Deep reinforcement learning (DRL) has proven to be an effective, general-purpose technology develop ‘good’ replenishment policies in inventory management. We show how transfer from existing, well-performing heuristics may stabilize the training process and improve performance of DRL control. While idea is general, we specifically implement potential-based reward shaping a deep Q-network algorit...

In multi-echelon distribution strategy freight is delivered to customers via intermediate depots. Rather than using direct shipments, this strategy is an increasingly popular one in urban logistics. This is primarily to alleviate the environmental (e.g., energy usage and congestion) and social (e.g., traffic-related air pollution, accidents and noise) consequences of logistics operations. This ...

Quick commerce has recently become the most important issue in logistics industry due to intensifying competition for a faster delivery. A fulfillment service that integrates various processes emerged conduct delivery as quickly possible. With need facilities can serve near customer, Micro Fulfillment Center (MFC) is suggested solution. MFC refers small center located city and functions an adva...