《DemandPlanningwithSAPAPO》由会员分享，可在线阅读，更多相关《DemandPlanningwithSAPAPO（46页珍藏版）》请在金锄头文库上搜索。
1、Demand-Planning-Demand-Planning-with-SAP-APOwith-SAP-APOAgenda1.Introduction to Demand Planning2.APO Demand Planning features 3.APO Forecasting 4.APO Demand Planning integration and master data 5.SAP APO Infos & References1Pain Points in Demand PlanningDo you have problems with Differences in planne
2、d demand and actual sales?Incorporating all necessary demand information like promotions, product life cycles or other events in your demand plan?Demand visibility and consistency across all your departments and users?Wrong forecasts? 2Implications for your Demand Management and therefore face Bad f
3、orecast qualityIncomplete and inaccurate demandHigh number of stock outsHigh inventory levelsSlow response to changing market Lack of information for right planning decisions3Do you therefore need a tool to shape demand considering promotions, events and the life cycle of products improve forecast a
4、ccuracy? achieve higher product availability? better sense and actively manage demand by considering various demand signals support effectively the synchronization of your supply and demand? improve the calculation of buffer or safety stocks4Synchronize supply with demand in your global supply chain
5、 by balancing push and pull network-planning processes and by handling replenishment and production based on actual demand. nAPO DP: Improves the forecast quality and planning accuracynAPO SNP: Improves visibility across your global supply chain and lowers inventorynAPO PP/DS: Supports you in creati
6、ng optimized production plansnAPO gATP: Allows state-of-the-art sales order confirmation planning processesnAPO TP/VS: Optimizes transportation loads and minimizes transportation costs nAPO Alert Monitor: Powerful exception message system integrated in all APO planning modulesGlobal Available-to-Pro
7、miseDemand PlanningSupply Network PlanningProduction Planning & Detailed SchedulingAdvanced Planning & OptimizationAlert MonitorTransportation Planning & Vehicle SchedulingDemand Planning is part of APO5Agenda1.Introduction to Demand Planning2.APO Demand Planning features 3.APO Forecasting 4.APO Dem
8、and Planning integration and master data 5.SAP APO Infos & References6APO DP (Demand Planning)nComprehensive forecasting tool-set nStatistical forecasting with causal and time-series methodsnSelection of best-fit modelnAutomatic outlier detection availablenHighly configurable planning books with mac
9、ro functionalitynSupporting aggregation / disaggregation logicnLifecycle planning nPlan promotions separately from the rest of your forecastnTrack & monitor forecast accuracynSeasonal planning nCollaborative demand planningnImproved forecast qualitynOne tool for power and business usernConsolidated
10、demand plan (different regions, countries, departments, )Key BenefitsFeaturesCalculate and determine future demand to improve demand quality and accuracy7Typical DP Planning CycleSense & RespondCompare and evaluate different scenariosImport relevant master and transactional data (e.g. historical dat
11、a)Generate forecast & apply life cycle managementReview final demand planPlan promotions and/or other eventsRelease final demand to APO SNP or SAP ERPSimulate 134567Analyze and prepare data2Collaborate 8Interactive PlanningnText can be added to any cell (notes management)nCopy and paste (within grid
12、 and from/to Microsoft Excel)nUser-specific customizationnFree definition of planning books and data viewsnCreation of data groups (selections) and user-specific assignment nMulti-level planning with full visibility (drill up/down)nSupporting different aggregation/disaggregation logicnData represent
13、ation on different periodicities and horizons9nUse graphs and charts to represent data in visually perspicuous way nGraphs with data manipulation possibilitiesnUse high flexibility in customizing and capability of displaying more chart types Interactive Planning - SAP Chart Engine10MacrosnUse advanc
14、ed macros to perform complex calculations quickly and easily.nAPO system comes with examples of the use of macro functions and operators nMacros are executed either directly by the user in interactive planning or automatically at a predefined point in time during a background job. 11Alert Management
15、Alerts are communicated to the user by:nVisualization in the alert monitornMailnSMS messageAlerts can be customized user specific Alerts are triggered during planning run or interactive planning for:nForecast errors exceeding borders defined by the usernAny kind of check carried out by a macro12Life
