Forecasting the returns in reusable containers’ closed-loop supply chains. A case in the LPG industry.

Resumen

In this paper we review the returns forecasting models described in the academic literature. Next, we build a model not requiring item-level information in the context of a case study conducted in Repsol GLP, the liquefied petroleum gases (LPG) division of Repsol group. Results are unexpected and put under scrutiny the estimates of the return rate and the return delay distribution obtained through this type of models when using a direct replacement policy. We think that the main cause for these results resides in the exchange of full for empty containers imposed by this policy; deliveries and returns are linked in time, and thus the hypothesis of unidirectional causality might not be respected.