Integrating Forecasts

A good way to improve the reliability of demand forecasts is to have multiple analysts generate forecasts independently and then combine their results. The problem here is figuring out how to integrate the forecasts in a meaningful way. One solution is simply to average them all together, but this can be risky. Just as aggregating across seasonal products with different peaks can cancel out the effects of seasonality, averaging independent forecasts can mask patterns that are evident in each forecast but that don't align precisely across forecasters. The better approach is to try to understand the reasoning behind each forecast and somehow combine the reasoning rather than just the numbers.

The obvious solution—just get ...

Get Supply Chains: A Manager's Guide now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.