dt bt 1 st m bt tt 1 smooth trend forecasts

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ffects • Smooth seasonality forecasts using γ by assuming m periods per season Forecast Ft+1 = (Bt + Tt )St+1 Ft+k = (Bt + kTt )St+k m m Algorithm 4: Exponential Smoothing with Trend & Seasonality • How to construct seasonality: • Smooth base forecast Bt ! ! • Dt Bt = ↵ + (1 St m ↵)(Bt + Tt 1) Smooth trend forecasts Tt ! Tt = (Bt Bt 1) + (1 ! • 1 Smooth seasonality forecasts St Dt St = + (1 Bt )St m )Tt 1 Measuring forecast accuracy • Mean Forecast Error (MFE or Bias): Measures average deviation of forecast from actuals. • Mean Absolute Deviation (MAD): Measures average absolute deviation of forecast from actuals. • Mean Absolute Percentage Error (MAPE): Measures absolute error as a percentage of the forecast. • Mean Squared Error (MSE): Measures variance of forecast error. Part 10: Pooling Strategies Some pooling we have seen • Capacity Pooling (economy of scale) ! ! ! ! • Process Pooling (flow design) Economy of Scale in Queueing Wq vs. n with load being constant 1.80 Waiting time in queue 1.35 Benefits of pooling or economies of scale in queueing! is huge 0.90 0.45 0.00 0.000 12.500 25.000 n 37.500 50.000 Inventory Pooling • Current operations: • • • Each sales representative has her own inventory to serve demand in her own territory. Inventory Territory 1 Inventory Territory 2 e.g., 3 territories, 3 stockpiles of inventory Inventory Territory 3 The pooling strategy: • • • A single inventory is used by several sales reps. (How to implement?) Inventory is automatically replenished at the pooled location as depleted by demand. e.g., 3 pooled territories, 1 stockpile of inventory Territory 1 Inventory Territory 2 Territory 3 Capacity Pooling class 1 class 2 class 3 For flexibility Lead Time Pooling Store Supplier Store Store Supplier Retail DC Store Probabilistic view • Joint distribution two random variables X and Y • coefficients of variation for X+Y 75 50 25 0 0 25 50 When to use pooling • Almost always better • But comes with some cost Pooling No Pooling...
View Full Document

This document was uploaded on 01/28/2014.

Ask a homework question - tutors are online