$\spadesuit$ SPADE $\spadesuit$ Split Peak Attention DEcomposition

TL;DR

SPADE is a neural forecasting model that effectively separates peak events from regular demand, reducing overreaction and improving forecast accuracy during and after peak periods in large retail datasets.

Contribution

The paper introduces SPADE, a novel neural network architecture with Peak Attention, specifically designed to handle demand spikes and improve forecast accuracy around peak events.

Findings

Overall PPE forecast accuracy improved by 4.5%.

30% improvement in forecasts after promotions and holidays.

Peak event forecast accuracy increased by 3.9%.

Abstract

Demand forecasting faces challenges induced by Peak Events (PEs) corresponding to special periods such as promotions and holidays. Peak events create significant spikes in demand followed by demand ramp down periods. Neural networks like MQCNN and MQT overreact to demand peaks by carrying over the elevated PE demand into subsequent Post-Peak-Event (PPE) periods, resulting in significantly over-biased forecasts. To tackle this challenge, we introduce a neural forecasting model called Split Peak Attention DEcomposition, SPADE. This model reduces the impact of PEs on subsequent forecasts by modeling forecasting as consisting of two separate tasks: one for PEs; and the other for the rest. Its architecture then uses masked convolution filters and a specialized Peak Attention module. We show SPADE's performance on a worldwide retail dataset with hundreds of millions of products. Our results reveal an overall PPE improvement of 4.5%, a 30% improvement for most affected forecasts after promotions and holidays, and an improvement in PE accuracy by 3.9%, relative to current production models.