoc modifications over time inresponse to problems (e.g., extra credit check addeddue to expensive nonpaying customer a few yearsago)
A major customer contact point withorganization
Can significantly impact customer perceptions
IT advances & high customer impact
A potential profitable target for improvement
Processing Demand
5/21/2018 Forecasting CPIM
9/1004-9
Example 1
Benetton
Electronic loop linking sales agent, factory, & warehouse
If not available, measurements transferred to knitting
machine for production
Benetton uses a single warehouse
Staffed by 8 people & about 230,000 pieces shipped daily
Processing Demand
5/21/2018 Forecasting CPIM
10/1004-10
Example 2
K-Mart and MasterLock
Policy for mistake in shipment or invoice
Strike 1: $10,000, Strike 2: $50,000, Strike 3: lose
business
MasterLock revamped their order processing function
Processing Demand
5/21/2018 Forecasting CPIM
11/1004-11
Example 3customer tools
Processing Demand
Amazon online order tracking
5/21/2018 Forecasting CPIM
12/1004-12
Example 4customer tools
Processing Demand
UPS online order tracking
5/21/2018 Forecasting CPIM
13/1004-13
Example 4continued
Processing Demand
UPS online tools
5/21/2018 Forecasting CPIM
14/1004-14
Road map Processing Demand
Influencing Demand
How to Improve Forecast Accuracy
Long Term Forecasting
Short Term Forecasting
Summary
5/21/2018 Forecasting CPIM
15/1004-15
Measure customer profit potential
Some customers are more profitable than others
Advancing technologies more practical to estimate profitpotential of individual customers
Can guide efforts/investments for customer retention &acquisition . . . investments to influence demand
E.g.,
Electronics manufacturer: reviews historical customer profit beforesending service contract renewal
Wireless phone firm: churn scores& lifetime valueestimatesinfluence # of customer contacts & attractiveness of offerings
Ongoing development of data mining methods
Influencing Demand
A simple idea
5/21/2018 Forecasting CPIM
16/1004-16
Road map
Processing Demand
Influencing Demand
How to Improve Forecast Accuracy
Long Term Forecasting
Short Term Forecasting
Summary
F ti Alt ti
5/21/2018 Forecasting CPIM
17/1004-17
Motivating example 1
Sunbeam
Improved forecasting led to 45% reduction ininventory
Included estimates from top 200 customers
Forecasting Alternatives
F ti Alt ti
5/21/2018 Forecasting CPIM
18/1004-18
Motivating example 2
Apple
A history of problems forecasting demand
Many components sourced from 1 supplier -accurate forecasts are critical
Over $1 billion in unfilled orders during thecrucial holiday season. The CEO (Spindler)ousted a few months later
Forecasting Alternatives
F ti Alt ti
5/21/2018 Forecasting CPIM
19/1004-19
Motivating example 3
IBM
Badly misjudged demand in PC business in 1996went from being profitable in 1995 to a $200million loss through 1sthalf of 1996
Forecasting Alternatives
F ti Alt ti
5/21/2018 Forecasting CPIM
20/1004-20
Motivating example 4
Christmas 1999 & e-commerce takesoff
Large unanticipated increase in Internet orders
didnt ship on time
E.g., Many Toys R Us Christmas orders not delivereduntil March I will never buy online again
Forecasting Alternatives
ForecastingAlternatives
5/21/2018 Forecasting CPIM
21/1004-21
Improvement alternatives
Change the forecasting method
Collect more or different data
Analyze the information differently
E.g., involve more people, new forecasting software, spend moretime manually reviewing, focus groups etc.
