Why Moving Averages Fail on Lumpy Demand
This lesson is part of the course
In this lesson
You forecast a lumpy part with a 3-month moving average, measure its absolute errors and discover that a forecast of zero every month scores a better MAE – the trap that makes intermittent demand special.
Buy the course once for $9.99 to unlock every lesson and lab, the final exam and the certificate – lifetime access, no subscription.
Lessons in this course
- Demand With Zeros: Demand Months, Average Interval and Demand Size Locked
- Smooth, Erratic, Intermittent or Lumpy: Classifying Parts by ADI and CV² Locked
- Why Moving Averages Fail on Lumpy Demand Locked
- Croston's Method: Smoothing Size and Interval Separately Locked
- Measuring Forecast Error Without the MAPE Trap Locked
- Service-Level Stock From a Probability Table: Min and Max Levels Locked
- Criticality × Value: A Stocking Policy for Every Cell of the Matrix Locked
- Dead Stock & the Spare Parts Dashboard Locked