The emergence of cyber-physical production systems poses new challenges for designing the interface between production systems and the human-in-the-loop. In this study, we investigate how human operators interact with risks in a supply chain scenario. We varied the financial magnitude and the expected value of the decisions, the combination of two types of risk (risk in delivery amount and risk in timeliness), as well as three different task displays as within-subject factors. As explanatory user factors we studied the influence of Need for Achievement and the Attitude towards Risk-taking on the dependent variables task speed and accuracy. Results of the user study with 33 participants show that each of the investigated factors either influences decision speed, decision accuracy, or both. Consequently, the human-in-the-loop profits from adequate decision support systems that help to increase decision efficiency and effectiveness and reduce uncertainty and workload. The article concludes with a research agenda to support the human-in-the-loop in production systems.