We explore using supervised learning with custom loss functions for multi-period inventory problems with feature-driven demand. This method directly considers feature information such as promotions and trends to make periodic order decisions, does not require distributional assumptions on...

Can there be side effects of improved forecast accuracy? In this study of the Belgian food retailer Colruyt Group, we show how adding explanatory variables (such as promotions, weather forecasts, national events, etc.) increases forecast accuracy compared to methods using only historical sales...

We propose a new policy architecture to apply deep reinforcement learning (DRL) with a continuous action representation. This contrasts most current DRL implementations that make use of a discrete action space. The latter are not scalable to large problems. To obtain feasible discrete actions,...

We study an inventory model with volume flexibility. A firm may replenish its inventory by using its base capacity at regular sourcing cost, after which additional supply is available at a premium. The structure of the optimal replenishment policy is known to be a modified base-stock policy, yet...

Artificial Intelligence (AI) applications are pervading logistics and supply chain management. The widespread availability of data, combined with sustained improvements in computing power, provide new opportunities to improve supply chain decision-making. Data may originate from digital...