Definition
The difference between forecast error as the organization reports it, weighted by volume, and forecast error weighted by the cost exposure each item-location actually carries.
The Measurement Gap
There is no metric that tells you how forecast accuracy is distributed across the consequence-weighted item population rather than the volume-weighted population, which means the engine can improve its score while service degrades in the tail where the expedite spend lives.
Formula
CWFAG = (E_c - E_v) / E_v, where E_v = sum(v_i * |f_i - a_i|) / sum(v_i * a_i) and E_c = sum(c_i * |f_i - a_i|) / sum(c_i * a_i)
Proposed formula, pending validation pilots.
At a Glance
Decision Supported
Whether continued investment in the reported accuracy metric will improve service, and which cohort should own the next forecasting improvement.