Post on 15-Oct-2014
Master Planning of Resources
Session 2
Forecasting Demand
What is a Forecast?
Forecast – An estimate of future demand. A forecast can be determined by mathematical means using historical data, it can be created subjectively by using estimates from informal sources, or it can represent a combination of both techniques.
Forecast Error – The difference between actual demand and forecast demand, stated as an absolute value or as a percentage.
Forecast Management – The process of making, checking, correcting, and using forecasts. It also includes determination of the forecast horizon.
Why Forecast? To plan for the future by reducing uncertainty To facilitate a company in taking control of operations.
Without forecast, it would be a chaos. To anticipate and manage change To increase communication and integration of planning
teams To anticipate inventory and capacity demands and
manage lead times To project costs of operations into budgeting processes To improve competitiveness and productivity through
decreased costs and improved delivery and responsiveness to customer needs
Areas Impacted by the Forecast
Investment decisions Capital equipment decisions Inventory planning Capacity planning Operations budgets Lead-time management
Forecast System Design Issues
Determine information that needs to be forecasted Assign responsibility for the forecast Set up forecast system parameters Select forecasting models and techniques Collect data Test models Record actual demand Report accuracy Determine root cause of variance Review forecasting system for improved performance
General Forecasting Techniques
Qualitative Techniques—based on intuitive or judgmental evaluation
Quantitative Techniques—based on computational projection of a numeric relationship
Qualitative Techniques
Expert opinion Market research Focus groups Historical analogy Delphi method Panel consensus
2-7
Quantitative Techniques
Moving average Exponential smoothing Regression analysis Adaptive smoothing Graphical methods Econometric modeling Life-cycle modeling
General Forecasting Data Methods
Intrinsic forecasting methods are based on historical patterns of the data itself from company data
Extrinsic forecasting methods are based on external patterns from information outside the company such as published data and data available from the Internet
Qualitative and quantitative forecasts may be generated based on intrinsic or extrinsic information.
Internal (Intrinsic) Factors
Product life-cycle management
Planned price changes
Changes in the sales force
Resource constraints Marketing and sales
promotion Advertising
2-10
External (Extrinsic) Factors
Competition New customers Plans of major
customers Government policies Regulatory concerns Economic conditions Environmental issues Weather conditions Global trends
2-11
Leading Indicators
Indicator(Causal Factor)
Housing starts
Birth rateHealth trends
Desire for Healthier lifestyle
Influences volume of
Building materialsHome furnishingsBaby productsMedical suppliesNutritional productsFitness products
2-12
Demand
A need for a particular product or component
2-13
Independent demand is demand for an item that is unrelated to the demand for other items. Independent demand items are saleable products or services that areadded to the master schedule.
Dependent demand can be calculated directly from the demand for other products.It is related to the bill of material structure.
Sources of Demand
Demand can come from many sources: Consumers Customers Referrers Dealers Distributors Interplant Service parts
Demand Characteristics
Internal Factors Product promotion Product substitution
External Factors Random fluctuation Seasonality Trend Economic cycle Changing customer
preferences and demands
Seasonality
0100
200
300
400
500600
700
800
J F M A M J J A S O N D
Sales in cases by month
Year 1Year 2
Seasonality Calculation
Measures seasonal variation of demand
Relates the average demand in a particular period to the average demand for all periods
The Seasonal Indexperiod average demand
average demand for all periods
Calculation of Seasonal IndexSales of Ice Cream
Month Year 1 Year 2 Total Calculation Index
January 10 12 22 22/409 0.05
February 10 12 22 22/409 0.05
March 10 12 22 22/409 0.05
April 50 55 105 105/409 0.26
May 150 160 310 310/409 0.76
June 400 420 820 820/409 2.00
July 600 620 1220 1220/409 2.98
August 700 730 1430 1430/409 3.49
September 350 360 710 710/409 1.74
October 100 105 205 205/409 0.50
November 10 12 22 22/409 0.05
December 10 12 22 22/409 0.05
Total 2400 2510 4910
Average 409.17 Round to 409
Seasonality Exercise
Economic Cycle
0
5
10
15
20
25
30
35
1 3 5 7 9 11 13 15 17 19
Quarter
Sales by Quarter
Pyramid Forecasting
Product/item volume(units)
Product family volume(units/dollars)
Totalbusinessvolume(dollars)
Rol
l Up
Fore
cast
Force Dow
n Adjustm
ent
Pyramid Forecasting
Pyramid Forecasting
Technique—Pyramid Forecasting Example
ROLL-UPProduct-level forecast
X1 units—8,200
price—$20.61 Family-level forecast
Family-adjusted forecastFORCE-DOWN
X1
X2
15,00013,045
15,00013,045
× 4,845 = 5,571 units
× 8,200 = 9,429 units
X2 units—4,845 price—$10.00 —units—13,045 Family avg price—$16.67 —units—15,000
Pyramid Forecasting Using Revenue
A B C D E F
X1 X2 Totals
units price units price Qty $
1 8,200 $20.61 4,845 $10.00 13,045 $217,452
2 1.15
3 9,429 $20.61 5,571 $10.00 15,000 $250,042
4 $250,070
2-25
2-26
Pyramid Forecasting Exercise
Historical DemandProduct A
Region 1 150Region 2 300Selling Price $4.50
Management has determined that next year’s demand will be $10,000 total.