Change operations or operating policies
Introduce early warning mechanisms
Take advantage of the law of large numbers
Reduce information delays & leadtimes (trumpet of doom)
Reduce demand volatility
Forecasting Alternatives
ForecastingAlternatives
5/21/2018 Forecasting CPIM
22/1004-22
Early warning
Change policies so that some (or more)customers provide earlier commitmentoffuture demand, e.g.,
Early bird program for builder markets discount for60-day advance order
Invite large buyers to Aspen in February to view nextyears skiwear line, & encourage orders
Commitment asking customers how muchthey are likely to buy next quarter
Forecasting Alternatives
ForecastingAlternatives
5/21/2018 Forecasting CPIM
23/1004-23
Law of large numbers
As volume increases, relative variability decreases
Postponement in form or place, e.g.,
Dell configure your own PC
From full product line at 12 regional DCs to full product line at asingle super DC, with 10% of product line stocked at 11 regionalDCs (i.e., fast movers that account for 70% of sales)
Part standardization, e.g.,
Arbys sandwich wrappers; plastic lids with push down drinkindicator
Intel Pentium processors all the same size
- 2.8 GHz tests out below 2.8 spec can be sold as a 2.66 GHz chip (down-binning)
Forecasting Alternatives
Principle of Nature
ForecastingAlternatives
5/21/2018 Forecasting CPIM
24/1004-24
Trumpet of doom
As forecast horizonincreases, accuracydecreases, e.g.,
Reduce production & delivery leadtimes
Dell pick-to-light system for assembly
Reduce information delays
EDI transmission of daily consumer demand up
through multiple levels in the supply chain
Forecasting Alternatives
0
Forecast Error Range over Time
Time Until Forecast Event0
PercentageForecast
Error
Principle of Nature
ForecastingAlternatives
5/21/2018 Forecasting CPIM
25/1004-25
Reduce demand volatility
Beware of product proliferation
Pareto analysis separating the important few from the trivial many
Periodic length of line analysis to critically assess whether to continuallyoffer slow movers
Principle of Nature: Pareto phenomenonthe lions share of anaggregate measure is determined by relatively few factors
E.g., the 80-20 rule 80% of demand is due to 20% of product line
Beware of perverse cycle of promotions customers wait forsale before buying, thereby forcing a sale
A step further dynamic pricing to stabilize demand & align with supply
Reduce the hockey stick effect
Forecasting Alternatives
2 Principles of Nature
ForecastingAlternatives
5/21/2018 Forecasting CPIM
26/100
4-26
Hockey stick effect
Volume tends to pick up towards the endof a reporting period . . . why?
Look for ways to lessen the effect contributes to demand volatility,inefficiency, poor service
Jan Feb
Principle of Nature
Forecasting Alternatives
ForecastingAlternatives
5/21/2018 Forecasting CPIM
27/100
4-27
Channel stuffing
Lots of sales booked near the end of a quarter,then sales drop off at the start of the nextquarter
E.g.,
A large brewer offered a vacation to the salespersonin each region who sold the most beer to stores over
a 3 month period
One winner was able to convince a few stores to freeup backroom space and fill it entirely with beer
One contributor to the hockey stick effect
Forecasting Alternatives
ForecastingAlternatives
5/21/2018 Forecasting CPIM
28/100
4-28
Were about to focus onmethods for predictingdemand
Forecasting Alternatives
Improvement alternatives
short pork bellies
But, important to remember . . . many creative ways toimprove forecast accuracy that have nothing to do withmethod
E.g., early warning incentives, law of large numbers, trumpet ofdoom, reduce demand volatility
5/21/2018 Forecasting CPIM
29/100
4-29
Road map Processing Demand
Influencing Demand
How to Improve Forecast Accuracy
Long Term Forecasting
Short Term Forecasting
Summary
Long Term Forecasting
5/21/2018 Forecasting CPIM
30/100
4-30
Characteristics of long term forecasts
Single or multi-year horizon
Monthly or annual time bucket
Aggregate units
Input to long term decisions
Accuracy generally more important than short termforecasts . . . why?
Tend to use expensive & time consuming methods . . .due to the preceding point & due to a PON . . . which is?
Long Term Forecasting
Long Term Forecasting
5/21/2018 Forecasting CPIM
31/100
4-31
Recency effect
Humans tend to overreact to (or be overlyinfluenced by) recent events
E.g.,
Hughes Electronics Corp. developed an artificialintelligence based financial trading system. Thedevelopers did this by encoding the wisdom of
Christine Downton, a successful portfolio manager.One motivation for creating the system is that it isimmune to the recency effect, i.e., humans tend toget overly fixated on the most recent information.
g g
Principle of Nature
Long Term Forecasting
5/21/2018 Forecasting CPIM
32/100
4-32
Some alternative methods
Judgment
Salesperson & customer input
Great information source, but beware of bias potential& recency effect= humans tend to be overlyinfluenced by recent events
Outside services
Causal methods . . . examples?
