In this meetup, we used the infamous Beer Distribution Game as a starting point to exploring and redesigning supply chains. After discussing the dynamics underlying the game and what this means for supply chains, we analyzed a robust playing strategy and then took a look at how to train AI to play the game using reinforcement learning.
Download the slide deck Login to Download.
You can find the code for the simulation models and machine learning algorithms along with explanatory Jupyter notebooks on our GitHub repository.
If you missed the event, you can watch a live recording here:
Further Reading:
Understanding The Beer Distribution Game provides an in-depth analysis of the beer game and of how to find the right ordering strategy.
Training Artificial Intelligence to Play The Beer Distribution Game first introduces an Agent-based simulation of the Beer Game and then uses reinforcement learning to train the agents to play the game.