16、cycle ManagementActuals for old productLike ModelingForecast for new productLifecyclePhase-in profile Phase-out profileLifecycle Planning simulates the launch, growth, maturity, and discontinuation phases of different productsMimics the sales curve that you expect the product to display during the f
17、ollowing phases:nLaunch and growth nDiscontinuation13Promotion PlanningnAttempts to predict the outcome of the effect of an event, e.g., monthly promotions or advertising campaign nEnables separation of base sales data from changes caused by the eventTools:nMultilevel promotionsnCannibalizationnPost
18、 promotion analysis pattern identification & repetitionnCollaboration through the Internet with trading partnernIntegration of campaigns created in SAP CRM to support Trade Promotion ManagementPast FutureCorrected forecastForecastHistory(with promotions)Corrected forecast+ promotions14Cannibalizatio
19、nYou use cannibalization groups to model the impact of a promotion on sales of related products Sales for special offer productM07/03M08/03M09/03M10/03TimeCorrected forecastOriginal forecastM07/03M08/03M09/03M10/03TimeOriginal forecastSales for similar productCorrected forecast15 Seasonal PlanningFr
20、eely definable seasons and planning years are introduced that can be flexibly assigned to characteristic combinations16ObjectivenCreate a demand plan by integrating all available informationnCollaborative process to gain “one number” consensus from sales, marketing, operationsn Combine various data:
21、 n Forecastn Promotionsn Budgets, sales plans, etc.n Manual changesConsensus Demand Planning17Duet Demand PlanningDuet Demand Planning enables sales and planners to utilize the full Microsoft Excel capabilities as a intuitive planning front end for mySAP SCMnFaster and accurate plans aligned to cust
22、omer demand requirementsnLeverages the full capabilities of Microsoft Excel (formulae, functions, graphics, etc.)nPresents an intuitive interface that requires minimum trainingnDelivers improved user acceptance and productivity through simple and timely access to relevant business informationnDesign
23、ed as a composite application based on Enterprise SOA 18Demand CombinationDemand Combination reconciles multiple demand streams in an automated manner using a flexible decision framework that represent business rules to get a more accurate projection of demandnAnalyses multiple demand streams in an
24、automated manner nCarries out reasonability checks to identify events or other demand componentsnAlerts user to discrepancies where automated correction is not possiblenDetermines the most appropriate demandstream for each time periodnPasses the reconciled demand to the downstream planning and execu
25、tion systemsnIs an enabler of a demand-driven supplynetwork e.g. for collaborative planning, dynamic S&OP, incorporation of the VMI order forecast into the demand plan19Agenda1.Introduction to Demand Planning2.APO Demand Planning features 3.APO Forecasting 4.APO Demand Planning integration and maste
26、r data 5.SAP APO Infos & References20nForecasting predicts future demand based on historical and judgmental datanForecasts can be created in various ways: nStatistical methodsnHuman judgementnCombination of aboveForecasting21ConstantnExponential smoothingnMoving averagenWeighted moving averageTrendn
27、Exponential smoothingnLinear RegressionSeason (without trend)nExponential smoothingTrend - SeasonnExponential smoothingnManual ForecastingnSeasonal linear regressionOthersnCroston (sporadic demand)nHistory nNo ForecastnExternal Forecast Causal AnalysisnMultiple Linear Regression (MLR)nInfluence Vari
28、ablesnClimate (e.g. Temperature)nPricenAdvertisingnDistributionn.AUTOMATEDPICKBESTComposite ForecastnOwn defined model selection based on error measure nCombine different forecastsnWeight each forecast (time independent or dynamic) Statistical Forecasting Methods22Finalized forecastUnivariate Foreca
29、sting MethodsCorrected current dataForecastCorrected forecastCurrent dataConstantnExponential smoothing 1st ordernExponential smoothing 1st order with a - a - adaptation nMoving averagenMoving weighted averagenCrostonTrendnExponential smoothing 1st order / HoltnExponential smoothing 2nd order nExpon
30、ential smoothing 2nd order with a - a - adaptation nLinear RegressionSeason (without trend)nExponential smoothing 1st order / WintersTrend - SeasonnExponential smoothing 1st ordernSeasonal linear regression OthersnHistory nManual Forecasting nNo Forecast nExternal Forecast / BAdI23(Weighted) Moving
31、AverageFirst Order Exponential Sm.Intermittent DemandExponential Sm. (Croston)Phase-In/OutSecond Order Expon. Sm.(Holt)Linear RegressionRegular DemandLife cycleConstantTrendTrend+ SeasonSeasonal Linear RegressionFirst Order Expon. Sm. (Winters) Choose Forecasting Model OverviewnDifferent univariate