g g
5/21/2018 Forecasting CPIM
33/100
4-33
Road map Processing Demand
Influencing Demand
How to Improve Forecast Accuracy
Long Term Forecasting
Short Term Forecasting
Characteristics
Components of demand
Moving average
Winters method
Focus forecasting
Filtering
Summary
Short Term Forecasting
5/21/2018 Forecasting CPIM
34/100
4-34
Long term/short term characteristics
Long term forecasts
Single or multi-year horizon
Monthly or annual time bucket
Aggregate units (e.g., product/service categories)
Input to long term decisions
Expensive & time consumingmethods
Accuracy importance
Trumpet of doom
Short term forecasts
Weekly or monthly horizon
Daily & weekly time bucket
Detailed units (e.g., SKU)
Input to short term decisions
Inexpensive & quick methods
Accuracy importance
Trumpet of doom
Could argue using 2 different principles of nature that its [easier?/harder?] to be
accurate with short term forecasting than with long term forecasting
g
5/21/2018 Forecasting CPIM
35/100
4-35
Definition of the Forecasting Process
The Art and Science of Predicting FutureEvents
Forecasting vs. Predicting
Based on Past Data
Economic vs. Demand Forecasting
5/21/2018 Forecasting CPIM
36/100
4-36
Elements of Demand Forecasting
Dynamic in Nature
Consider Uncertainty (Stochastic)
Rely on Information contained in PastData
Applied to various time horizons
short term
medium term forecasts
long term forecasts
5/21/2018 Forecasting CPIM
37/100
4-37
Steps in the Forecasting Process
Determine the Use of the Forecast Select the Items to be Forecasted Determine a Suitable Time Horizon Select an appropriate Set of Forecasting Models
Gather Relevant Data Conduct the Analysis Validate the Model - Assess its Accuracy Make the Forecast Implement the Results
5/21/2018 Forecasting CPIM
38/100
4-38
Independent Demand:What a firm can do to manage it?
Can take an active role to influencedemand
FORECASTING
Can take a passive role and simplyrespond to demand
5/21/2018 Forecasting CPIM
39/100
4-39
Types of Forecasts
Qualitative (Judgmental)
Quantitative Time Series Analysis
Causal Relationships
Simulation
5/21/2018 Forecasting CPIM
40/100
4-40
Qualitative Methods
Grass Roots
Market Research
Panel Consensus
Executive Judgment
Historical analogy
Delphi Method
Qualitative
Methods
5/21/2018 Forecasting CPIM
41/100
4-41
Delphi Method
1. Choose the experts to participate representing avariety of knowledgeable people in differentareas
2. Through a questionnaire (or E-mail), obtainforecasts (and any premises or qualifications forthe forecasts) from all participants
3. Summarize the results and redistribute them tothe participants along with appropriate newquestions
4. Summarize again, refining forecasts andconditions, and again develop new questions
5. Repeat Step 4 as necessary and distribute thefinal results to all participants
5/21/2018 Forecasting CPIM
42/100
4-42
Quantitative Forecasting Models
Both Pattern Based and CorrelationalModels rest on the assumption that therelationships of the past will continueinto the Future
Both can Mathematically Characterize theProbabilistic Nature of the Forecast
Both Use Information from RelevantTime Frames
5/21/2018 Forecasting CPIM
43/100
4-43
Road map Processing Demand
Influencing Demand
How to Improve Forecast Accuracy
Long Term Forecasting
Short Term Forecasting
Characteristics
Components of demand
Moving average
Winters method
Focus forecasting
Filtering
Summary
5/21/2018 Forecasting CPIM
44/100
4-44
Components of Demand
Average demand for a period oftime
Trend
Seasonal element
Cyclical elements
Random variation
Autocorrelation
5/21/2018 Forecasting CPIM
45/100
4-45
Pattern Based Analyses
Definition Identifying an underlying pattern in historical
data, describe it in mathematical terms, andthen extrapolate it into the future
Uses a Time Series of Past Data
5/21/2018 Forecasting CPIM
46/100
4-46
Time Series Variation
