18. how to improve forecast accuracy with sap apo demand planning 26125209

© SAP 2007-2008 / 0

How to Improve Forecast Accuracy with SAP APO Demand Planning

Dr. Gerald HeisigSAP AG

© SAP 2007-2008 / 2

In This Session, You’ll Get An Overview About ...

SAP APO Demand Planning (DP) basic architectureKey DP features

Promotion PlanningLifecycle ManagementSeasonal PlanningEtc.

The broad range of statistical forecasting methods in DP and guidelines for selecting the right method Integration of DP into other SAP solutions and SAP APO modules

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

© SAP 2007-2008 / 3

What We’ll Cover …

Introduction to Demand PlanningUnderstanding the basic structures and architecture Tapping into Demand Planning features Looking at different forecasting techniquesDemonstration: Demand PlanningWrap-up

© SAP 2007-2008 / 4

Pain Points in Demand Planning

Differences in planned demand and actual sales

Incorporation of all necessary demand information like promotions, product lifecycles, or other events in your demand plan

Demand visibility and consistency across all your departments and users

Sophisticated statistical forecasting

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Actual Planned

© SAP 2007-2008 / 5

Implications for Your Demand Management

Bad forecast quality

Incomplete and inaccurate demand

High number of stock outs

High inventory levels

Slow response to changing market

Lack of information for right planning decisions

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 2 2 23 24

Actual Planned

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

© SAP 2007-2008 / 6

SAP APO Demand Planning (DP)

Calculates future demand as accurate as possible

Comprehensive forecasting toolset Statistical forecasting with causal and time-series methodsAutomatic outlier detection availableHighly configurable planning books with macro functionalitySupporting aggregation/ disaggregation logicLifecycle Planning Plan promotions separately from the rest of your forecastOffline PlanningSeasonal Planning Collaborative Demand Planning

Improved forecast qualityOne tool for power and business userConsolidated demand plan (different regions, countries, departments, … )

Key BenefitsFeatures

© SAP 2007-2008 / 7

DP Interactive Planning

© SAP 2007-2008 / 8


Introduction to Demand PlanningUnderstanding the basic structures and architectureTapping into Demand Planning features Looking at different forecasting techniquesDemonstration: Demand PlanningWrap-up

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

© SAP 2007-2008 / 9

Integration Between SAP SCM and SAP NetWeaver Business Intelligence (SAP NetWeaver BI)

SAP SCMSAP SCM

With internal SAP BI

APO Demand Planning

Demand History

External SAP BI

InfoCube

InfoCubeCentral data store for reporting and analyzing

Source systems include:SAP ERP ExcelNon-SAP systems

POS dataCost informationOrder and shipping data

Demand History

Demand History

Forecasting results

© SAP 2007-2008 / 10

Data in SAP Supply Chain Management (SAP SCM)

DP master Data: CVCs

TransactionalData SAP BI extraction

structures

SAP SCMSAP SCMSAP ERPSAP ERP

MasterData

TransactionalData

MasterData

ATP DP

PP/DS SNP

LC

CIF

SAP BI

LocationProductResourcePPM/PDS

ATP = Available-to-Promise, SAP BI = SAP NetWeaver BI, CVCs = Characteristic Value Combinations, CIF = Core Interface, DP = Demand Planning, LC = liveCache, SNP = Supply Network Planning, PP/DS = Production Planning and Detailed Scheduling, PPM = Production Process Model, PDS = Product Data Structure

© SAP 2007-2008 / 11

Demand Planning Master Data: Characteristic Value Combinations

Characteristics

Product

Location

Customer

Characteristicvalues

Prod01Prod02Prod03

DC01DC02DC03

Cust01Cust02Cust03

Characteristicvalue combinations

(CVCs)

Prod01, DC01, Cust01Prod01, DC01, Cust02Prod01, DC01, Cust03Prod01, DC02, Cust01Prod01, DC03, Cust02Prod01, DC03, Cust03… Planning

ObjectStructure

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

© SAP 2007-2008 / 12

Planning Object Structure and Planning Area

The planning object structure is an APO InfoCube saved in the database of the internal SAP NetWeaver® BI systemThe characteristic value combinations can be created automaticallyOptionally, for better performance you can create additional aggregatesThe planning object structure with the relevant characteristic value combinations (DP master data) is assigned to a planning area

© SAP 2007-2008 / 13

Planning Area

A planning area is the central data structure for saving planning data for Demand Planning and Supply Network Planning

Characteristics and key figures and their functions for planning are determined hereIt groups together the central parameters that define the scope of the planning activities

