Predictive Model Selection for Forecasting Product Returns

Abstract

As awareness of environmental issues increases, the pressures from the public and policy makers have forced original equipment manufacturers (OEMs) to consider remanufacturing as the key product design option. In order to make the remanufacturing operations more profitable, forecasting product returns is critical due to the uncertainty in quantity and timing. This paper proposes a predictive model selection algorithm to deal with the uncertainty by identifying a better predictive model. Unlike other major approaches in literature such as distributed lag models or DLMs, the predictive model selection algorithm focuses on the predictive power over new or future returns and extends the set of candidate models. The case study of reusable bottles shows that the proposed algorithm can find a better predictive model than the DLM.