Time Series of Demand Data TypicallyContain Four Components of VariationAbout the Mean or Average
Pattern Based Forecasting Needs toMathematically Characterize Each ofthese
Fi di C t f D d
5/21/2018 Forecasting CPIM
47/100
4-47
Finding Components of Demand
1 2 3 4
x
x xx
xx
x xx
xxx x x
xxxxxx x x
xx
x x xx
xx
xx
x
xx
xx
xx
x
xx
x xx
x
x
Year
Sales
Seasonal variation
Linear
Trend
Average
5/21/2018 Forecasting CPIM
48/100
4-48
Time Series Analysis
Time series forecasting models try topredict the future based on past data
You can pick models based on:
1. Time horizon to forecast2. Data availability
3. Accuracy required
4. Size of forecasting budget
5. Availability of qualified personnel
5/21/2018 Forecasting CPIM
49/100
4-49
Simple Moving Average Formula
F =A + A + A +...+A
nt
t-1 t-2 t-3 t-n
The simple moving average model assumes
an average is a good estimator of futurebehavior
The formula for the simple moving averageis:
Ft= Forecast for the coming periodn = Number of periods to be averaged
A t-1= Actual occurrence in the past period for up to n
periods
5/21/2018 Forecasting CPIM
50/100
4-50
Simple Moving Average Problem (1)
Week Demand
1 650
2 678
3 720
4 785
5 859
6 920
7 850
8 758
9 892
10 920
11 789
12 844
F =A + A + A +...+A
nt t-1 t-2 t-3 t-n
Question: What are the 3-week and 6-week movingaverage forecasts fordemand?
Assume you only have 3weeks and 6 weeks ofactual demand data for the
respective forecasts
Calculating the moving averages gives us:
5/21/2018 Forecasting CPIM
51/100
4-51
Week Demand 3-Week 6-Week
1 650
2 678
3 720
4 785 682.67
5 859 727.67
6 920 788.00
7 850 854.67 768.67
8 758 876.33 802.00
9 892 842.67 815.3310 920 833.33 844.00
11 789 856.67 866.50
12 844 867.00 854.83
F4=(650+678+720)/3
=682.67
F7=(650+678+720
+785+859+920)/6
=768.67
5/21/2018 Forecasting CPIM
52/100
4-52
500
600
700
800
900
1000
1 2 3 4 5 6 7 8 9 10 11 12
Week
Deman
d Demand
3-Week
6-Week
Plotting the moving averages and comparing
them shows how the lines smooth out to reveal
the overall upward trend in this example
Note how the
3-Week issmoother than
the Demand,
and 6-Week is
even smoother
5/21/2018 Forecasting CPIM
53/100
4-53
Simple Moving Average Problem (2) Data
Week Demand
1 820
2 775
3 680
4 655
5 620
6 600
7 575
Question: What is the 3week moving averageforecast for this data?
Assume you only have3 weeks and 5 weeksof actual demanddata for therespective forecasts
5/21/2018 Forecasting CPIM
54/100
4-54
Simple Moving Average Problem (2) Solution
Week Demand 3-Week 5-Week
1 820
2 7753 680
4 655 758.33
5 620 703.33
6 600 651.67 710.00
7 575 625.00 666.00
F4=(820+775+680)/3
=758.33 F6=(820+775+680
+655+620)/5
=710.00
W i ht d M i A F l
5/21/2018 Forecasting CPIM
55/100
4-55
Weighted Moving Average Formula
F = w A + w A + w A + ...+ w At 1 t -1 2 t -2 3 t -3 n t - n
w = 1ii=1
n
While the moving average formula implies an equalweight being placed on each value that is being
averaged, the weighted moving average permits an
unequal weighting on prior time periods
wt = weight given to time period t
occurrence (weights must add to one)
The formula for the moving average is:
5/21/2018 Forecasting CPIM
56/100
4-56
Weighted Moving Average Problem (1) Data
Weights:t-1 .5
t-2 .3
t-3 .2
Week Demand
1 650
2 678
3 720
4
Question: Given the weekly demand and weights, what is
the forecast for the 4thperiod or Week 4?
Note that the weights place more emphasis on the
most recent data, that is time period t-1
5/21/2018 Forecasting CPIM
57/100
4-57
Weighted Moving Average Problem (1) Solution
Week Demand Forecast
1 650
2 6783 720
4 693.4
F4= 0.5(720)+0.3(678)+0.2(650)=693.4
5/21/2018 Forecasting CPIM
58/100
4-58
Weighted Moving Average Problem (2) Data
Weights:
t-1 .7
t-2 .2t-3 .1
Week Demand
1 820
2 775
3 680
4 655
Question: Given the weekly demand information and
weights, what is the weighted moving average forecast
of the 5thperiod or week?
