The MPC elucidator: a case study in the design for human-automation interaction

Abstract

In this paper, we describe the design of a decision support system for operators of model-based predictive controllers (MPC). MPC is a form of advanced automatic control that is increasingly common in process operations due to its ability to control and optimize large sections of a process. A cognitive task analysis revealed that current operating displays, which rely on displaying tables of numeric information across several display pages, do not effectively support human operator monitoring, diagnosis, and control of MPC. This case study shows how we applied representation aiding and workspace management design principles to better support the human-automation interaction requirements of monitoring, understanding, and adjusting these complex, semi-autonomous; process controllers. We show how effective user interface design can significantly reduce the complexity of operating with advanced automation, and can lead to improved understanding of how the automation works.