Paper Title
DETERMINING THE EFFECTIVENESS OF RL IN TACKLING INVENTORY CONTROL COMPLEXITIES

Abstract
Abstract - Determining inventory control (IC) with uncertain demand becomes challenging when inventory analysis is done with multiple products and with different lead times including non-zero lead times. The objective of this paper is to verify the effectiveness of Deep Reinforcement learning (DRL) algorithms and their impact on IC complexities. Our contributions include quantifying the extent of the application of certain DRL algorithms in complex IC scenarios and identifying that more than one DRL algorithm can be effective. We also identify potential extensions involving more complexities and recommend specific future research directions and novel approaches to categorizing DRL algorithm. Keywords - Reinforcement Learning, Inventory Control Complexities, DRL Categorization