5/21/2018 Forecasting CPIM
59/100
4-59
Weighted Moving Average Problem (2) Solution
Week Demand Forecast
1 820
2 775
3 680
4 655
5 672
F5= (0.1)(755)+(0.2)(680)+(0.7)(655)= 672
Short Term ForecastingMoving Average and Weighted Moving Average
5/21/2018 Forecasting CPIM
60/100
4-60
Some pros/cons
1. Simple (+)
2. Designated weights of history (-)
3. History cut-off beyond mperiods (-)
Exponential Smoothing Model
5/21/2018 Forecasting CPIM
61/100
4-61
Exponential Smoothing Model
Premise: The most recent observations mighthave the highest predictive value
Therefore, we should give more weight to themore recent time periods when forecasting
Ft= Ft-1 + a(At-1 - Ft-1)
constantsmoothingAlpha
periodepast t timin theoccuranceActualA
periodpast time1inalueForecast vF
periodt timecomingfor thelueForcast vaF
:Where
1-t
1-t
t
a
5/21/2018 Forecasting CPIM
62/100
4-62
Exponential Smoothing Problem (1) Data
Week Demand
1 820
2 775
3 6804 655
5 750
6 802
7 7988 689
9 775
10
Question: Given theweekly demand data,what are theexponential smoothingforecasts for periods 2-
10 using a=0.10 anda
=0.60?Assume F1=D1
Answer: The respective alphas columns denote the forecast
5/21/2018 Forecasting CPIM
63/100
4-63
Week Demand 0.1 0.6 1 820 820.00 820.00
2 775 820.00 820.00
3 680 815.50 793.00
4 655 801.95 725.20
5 750 787.26 683.08
6 802 783.53 723.23
7 798 785.38 770.498 689 786.64 787.00
9 775 776.88 728.20
10 776.69 756.28
p p
values. Note that you can only forecast one time period into
the future.
5/21/2018 Forecasting CPIM
64/100
4-64
Exponential Smoothing Problem (1) Plotting
500
600
700
800
900
1 2 3 4 5 6 7 8 9 10
Week
Demand Demand
0.1
0.6
Note how that the smaller alpha results in a smoother line inthis example
5/21/2018 Forecasting CPIM
65/100
4-65
Exponential Smoothing Problem (2) Data
Question: What are
the exponential
smoothing forecastsfor periods 2-5 using
a =0.5?
Assume F1=D1
Week Demand
1 8202 775
3 680
4 6555
5/21/2018 Forecasting CPIM
66/100
4-66
Exponential Smoothing Problem (2) Solution
Week Demand 0.5
1 820 820.00
2 775 820.00
3 680 797.504 655 738.75
5 696.88
F1=820+(0.5)(820-820)=820 F3=820+(0.5)(775-820)=797.75
5/21/2018 Forecasting CPIM
67/100
4-67
Seasonal Adjustments
Applied to Moving Averages and TimeSeries Regression
First, Calculate a Seasonal Index (SI)Factor for Each Relevant Time Period(day, week, month, quarter)
Each Seasonal Periods SI isCalculated by Averaging the Ratio ofits Actual Demand to the ForecastDemand for all Corresponding Periods
S l Adj t t
5/21/2018 Forecasting CPIM
68/100
4-68
Seasonal Adjustments
Forecast for Future Periods is Calculatedby Multiplying the Unadjusted MovingAverage or Time Series Forecast for a
given Period by the CorrespondingSeasonal Index for that Period
i.e. if the SMA forecast for the month ofMarch is 27 and the SI for March is
1.125, then
Emar= 27*1.125 = 30.375
Seasonal Adjustment Example
5/21/2018 Forecasting CPIM
69/100
4-69
Seasonal Adjustment Example
Seasonal Adjustments
Sales Demand
Month 1993 1994Monthly
Average