It also determines whether planning results are to be saved as orders or time series

Planning book

Planning area

Interactive Planning

Characteristics Key figures

Planning version


© SAP 2007-2008 / 14

Planning Book

A planning book is based on information or a subgroup of information from a planning areaIn the planning book, select the characteristics and key figures required for the demand planner’s individual tasksEach planning book can contain several views where you can store key figures for detailed analyses and planning tasksIn each view you must also determine the planning horizon and time buckets profileA planning area can have more than one planning book, but a planning book can be linked to only one planning area

Planning book

Planning area

Sales Rep Data View

Characteristics Key figures Planningversion


Sales ManagerData View

Key figuresKey figures

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

© SAP 2007-2008 / 15

SAP NetWeaver BI and Demand Planning Reporting

Reporting for Demand Planner and Sales RepsReporting for Demand Planner and Sales Reps

Regional ForecastReporting

Planned ActualDeviation

Top 10 DeviationsPlanned/Actual

ForecastAccuracy

SAP APOSAP APO

Demand PlanningDMI

DP delivers planning data through Data Mart Interface (DMI)SAP NetWeaver BI InfoCube for information consumers

Historical

Planned

© SAP 2007-2008 / 16



© SAP 2007-2008 / 17

Interactive Planning

FlexibilityFree definition of planning books and data viewsCreation of data groups (selections) and user-specific assignment Multi-level planning with full visibility (drill up/down)Supporting different aggregation/disaggregation logicData representation on different periodicities and horizonsText can be added to any cell (notes management)Copy and paste (within grid and from/to Microsoft Excel)Graphic with data manipulation possibilitiesUser-specific customization

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

© SAP 2007-2008 / 18

Interactive Planning (cont.)

Macros Enable any kind of calculationCan be started any time on any level

© SAP 2007-2008 / 19

Interactive Planning – Alerts

Alerts can be customized user specific Alerts are triggered during batch processing or interactive planning for:

Forecast errors exceeding borders defined by the userAny kind of check carried out by a macro

Alerts are communicated to the user by:

Visualization in the alert monitorMailSMS message

© SAP 2007-2008 / 20

Lifecycle Management

Lifecycle Planning simulates the launch, growth, maturity, and discontinuation phases of different productsMimics the sales curve that you expect the product to display during the following phases:

Launch and growth Discontinuation

Actuals for old product

Like Modeling

Forecast for new product

Lifecycle

Phase-in profile Phase-out profile

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

© SAP 2007-2008 / 21

Realignment

Generate new characteristic value combinations based on existing combinations The key figures for realignment can be selectedAn empty Excel file can be created with the structure of the required realignment steps to upload and execute realignment

© SAP 2007-2008 / 22

Promotion Planning

Attempts to predict the outcome of the effect of an event, e.g., an annual promotion or advertising campaignCharacteristics of promotion planning include:

Separation of base sales data from changes caused by the eventEvaluating the effect of the promotional spending

Past Future

Corrected forecast Forecast

History(with promotions)

Corrected forecast+ promotions

Promotions can be imported from SAP CRM Marketing Planner

© SAP 2007-2008 / 23

Cannibalization

You use cannibalization groups to model the impact of a promotion on sales of related products

Sales for special offer product

M07/03

M08/03

M09/03

M10/03

Time

Corrected forecast

Original forecast

M07/03

M08/03

M09/03

M10/03

Time

Original forecast

Sales for similar product

Corrected forecast

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

© SAP 2007-2008 / 24

Seasonal Planning

Freely definable seasons and planning years are introduced that can be flexibly assigned to characteristic combinations

2003 2004 2005 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan

Product A (Apparel) SRING '03 SUMMER '03 FALL '03 HOLIDAY '03 SRING '04 SUMMER '04 FALL '04 HOLIDAY '04

Product B (Apparel) SRING '03 SUMMER '03 FALL '03 HOLIDAY '03 SRING '04 SUMMER '04 FALL '04 HOLIDAY '04

Product 1 (Footwear) SEASON C '02 SEASON A '03 SEASON B '03 SEASON C '03 SEASON A '04 SEASON B '04

Product 2 (Footwear) SEASON C '02 SEASON A '03 SEASON B '03 SEASON C '03 SEASON A '04 SEASON B '04

Apparel Planning Year Footwear Planning Year

© SAP 2007-2008 / 25

ep = ex-post forecast

Today0

1

2

3

4

5

6

7

8

Tolerance range

= ep ± σ *1.25* MAD

Automatic Outlier Correction

© SAP 2007-2008 / 26

Batch Processing/Process Chains

A job scheduling tool for creation, scheduling, and monitoring complex job chains is offered

