Intermittency and Obsolescence: a Croston Method With Linear Decay

TL;DR

This paper introduces a new unbiased Croston-style forecasting method called Linear-Exponential Smoothing, which decays linearly to zero in finite time and outperforms existing methods in handling demand obsolescence.

Contribution

The paper proposes a novel unbiased Croston variant that decays linearly, providing a finite-time obsolescence response and improved performance over existing methods.

Findings

Linear-Exponential Smoothing is unbiased and decays linearly to zero.

It is asymptotically the best for handling obsolescence.

Experimental results show strong performance in practical scenarios.

Abstract

Only two Croston-style forecasting methods are currently known for handling stochastic intermittent demand with possible demand obsolescence: TSB and HES, both shown to be unbiased. When an item becomes obsolescent then TSB's forecasts decay exponentially, while HES's decay hyperbolically. We describe a third variant called Linear-Exponential Smoothing that is also unbiased, decays linearly to zero in a finite time, is asymptotically the best variant for handling obsolescence, and performs well in experiments.