Overall
AverageSeasonal Index
SI Adjusted
Forecast
Jan 80 100 90.00 94.00 0.96 86.17
Feb 75 85 80.00 94.00 0.85 68.09
Mar 80 90 85.00 94.00 0.90 76.86
Apr 90 110 100.00 94.00 1.06 106.38May 115 131 123.00 94.00 1.31 160.95
Jun 110 120 115.00 94.00 1.22 140.69
Jul 100 110 105.00 94.00 1.12 117.29
Aug 90 110 100.00 94.00 1.06 106.38
Sep 85 95 90.00 94.00 0.96 86.17
Oct 75 85 80.00 94.00 0.85 68.09
Nov 75 85 80.00 94.00 0.85 68.09
Dec 80 80 80.00 94.00 0.85 68.09
Average 87.92 100.08
Expected Demand for 1995 = 1153.23
Seasonal Adjustments
5/21/2018 Forecasting CPIM
70/100
4-70
Seasonal AdjustmentsExample Graph
Seasonal Adjusted Forecasting
50
70
90
110
130
150
170
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
1993
1994
SI Adjusted
Forecast
Overall
Average
Evaluating Forecast Accuracy
5/21/2018 Forecasting CPIM
71/100
4-71
Evaluating Forecast Accuracy
Use of Residuals Analyses Residuals are the Difference Between the
Forecast and the Actual Demand for a GivenPeriod
Assessed by Several Measures Mean Absolute Deviation - MAD
Mean Squared Error - MSE
Tracking Signal
5/21/2018 Forecasting CPIM
72/100
4-72
The MAD Statistic to DetermineForecasting Error
MAD =
A - F
n
t t
t=1
n
1 MAD 0.8 standard deviation
1 standard deviation 1.25 MAD
The ideal MAD is zero which wouldmean there is no forecasting error
The larger the MAD, the less theaccurate the resulting model
MAD Problem Data
5/21/2018 Forecasting CPIM
73/100
4-73
MAD Problem Data
Month Sales Forecast
1 220 n/a
2 250 255
3 210 205
4 300 320
5 325 315
Question: What is the MAD value given
the forecast values in the table below?
MAD Problem Solution
5/21/2018 Forecasting CPIM
74/100
4-74
MAD Problem Solution
MAD =
A - F
n=
40
4= 10
t t
t=1
n
Month Sales Forecast Abs Error1 220 n/a
2 250 255 5
3 210 205 5
4 300 320 20
5 325 315 10
40
Note that by itself, the MAD
only lets us know the mean
error in a set of forecasts
5/21/2018 Forecasting CPIM
75/100
4-75
Evaluating Forecast AccuracyMean Absolute Deviation - MAD
Exponentially Smoothed MAD
MADt= aMAD|Dt- Forecastt| + (1- aMAD)MADt-1
Evaluating Forecast Accuracy
5/21/2018 Forecasting CPIM
76/100
4-76
g yMean Squared Error - MSE
MSE = ((Di- Forecasti)
2
)/nPeriod
Actual
Demand
Time
Series
Forecast
Time
Series
Residual
Squared
Error
1 12 12.16 -0.16 0.03
2 13 12.13 0.87 0.76
3 10 12.09 -2.09 4.39
4 11 12.06 -1.06 1.135 10 12.03 -2.03 4.12
6 14 12.00 2.00 4.01
7 16 11.97 4.03 16.28
8 15 11.93 3.07 9.40
9 13 11.90 1.10 1.21
10 8 11.87 -3.87 14.97
11 10 11.84 -1.84 3.37
12 12 11.80 0.20 0.04
13 9 11.77 -2.77 7.69
14 13 11.74 1.26 1.59
15 13 11.71 1.29 1.67
MSE = 4.71
RMSE = 2.17
Tracking Signal Formula
5/21/2018 Forecasting CPIM
77/100
4-77
Tracking Signal Formula
The Tracking Signal or TS is a measure thatindicates whether the forecast average iskeeping pace with any genuine upward ordownward changes in demand.
Depending on the number of MADs selected, theTS can be used like a quality control chart
indicating when the model is generating toomuch error in its forecasts.
