State space ARIMA for supply-chain forecasting

John Boylan and I have been working lately on a paper, explaining the logic behind the ssarima() function from the smooth package. This paper has finally been accepted and published.

Also, based on a modified version of the ssarima() function, I have developed a SSARIMA module for Smoothie software, developed by DemandWorks company. Both the module and the ssarima() function work quite well on short seasonal time series, which is typical for many products in supply chain context.

The abstract of the paper:

ARIMA is seldom used in supply chains in practice. There are several reasons, not the least of which is the small sample size of available data, which restricts the usage of the model. Keeping in mind this restriction, we discuss in this paper a state-space ARIMA model with a single source of error and show how it can be efficiently used in the supply-chain context, especially in cases when only two seasonal cycles of data are available. We propose a new order selection algorithm for the model and compare its performance with the conventional ARIMA on real data. We show that the proposed model performs well in terms of both accuracy and computational time in comparison with other ARIMA implementations, which makes it efficient in the supply-chain context.

And here’s the postprint of the paper.

Leave a Reply