32、forecasting methods can be assigned based on regular and intermittent demand patternsnAdditionally, life cycle profiles can be added to simulate phase-in/out of products.24Automated Pick the Best Model SelectionYou can choose Automatic Model Selection if there is no knowledge of the patterns in the
33、historical datanThe historical data are checked for constant, trend, seasonal, and seasonal trend patternsProcedure 1nThe relevant forecast parameters (alpha, beta, and gamma) are constantnNo consideration of error measurenThe forecasting model that corresponds most closely to the pattern detected i
34、s appliedProcedure 2nTests for constant, trend, seasonal, and seasonal trend patterns, using all possible combinations for the alpha, beta, and gamma smoothing factors nThe model with the lowest error measure customized (e.g., mean absolute deviation MAD) is chosen25Multiple Linear Regression (MLR)M
35、LR can assess how the development of one (dependent) variable can be explained by several (independent) variables For a causal analysis, MLR does the final calculation of the regression coefficientsTypical variables:n Trend n Seasonality n Climatic conditions (e.g., temperature, precipitation)n Econ
36、omy (e.g., GDP, inflation, unemployment rate)n Product specific (e.g., price/costs, new model/version, marketing activities)n Demography (e.g., population in age classes)n Others (e.g., life cycle, )26Multiple Linear Regression (MLR)Causal Analysis can assess how the development of one (dependent) v
37、ariable can be explained by several (independent) variables The input data for the MLR, i.e. the modelling of the causal effects is the key issueYi = b bo + b b1X1 + b b2X2. + b bn XnSalesHistoryMeanCoefficientsCausalFactor27Composite ForecastingnCombine different forecastsnOwn defined model selecti
38、on based on error measure nWeight each forecast (time independent or dynamic)nEnables the combination of different forecasts with a constant or time dependent weightingnThe weighting will, in general, be purely arbitraryForecast1n.Combine & ReconcileUnivariateMLRUnivariateResultMLR28Forecast Error M
39、easures for Univariate Models Look at errors over timeCumulative measures summed or averaged over all datanError Total (ET)nMean Percentage Error (MPE)nMean Absolute Percentage Error (MAPE)nMean Squared Error (MSE)nRoot Mean Squared Error (RMSE)Smoothed measures reflects errors in the recent pastnMe
40、an Absolute Deviation (MAD)Measure BiasMeasure errormagnitude29Measures of Fit for MLRInfluence of an independent variablenCoefficient, ElasticityA Measure Of Goodness-Of-FitnR2nAdjusted R2Correlation of the dependent to an independent variablent-statistic Autocorrelation (past periods are influenci
41、ng current periods)nDurbin-Watson nDurbin-h 30Agenda1.Introduction to Demand Planning2.APO Demand Planning features 3.APO Forecasting 4.APO Demand Planning integration and master data 5.SAP APO Infos & References31Agenda1.Introduction to Demand Planning2.APO Demand Planning features 3.APO Forecastin
42、g 4.APO Demand Planning integration and master data 5.SAP APO Infos & References35Some figures about SAP APO & Informationn First release SAP APO 1.1 (1998)n Current release: SAP APO 5.1 (8th APO release) part of SCM 2007 - in ramp-up since August 31st 2007 - mass shipment planned for Q1/2008 n Plan
43、ned next release: SAP APO 7.0 (Start ramp-up: Q4/2008)n Number of active APO customers: 1800n Service marketplace for SCM: http:/ http:/ figures about SAP APO36Molex - SAP Demand Planning (1)37Molex - SAP Demand Planning (2)38Conair - SAP Demand Planning39Oxford University Press - SAP Supply Chain M
44、anagement (1)40Oxford University Press - SAP Supply Chain Management (2)41Analyst Quotes“SAP Extends Its Vision for Supply Chain Management 1n “As implementations of SAP Advanced Planner and Optimizer (APO) have increased, APOs strengths (and challenges) have become apparent.” n “APO 5.0 is the most
45、 mature product in the mySAP SCM portfolio With APO 5.0, SAP has moved the investment towards innovation and new functionality” n “Functionally, SAP is a good contender when companies are considering using SAP APO to power their ATP processes”. n 1 White, et al., March 20, 2006 (Gartner)42SAP SCMCla
46、us BoschC Q & AThank you for your interest!“This presentation 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. Please note that this document is subject to change and may be changed by SAP at any time without notice. SAP assumes no responsibility for errors or omissions in thisdocument.”43结束结束