The TS formula is:
TS =RSFE
MAD=
Running sum of forecast errors
Mean absolute deviation
5/21/2018 Forecasting CPIM
78/100
4-78
Evaluating Forecast AccuracyTracking Signal
Tracking Signal = Running Sum ofForecast Error / MAD = RSFE/MAD
Period Actual
Demand
Time
Series
Forecast
Time
Series
Residual
RSFE MAD Tracking
Signal1 12 12.16 -0.16 -0.16 0.03 -5.00
2 13 12.13 0.87 0.72 0.20 3.58
3 10 12.09 -2.09 -1.38 0.58 -2.38
4 11 12.06 -1.06 -2.44 0.68 -3.61
5 10 12.03 -2.03 -4.47 0.95 -4.72
6 14 12.00 2.00 -2.47 1.16 -2.13
7 16 11.97 4.03 1.57 1.73 0.90
8 15 11.93 3.07 4.63 2.00 2.329 13 11.90 1.10 5.73 1.82 3.15
10 8 11.87 -3.87 1.86 2.23 0.84
11 10 11.84 -1.84 0.03 2.15 0.01
12 12 11.80 0.20 0.22 1.76 0.13
13 9 11.77 -2.77 -2.55 1.96 -1.30
14 13 11.74 1.26 -1.29 1.82 -0.71
15 13 11.71 1.29 0.00 1.72 0.00
Road map
5/21/2018 Forecasting CPIM
79/100
4-79
Road map
Processing Demand
Influencing Demand
How to Improve Forecast Accuracy
Long Term Forecasting
Short Term Forecasting
Characteristics
Components of demand
Moving average
Winters method
Focus forecasting
Filtering
Summary
Old i t
Short Term ForecastingWinters
5/21/2018 Forecasting CPIM
80/100
4-80
Old man winters
Winters method used to forecast one period into the future
See how method detects patterns & adapts to market changes overtime
Old Man Winters in Action
0.00
100.00
200.00
300.00
400.00
500.00
600.00
0 20 40 60 80 100
Time
Volume
Actual
Forecast
Short Term ForecastingWinters
5/21/2018 Forecasting CPIM
81/100
4-81
Key to Winters method
Winters is an exponential smoothingmethod
Smoothing is based on a key idea
For each component (which are?), a portionof difference between estimate & actual isdue to randomness& certain portion due
to real change
Short Term ForecastingWinters
5/21/2018 Forecasting CPIM
82/100
4-82
Smoothing in action...
New estimate = old estimate + (somepercentage)(error)
Smoothes out peaks & valleys (i.e.,randomness) of actual
Road map
5/21/2018 Forecasting CPIM
83/100
4-83
Road map Processing Demand
Influencing Demand
How to Improve Forecast Accuracy
Long Term Forecasting
Short Term Forecasting
Characteristics
Components of demand
Moving average
Winters method
Focus forecasting
Filtering
Summary
B i i i ht
Short Term ForecastingFocus
5/21/2018 Forecasting CPIM
84/100
4-84
Bernies insight
An intuitive & successful idea
Regularly use a # of different methods togenerate forecasts
Maintain historical accuracy information on each
method
Use the most accurate method to generateofficial forecasts
or what is focus forecasting?
Short Term ForecastingFocus
5/21/2018 Forecasting CPIM
85/100
4-85
Advertisementappearing in
APICS The
Performance
Advantage
Road map
5/21/2018 Forecasting CPIM
86/100
4-86
Road map
Processing Demand
Influencing Demand
How to Improve Forecast Accuracy
Long Term Forecasting
Short Term Forecasting
Characteristics
Components of demand
Moving average
Winters method
Focus forecasting
Filtering
Summary
Short Term ForecastingFiltering
5/21/2018 Forecasting CPIM
87/100
4-87
Two types of filters
An important feature of computer-based forecastingsystems
Large amounts of data impractical to manually review all
1. For data input errors (e.g., typos, scanner errors)
If |actual - forecast| > limit, then report
2. For unacceptable forecast errors (e.g., warranting
management attention)
If average absolute error > limit, then report
If average error (i.e., bias) > limit, then report
R d
5/21/2018 Forecasting CPIM
88/100
4-88
Road map Processing Demand
Influencing Demand
How to Improve Forecast Accuracy
Long Term Forecasting
Short Term Forecasting
Dependent Demand
Correlational Forecasting Summary
Demand Management
5/21/2018 Forecasting CPIM
89/100
4-89
Demand ManagementBill of Materials (BOM)
A
B(4) C(2)
D(2) E(1) D(3) F(2)
Dependent Demand:
Raw Materials,
Component parts,Sub-assemblies, etc.
