Deep reinforcement learning to support dynamic decision-making in a transport network amid travel- and handling time uncertainty

This thesis explores efficient route planning techniques within the transport network of Gam Bakker. Specifically, the effectiveness of a deep reinforcement learning-based route planning approach is evaluated using a simulation study. Conceptually and experimentally deep reinforcement learning has demonstrated its efficacy in training agents that contribute to efficient route planning within the model of Gam Bakker’s proposed transport network.