The SAP NetWeaver BI tool for process chains is implemented as a frameworkMost DP processes are enabled for use by this framework

Automatic parallelization is offered for most of the DP processes

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

© SAP 2007-2008 / 27

Offline Planning

1. Download data

2. Working on the file

3. Upload file data

© SAP 2007-2008 / 28

Collaborative Planning

Any planning book can be accessed through the Internet

© SAP 2007-2008 / 29

Duet™ Demand Planning enables sales and planners to utilize the full Microsoft Excel capabilities as an intuitive planning frontend for SAP SCM

Load data from APO Demand Planning Analyze and contribute to demand planUse MS Excel features like additional lines, columns, graphics, and formulaeSave local file, distribute (e.g., by mail) with possibility for a later data synchronization, work online or offline

Duet™ Demand Planning

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

© SAP 2007-2008 / 30

Consensus Demand Planning

ObjectiveCreate a demand plan by integrating all available informationCollaborative process to gain “one number” consensus from sales, marketing, operations

Combine various data: – Forecast– Promotions– Budgets, sales plans, etc.– Manual changes

© SAP 2007-2008 / 31



© SAP 2007-2008 / 32

Forecasting

Forecasting predicts future demand based on historical and judgmental dataForecasts can be created in various ways

Statistical methodsHuman judgmentCombination of above

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

© SAP 2007-2008 / 33

Use of Statistical Methods

Statistical methods can support the planning process but they cannot solve basic planning problemsPowerful forecasting software can calculate millions of forecasts on the lowest level of detail but this is not always the appropriate planning level

Nobody can control/check millions of forecastsWith millions of data sets everything happens (Murphy’s law)It is sometimes better to plan on a higher (controllable) level and break down the results into details using fixed rules

Demand Planning ≠ Forecasting

© SAP 2007-2008 / 34

Data Preparation

Statistical methods can only run on appropriate dataAdaptations may be necessary for:

Start of real historyNegative/zero valuesMissing valuesSpecial events (e.g., strike, promotions, …)Causal effects

© SAP 2007-2008 / 35

Different Demand Patterns

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

© SAP 2007-2008 / 36

ConstantExponential smoothingMoving averageWeighted moving average

TrendExponential smoothingLinear regression

Season (without trend)Exponential smoothing

Trend – SeasonExponential smoothingManual forecastingSeasonal linear regression

OthersCroston method (sporadic demand)History No forecastExternal forecast

Causal Analysis

Multiple Linear Regression (MLR)Influence variables

Climate (e.g., temperature)PriceAdvertisingDistribution...

AUTOMATED

PICK

BEST

Composite ForecastCombine different forecastsWeight each forecast (time independent or dynamic)

Statistical Forecasting Methods

© SAP 2007-2008 / 37

Selection of Forecasting Methods

It is not appropriate to use the same forecasting method for all itemsBasic classifications include:

Forecast/Planning – Horizon: short ↔ medium ↔ longLinear ↔ Nonlinear development of the trendUnivariate forecast ↔ Causal analysis (MLR)Product type: New, mature, sporadic Different parameter settings

The assignment can be based on:Logical reasons– Product classification (e.g., spare part, standard product)– Planning purpose/business requirementsPilot study– Grouping of products– Assignment of parameters Ex-post error measures, but should not be the only criteria

The assignment should be checked in regular intervals

© SAP 2007-2008 / 38

(Weighted) Moving Average

Moving Average

93,5

94

94,5

95

95,5

96

96,5

97

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Periods

DemandForecast

Period Demand Forecast (M)1 95,12 94,93 94,64 94,7 95,15 95,2 95,26 95,6 95,37 95,7 95,48 95,6 95,59 95,5 95,710 95,3 95,811 95,9 95,912 96,2 96,013 96,4 96,014 96,315 96,116 95,9 n = 7

(Weighted) Moving Average: Gt+1 = Σ (Wt-j+1) Vt-j+1 / nj=1

n

Only suitable for constant demand patterns (with no trend-like or season-like patterns)

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

© SAP 2007-2008 / 39

First-Order Exponential Smoothing

Exponential Smoothing

0

5

10

15

20

25

30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Periods

DemandForecast (alpha = 0.3)Forecast (alpha = 0.1)

Period Demand Forecast Forecast 1 202 22 20,0 20,03 18 20,6 20,24 23 19,8 20,05 19 20,8 20,36 17 20,8 20,37 20 19,6 20,08 24 19,7 20,09 23 21,0 20,410 18 21,6 20,611 16 20,5 20,412 23 19,2 19,913 22 20,3 20,214 17 20,8 20,415 20 19,7 20,116 21 19,8 20,1

