ch7供應(yīng)鏈的需求預(yù)測(cè)-資料下載頁(yè)
2025-02-04 12:06本頁(yè)面
【正文】 38 = 14446 F5 = L1 + 4T1 = 13008 + (4)(1438) = 18760 30 需求趨勢(shì)和季節(jié)性需求修正后的指數(shù)平滑法 Trend and SeasonalityCorrected Exponential Smoothing （ Winter Model） ? 系統(tǒng)需求有需求水平、需求趨勢(shì)和季節(jié)性變動(dòng) 系統(tǒng)需求＝（需求水平＋需求趨勢(shì)） 季節(jié)性需求 ? Assume periodicity p ? Obtain initial estimates of level (L0), trend (T0), seasonal factors (S1,…, Sp) using procedure for static forecasting ? In period t, the forecast for future periods is given by: Ft+1 = (Lt+Tt)(St+1) and Ft+n = (Lt + nTt)St+n 31 Trend and SeasonalityCorrected Exponential Smoothing (continued) After observing demand for period t+1, revise estimates for level, trend, and seasonal factors as follows: Lt+1 = ? (Dt+1/St+1) + (1?)(Lt+Tt) Tt+1 = b(Lt+1 Lt) + (1b)Tt St+p+1 =? (Dt+1/Lt+1) + (1?)St+1 a 為需求水平的平滑系數(shù)， 0?1 b 為需求趨勢(shì)的平滑系數(shù)， 0 b 1 g為季節(jié)性需求的平滑系數(shù)， 0?1 32 Trend and SeasonalityCorrected Exponential Smoothing Example Example: Tahoe Salt data. Forecast demand for period 1 using Winter’s model. Initial estimates of level, trend, and seasonal factors are obtained as in the static forecasting case L0 = 18439 T0 = 524 S1=, S2=, S3=, S4= F1 = (L0 + T0)S1 = (18439+524)() = 8913 The observed demand for period 1 = D1 = 8000 Forecast error for period 1 = E1 = F1D1 = 8913 8000 = 913 Assume a = , b=, g=。 revise estimates for level and trend for period 1 and for seasonal factor for period 5 L1 = a(D1/S1)+(1a)(L0+T0) = )(8000/)+()(18439+524)=18769 T1 = b(L1L0)+(1b)T0 = ()(1876918439)+()(524) = 485 S5 = g(D1/L1)+(1g)S1 = ()(8000/18769)+()() = F2 = (L1+T1)S2 = (18769 + 485)() = 13093 33 預(yù)測(cè)誤差的度量 ? 一個(gè)好的預(yù)測(cè)方法應(yīng)該反映系統(tǒng)需求部分而不是隨機(jī)需求部分。隨機(jī)需求部分會(huì)以預(yù)測(cè)誤差的形式表現(xiàn)出來(lái)。 ? t期的預(yù)測(cè)誤差 Forecast error = Et = Ft Dt ? 平均方差 Mean squared error (MSE) 絕對(duì)離差 Absolute deviation = At = |Et| ? 平均絕對(duì)離差 Mean absolute deviation (MAD) When the random ponent is normally distributed = ?34 ? 平均絕對(duì)百分比誤差 Mean absolute percentage error (MAPE) ? 偏差 Bias 用來(lái)判斷預(yù)測(cè)方法持續(xù)低估或高估了需求 ? 跟蹤信號(hào) Tracking signal TSt = bias / MADt ? 應(yīng)該在 +6之間，否則 TS6，高估了需求； TS6，低估了需求，應(yīng)該更換 一種新的預(yù)測(cè)方法 1| | 100ntt tnEDM AP En???35 思考題 ? 預(yù)測(cè)在像戴爾公司這樣的按單生產(chǎn)的制造商的供應(yīng)鏈中起到什么作用？ ? 你認(rèn)為在巧克力和面粉的需求中系統(tǒng)成分和隨機(jī)成分各是什么？ 36 37 演講完畢，謝謝觀看！
環(huán)評(píng)公示相關(guān)推薦
鄂ICP備17016276號(hào)-1