Peer groups that actually catch anomalies
Compare a merchant to the wrong 'normal' and the anomaly disappears. Measured: 91% of cases caught with peers built on identity, 24% with peers built on behavior.
Compare a merchant to the wrong 'normal' and the anomaly disappears. Measured: 91% of cases caught with peers built on identity, 24% with peers built on behavior.
SKU-level demand forecasts feeding a reorder/safety-stock simulator — the £ cost of forecast error made visible.