Toward Human in the Loop Optimization Through Game-Based Experiments.

Optimizing operations on critical infrastructure networks is key to reducing the impact of disruptive events. In this paper, we aim to explore the potential of having humans and algorithms work together to address this difficult task, which we refer to as human in the loop optimization. Additionally, we aim to use game-based experiments to build and assess this potential in the context of the debris collection problem. To start, we developed an Excel-based prototype which allow humans to propose their own solutions for simplified problem contexts. Through two within-subjects experimental studies we assessed how humans optimized under various circumstances with this prototype compared to the computer on its own. The results suggest that human input may indeed help in solving complex problems but that guidance is needed. Future work will focus on running additional game-based experiments.