AI-based container usage optimization tool

Amine Kerkeni walks you through a system that demonstrates an agent that learns to pack boxes efficiently in containers while respecting multiple physical constraints. The reinforcement learning system demonstrates an agent that learns to pack boxes of different sizes efficiently in containers while respecting multiple operational constraints, for example, preventing items from overlapping, the need for physical support, weight distribution, etc. The agent is trained using reinforcement learning to minimize the wasted space. Without any human knowledge, the agent achieves superhuman performance and outperforms commercial optimization software. The system was trained on an Intel multicore system that helped to parallelize simulations and generate data for the agent to accelerate the learning process.