CALCULATE the projected demand in units for products A and B in each region.
Product BRegion 1 300Region 2 450Selling Price $8.50
2-27
Pyramid Forecasting Exercise—Solution
Based upon historical demandA = 150 + 300 = 450 × $4.50 = $2,025B = 300 + 450 = 750 × $8.50 = $6,375
Total = $8,400
A: Region 1 = 1.19 × 150 = 178.5Region 2 = 1.19 × 300 = 357.0
B: Region 1 = 1.19 × 300 = 357.0Region 2 = 1.19 × 450 = 535.5
178.5 + 357.0 = 535.5 × $4.50 = $2,409.75357.0 + 535.5 = 892.5 × $8.50 = $7,586.25
= 1.19 (19% increase)$10,000
$8,400
$9,996.00
Moving Average Forecasting
Advantages A simple technique that is easy to calculate It can be used to filter out random variation Longer periods provide more smoothing
Limitations If a trend exists, it is hard to detect Moving averages lag trends
2-28
Moving Average ExerciseActual sales Next month’s
forecastvariation
Jan 100
Feb 500
Mar 1000
Apr 1500
May 2800
June 5100
Jul 6200
Aug 5700
Sep 3200
Oct 1200
Nov 500
Dec 100
Exponential Smoothing
Provides a routine method of updating item forecasts
Alpha is a weighting factor applied to the demand element
Works well for items with fairly constant demand
Is satisfactory for short-range forecasts Lags trends
2-31
New Forecast = ∝x Actual Demand + (1 - ) x Old Forecast∝
New Forecast = Old Forecast + x (Actual Demand – Old Forecast)∝
2-32
Smoothing Factor
Referred to as Alpha ( Determines the weight of historical
data on projection Sets responsiveness to changes in
demand Range 0 1
=
2n + 1
2-33
Smoothing Factor (cont.)
Determines how many periods of actual demand will influence forecast
1.00 = 1 period
0.50 = 3 periods
0.29 = 6 periods
0.15 = 12 periods
0.10 = 19 periods
0.1 Low weighting -most smoothing
0.9 High weighting - close to actual
Comparison of Exponential Smoothing Alpha Factors
Actual sales
2-34
2-35
Exponential Smoothing Examples
New forecast = Old forecast + smoothing factor ( (actual demand - old forecast)
Example: old forecast = 160, actual = 200, = 0.1
new forecast = 160 + (0.1 (200 - 160))
= 160 + (0.1 40) = 164
Example: old forecast = 160, actual = 200, = 0.8
new forecast = 160 + (0.8 (200 - 160))
= 160 + (0.8 40) = 192
Adapted from: Manufacturing for Survival, B.R. Williams, Addison Wesley, 1996
New Product Introduction
Every new product/service is a calculated risk.
Every new product/service has the potential to be the next
Blockbuster Lifesaver Money loser Disaster Liability nightmare.
2-36
2-37
Product Life Cycle
Introduction Growth Maturity Decline
Product Life Cycle Stages
Volume
Time
Focus Forecasting—Assumptions/Methods
Assumptions The most recent past is the best indicator of the
future One forecasting model is better than the others
Methods All forecasting models for all items forecasted will
be compared against recent sales history The model that achieves the closest fit will be
used to forecast this item this time Next time, a different model may be selected
2-38
Data Issues for Forecasting
Availability of data Consistency of data Amount of history required Forecast frequency Frequency of model reevaluation Cost and time issues Recording true demand Order date vs. ship date Product units vs. financial units Level of aggregation Customer partnering
2-39
Planning Horizon and Time Periods
Time Periods (week numbers)
Forecast Length
Short Mid Long
Weeks Months Quarters
1 2 3 4 5 6 7 8 9 1011 12 13 17 21 25 29 33 37 41 45 49 53 65 78 91 104
PlanningHorizon
2-40
Data Preparation and Collection Record sales data in same
periods as forecast data (daily, weekly, or monthly)
Monitor demand, not sales and/or shipments
Record the circumstances of exceptional demand
Record demand separately for unique customer groupings and market sectors
2-41
Dealing with Outliers
0
5
10
15
20
25
J F M A M J J A S O N D J F M A M J J A S O N D
2-42
50
55
Decomposition of Data Purify the data Adjust the data Take out the baseline and components Identify demand components
– Trend
– Seasonality
– Nonannual cycle
– Random error Measure the random error Project the series Recompose
2-43
Session 2 Review
You should now be able to Explain why forecasting is important Identify and describe general methods of
forecasting Identify factors influencing demand Describe considerations in using data for
forecasts Outline the process of data decomposition
2-44