预览加载中，请您耐心等待几秒...
1/10
2/10
3/10
4/10
5/10
6/10
7/10
8/10
9/10
10/10
亲，该文档总共16页，到这已经超出免费预览范围，如果喜欢就直接下载吧～
如果您无法下载资料，请参考说明：
1、部分资料下载需要金币，请确保您的账户上有足够的金币
2、已购买过的文档，再次下载不重复扣费
3、资料包下载后请先用软件解压，在使用对应软件打开
ForecastingSupplyChainRequirements(2)ClassicTimeSeriesDecompositionClassictimeseriesdecompositionforecastingisbuiltonthephilosophythatahistoricalsalespatterncanbedecomposedintofourcategories:trend,seasonalvariation,cyclicalvariation,andresidual,orrandomvaviation.Classictimeseriesanalysiscombineseachtypeofsalesvariationinthefollowingway:
F=T*S*C*R
F=demandforecast
T=trendlevel
S=seasonalindex
C=cyclicalindex
R=residualindexInpractice,themodelisoftenreducedtoonlytrendandseasonalcomponents.
Thisisdonebecauseawell-specifiedmodelhasaresidualindexvalue(R)of1.0andthusdoesnotaffecttheforecast,andbecauseitisdifficultinmanycasestodecomposecyclicalvariationfromrandomvariation.
Treatingthecyclicalindex(C)asequalto1.0isbecausethemodelisusuallyupdatedwhennewdatabecomeavailable.ThemathematicalexpressionforalineartrendlineisT=a+bt,thecoefficientsarefoundby
∑Dt(t)-N(D)(t)
b=∑t2-Nt2
a=D-btN=thenumberofobservationsusedinthedevelopmentofthetrendline
Dt=theactualdemandintimeperiodt
D=averagedemandforNtimeperiods
t=averageoftoverNtimeperiodsSt=Dt/Tt
Ft=(Tt)(St-L)Example:
Amanufactureofyoungwomen’sclothinghadtomakepurchasequantitydecisionsandsetproductionandlogisticsschedulesbasedonforecastsofmarketsales.Fiveseasonsoftheyearwerespecifiedforplanningandpromotionalpurposes-summer,trans-season,fall,holiday,andspring.Salesdataforapproximatelytwoandone-halfyearswereobtained(SeeExcel).Aforecastwasneededfortwoseasonsaheadofthecurrentaccountingperiodtoensureadequatepurchasingandproductionleadtime.Whatisforcastfornextholidayseason?由于室内空调产品销售中显著的季节性特征，高压电机公司（Thehigh-voltElectricCompany）在预测季度销售量时面临很大的困难。下表是过去三年个季度销售数据。
1）找到最佳的线性趋势线
2）利用经典的时间序列分解法预测以后四个季度的销售量
去年二年前三年前MultipleregressionanalysisSpecialpredictionproblemsforlogisticsLumpydemandpatternsaredifficulttopredictaccuratelybymathematicalmethods,however,somesuggestionsonhowtotreatthemcanbeoffered:
1lookforobviousreasonsforthelumpinessandusethemtoproducetheforecast.Separatetheforecastingoflumpydemandproductsfromthoseshowingaregularpatternandusefor
内容提供者
2024-09-17
16金币
45KB
16页
实名认证