Independent Demand:
Finished Goods
Web-Based Forecasting: CPFR
5/21/2018 Forecasting CPIM
90/100
4-90
Web Based Forecasting: CPFR
Collaborative Planning, Forecasting, and
Replenishment(CPFR) a Web-based tool used tocoordinate demand forecasting, production andpurchase planning, and inventory replenishmentbetween supply chain trading partners.
Used to integrate the multi-tier or n-Tier supplychain, including manufacturers, distributors andretailers.
CPFRs objective is to exchange selected internalinformation to provide for a reliable, longer termfuture views of demand in the supply chain.
CPFR uses a cyclic and iterative approach to
derive consensus forecasts.
5/21/2018 Forecasting CPIM
91/100
4-91
Web-Based Forecasting:Steps in CPFR
1. Creation of a front-end partnership
agreement
2. Joint business planning
3. Development of demand forecasts
4. Sharing forecasts
5. Inventory replenishment
Correlational Forecasting
5/21/2018 Forecasting CPIM
92/100
4-92
g
Assumes an Outcome is Dependent anExisting Relationship Between theDemand Variable and Some otherIndependent Variable(s) Demand Variable is Dependent Variable Other Related Variables are Independent
Variables Generally Expressed as a Multiple Linear
Regression Model Y =
+
X1+ X2+ X2+ . . . nXn+ i
Simple Linear Regression Model
5/21/2018 Forecasting CPIM
93/100
4-93
Simple Linear Regression Model
Yt= a + bx
0 1 2 3 4 5 x (Time)
YThe simple linear regression
model seeks to fit a line
through various data over
time
Is the linear regression model
a
- Ytis the regressed forecast value or dependentvariable in the model
-a is the intercept value of the the regression line, and-b is similar to the slope of the regression line.
- However, since it is calculated with the variability ofthe data in mind, its formulation is not as straightforward as our usual notion of slope.
Si l Li R i F l f
5/21/2018 Forecasting CPIM
94/100
4-94
Simple Linear Regression Formulas forCalculatinga and b
a = y - b x
b =xy- n(y)(x)
x - n(x2 2
)
Simple Linear Regression Problem Data
5/21/2018 Forecasting CPIM
95/100
4-95
Simple Linear Regression Problem Data
Week Sales
1 150
2 157
3 1624 166
5 177
Question: Given the data below, what is the simple linearregression model that can be used to predict sales in future
weeks?
Answer: First, using the linear regression formulas, we
can compute a and b
5/21/2018 Forecasting CPIM
96/100
4-96
Week Week*Week Sales Week*Sales
1 1 150 1502 4 157 314
3 9 162 486
4 16 166 664
5 25 177 8853 55 162.4 2499
Average Sum Average Sum
b = xy - n( y)(x)x - n( x
= 2499 - 5(162.4)(3) =
a = y - bx = 162.4 - (6.3)(3) =
2 2
) ( )55 5 96310
6.3
143.5
can compute a and b
Y = 143 5 + 6 3xThe resulting regression model
97
5/21/2018 Forecasting CPIM
97/100
4-97
Yt= 143.5 + 6.3x
180
Perio
d
135
140145
150
155
160165
170
175
1 2 3 4 5
Sales Sales
Forecast
is:
Now if we plot the regression generated forecasts against the
actual sales we obtain the following chart:
Statistical Assumptions of Multiple LinearR i
5/21/2018 Forecasting CPIM
98/100
4-98
Regression
The Error Term (the residual i) isNormally Distributed
There is no Serial Correlation Among
Error Terms Magnitude of the Error Term is
Independent of the Size of Any of theIndependent Variables - Xi
Assumptions Can be Tested ThroughAnalyses of the Residuals - i
Major Statistical Problems of Multiple
5/21/2018 Forecasting CPIM
99/100
4-99
Major Statistical Problems of MultipleLinear Regression
Multicolinarity
Use of Time-Lagged IndependentVariables
Both of These Problems Result in Modelswith Potentially Valid Predictions, but theReliability of the Coefficients isQuestionable
Demand Management
5/21/2018 Forecasting CPIM
100/100
gThe End
Processing,Influencing, &Anticipating
Demand
Store SellMake
top related