Identifying the Robust Number of Intelligent Autonomous Vehicles in Container Terminals

The purpose of this research is to provide an improved Evolutionary Algorithm (EA) in combination with Monte Carlo Simulation (MCS) to identify the robust number of a new type of intelligent vehicles in container terminals. This type of vehicles, named Intelligent Autonomous Vehicles (IAVs), has been developed in a European project. This research extends our previous study on combining MCS with EAs. This paper has three main contributions: first, it proposes a dynamic strategy to adjust the number of samples used by MCS to improve the performance of the EA; second, it incorporates different robustness measures into the EA to produce different robust solutions depending on user requirements; and third, it investigates the relation between different robust solutions using statistical analyses to provide insights into what would be the most appropriate robust solutions for port operators. These contributions have been verified using empirical experiments.