α = 0.3 α = 0.1

Gt+1(t) = (1- α)Gt(t-1) + αVt for all t = 2,…..,n ; G1 = V1

Only suitable for constant demand patterns (with no trend-like or season-like patterns)

© SAP 2007-2008 / 40

Exponential Smoothing

In a trend, seasonal, or seasonal trend model

© SAP 2007-2008 / 41

Linear Regression

For demand patterns With trendWith trend + season

t y1 1048,352 1102,063 1155,774 1209,485 1263,196 1316,907 1370,608 1424,319 1478,02

10 1531,7311 1585,4412 1639,1513 1692,8614 1746,5715 1800,2816 1853,9917 1907,7018 1961,41

Linear Regression

800900

100011001200130014001500160017001800190020002100

0 2 4 6 8 10 12 14 16 18

Periods

y

( )( )

( )∑

∑

=

=

−

−−= n

ii

n

iii

tt

yyttb

1

2

11

tbyb 10 −=

yt = b0 + b1* tyt = b0 + b1* t + Smod, t

seasonal adjustment factor for each period within the season

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

© SAP 2007-2008 / 42

Croston Method

Useful for intermittent, erratic, or slow-moving demandExample: When demand is zero most of the time (say 2/3 of the time)Might be caused by:

Short forecasting intervals (e.g., daily)A handful of customers that order periodicallyAggregation of demand elsewhere (e.g., reorder points)

An intermittent Demand Series

0

0.5

1

1.5

2

2.5

3

3.5

1 14 27 40 53 66 79 92 105

118

131

144

157

170

183

196

209

222

235

248

261

274

287

300

313

326

339

352

365

378

391

Period

Dem

and

© SAP 2007-2008 / 43

(Wei

ghte

d) M

ovin

g A

vera

ge

Firs

t Ord

er E

xpon

entia

l Sm

.

Intermittent Demand

Exp

onen

tial S

m. (

Cro

ston

)

Phas

e-In

/Out

Sec

ond

Ord

er E

xpon

. Sm

. (H

olt)

Line

ar R

egre

ssio

n

Regular Demand

Life-cycle

Constant Trend Trend+ Season

Sea

sona

l Lin

ear R

egre

ssio

n

Firs

t Ord

er E

xpon

. Sm

. (W

inte

rs)

Choose Forecasting Model –Overview

Different univariate forecasting methods can be assigned based on regular and intermittent demand patternsAdditionally, life-cycle profiles can be added to simulate phase-in/out of products

© SAP 2007-2008 / 44

Automatic Model Selection

You can choose Automatic Model Selection if there is no knowledge of the patterns in the historical data

The historical data are checked for constant, trend, seasonal, and seasonal trend patternsThe forecasting model that corresponds most closely to the pattern detected is applied

If no regular pattern is detected, the system runs the forecast as if the data revealed a constant pattern

You can also restrict the Automatic Model Selection to:Test for trendTest for seasonTest for trend and seasonSeasonal model and test for trend Trend model and test for seasonal pattern

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

© SAP 2007-2008 / 45

Automatic Model Selection (cont.)

Procedure 1The relevant forecast parameters (alpha, beta, and gamma) are constantNo consideration of error measure

Procedure 2Tests for constant, trend, seasonal, and seasonal trend patterns, using all possible combinations for the alpha, beta, and gamma smoothing factors The model with the lowest error measure customized (e.g., Mean Absolute Deviation [MAD]) is chosen

Procedure 2 is more precise than Procedure 1, but it takes longer

© SAP 2007-2008 / 46

Multiple Linear Regression (MLR)

MLR can assess how the development of one (dependent) variable can be explained by several (independent) variables (and a constant value)For a causal analysis, MLR does the final calculation of the regression coefficientsThe input data for the MLR (i.e., the modeling of the causal effects) is the key issueTypical variables:

Trend Seasonality Climatic conditions (e.g., temperature, precipitation)Economy (e.g., GDP, inflation, unemployment rate)Product specific (e.g., price/costs, new model/version, marketing activities)Demography (e.g., population in age classes)Others (e.g., lifecycle, replacement demand, distribution)

Substantial experience is required for modeling causal effects!

