摘要:AbstractExplaining concepts of optimization or even relevant algorithms to outsiders, like pupils or contacts from the industry, is a difficult task. Based on the idea of an augmented reality sandbox explaining geoscience processes we developed an exhibition piece to visually demonstrate fundamental ideas of numerical algorithms in the field of optimization and optimal control. In this paper we give an overview of our motivation to build the sandbox and how we apply it in our outreach projects. We discuss, how elementary problems from optimization and optimal control (as finding the best route) can be posed by modeling a landscape from sand. The sandbox features an interface to try out own basic algorithms and allows a comparison to the included state-of-the-art NLP solver WORHP. Finally, we present some further ideas to use the sandbox, which might be included in the future.