强化学习
计算机科学
供应链
启发式
随机规划
供应链管理
数学优化
集合(抽象数据类型)
运筹学
生产(经济)
人工智能
数学
经济
政治学
法学
宏观经济学
程序设计语言
作者
Francesco Stranieri,Edoardo Fadda,Fabio Stella
标识
DOI:10.1016/j.ijpe.2023.109099
摘要
We introduce a novel heuristic for solving the supply chain inventory management problem in the case of two-echelon divergent supply chains. The proposed heuristic enhances the current state-of-the-art by combining deep reinforcement learning with multi-stage stochastic programming. In particular, we employ deep reinforcement learning to determine the number of production batches, while multi-stage stochastic programming is used for shipping decisions. To support further research, we make publicly available a software environment that simulates a wide range of two-echelon divergent supply chain settings, including different types of seasonal demands. We present a rich set of numerical experiments considering constraints on production and warehouse capacities under fixed and variable logistic costs. The results demonstrate that the proposed heuristic significantly and consistently outperforms pure deep reinforcement learning algorithms in terms of total costs. Moreover, it overcomes several inherent limitations of multi-stage stochastic programming models, thus further highlighting its potential advantages in solving the supply chain inventory management problem.
科研通智能强力驱动
Strongly Powered by AbleSci AI