© SAP 2007-2008 / 47

Composite Forecast

Combine different forecastsOwn defined model selection based on error measure Weight each forecast (time independent or dynamic)Enables the combination of different forecasts with a constant or time-dependent weightingThe weighting will, in general, be purely arbitrary

Forecast1

n

... Combine and

Reconcile

Combine and

Reconcile

Univariate

MLR

UnivariateResultMLR

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

© SAP 2007-2008 / 48

Forecasting and Forecast Errors

© SAP 2007-2008 / 49

Error Measures

Look at errors over timeCumulative measures summed or averaged over all data

Error Total (ET)Mean Percentage Error (MPE)Mean Absolute Percentage Error (MAPE)Mean Squared Error (MSE)Root Mean Squared Error (RMSE)

Smoothed measures reflects errors in the recent pastMean Absolute Deviation (MAD)

Measure Bias

© SAP 2007-2008 / 50

Error Measures (cont.)

Look at errors over timeCumulative measures summed or averaged over all data

Error Total (ET)Mean Percentage Error (MPE)Mean Absolute Percentage Error (MAPE)Mean Squared Error (MSE)Root Mean Squared Error (RMSE)

Smoothed measures reflects errors in the recent pastMean Absolute Deviation (MAD)

Measure errormagnitude

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

© SAP 2007-2008 / 51

Evaluation

The planning/forecasting process has to be reviewed permanently/in regular intervals

This can include the analysis of:KPIs (e.g., service level, out of stock)Financial data (e.g., turnover, profit)Promotions/advertising Special influences (e.g., strike)Causal effects

The results should be documented and archived

© SAP 2007-2008 / 52



© SAP 2007-2008 / 53

Demonstration: Demand Planning

1. DP Interactive Planning

2. DP Features and Forecasting Run

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

© SAP 2007-2008 / 54



© SAP 2007-2008 / 55

Resources

help.sap.comSAP Business Suite SAP Supply Chain Management SAP SCM 5.0 Application Help EN SAP Advanced Planning and Optimization (SAP APO) Demand Planning Demand Planning Process

Comparison of the Planning MethodsSAP Business Suite SAP Supply Chain Management SAP SCM 5.0 Application Help EN SAP Advanced Planning and Optimization (SAP APO) Demand Planning Technical Aspects of Demand Planning

SAP NotesSAP Note 832393 (Release Restrictions for SCM 5.0)SAP Note 576015 (Collective Consulting Note for Demand Planning)

© SAP 2007-2008 / 56

7 Key Points to Take Home

Demand Planning (DP) is mid- to long-term planning that will help you to create forecasts for your productsDemand Management is not just forecasting – data preparation is key to getting reasonable forecasts DP offers a broad range of features for demand management like promotion or lifecycle managementDP offers a broad range of statistical forecasting methods that are applicable for all kinds of demand patternsUsually, the same statistical forecasting method cannot be used for all products Quality of resulting forecasts should be evaluated regularlyDP is tightly integrated to SAP NetWeaver BI, SAP ERP, and other SAP APO modules like SNP, PP/DS, and GATP

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

© SAP 2007-2008 / 57

Your Turn!

How to contact me:Gerald Heisig

[email protected]

© SAP 2007-2008 / 58

Copyright 2007-2008 SAP AGAll Rights ReservedNo part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of SAP AG. The information contained herein may be changed without prior notice.Some software products marketed by SAP AG and its distributors contain proprietary software components of other software vendors.SAP, R/3, mySAP, mySAP.com, xApps, xApp, SAP NetWeaver, Duet, Business ByDesign, ByDesign, PartnerEdge and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of SAP AG in Germany and in several other countries all over the world. All other product and service names mentioned and associated logos displayed are the trademarks of their respective companies. Data contained in this document serves informational purposes only. National product specifications may vary.

The information in this document is proprietary to SAP. This document is a preliminary version and not subject to your license agreement or any other agreement with SAP. This document contains only intended strategies, developments, and functionalities of the SAP® product and is not intended to be binding upon SAP to any particular course of business, product strategy, and/or development. SAP assumes no responsibility for errors or omissions in this document. SAP does not warrant the accuracy or completeness of the information, text, graphics, links, or other items contained within this material. This document is provided without a warranty of any kind, either express or implied, including but not limited to the implied warranties of merchantability, fitness for a particular purpose, or non-infringement.SAP shall have no liability for damages of any kind including without limitation direct, special, indirect, or consequential damages that may result from the use of these materials. This limitation shall not apply in cases of intent or gross negligence.The statutory liability for personal injury and defective products is not affected. SAP has no control over the information that you may access through the use of hot links contained in these materials and does not endorse your use of third-party Web pages nor provide any warranty whatsoever relating to third-party Web pages.

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

_________________________________

Notes: ______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

______________________________________________________________________________

18. how to improve forecast accuracy with sap apo demand planning 26125209

Documents