Advanced visualization tools for training and trust in highly complex process control work domains.

The collaboration aims to build a long-term collaboration in the Human-Systems Interaction field. New knowledge and solutions in this area will help improve performance in complex systems whose performance and safety rely on effective joint human and auto matic control and decision-making. Cognitive Work Analysis (CWA) and Ecological Interface Design (EID) are frameworks for developing advanced graphical interfaces for complex systems such as chemical and nuclear plants. Such interfaces provide users with the information needed to monitor the system, make control actions and handle emergency situations effectively. The two potential directions for collaborative effort are both based on and expand on these frameworks. "Transparent automation" The introducti on of advanced automation technologies such as steady-state optimizers and model-based predictive controllers raises new challenges related to the joint performance of the human/technology/organization system. Both research and serious incidents in differ ent domains have shown that as automation technology grows more complex, both misuse and disuse are basic problems that arise in the human-automation partnership. Advanced human system interface techniques and visualization is one means by which the quali ty of the human-automation partnership can be influenced. "Training interfaces for complex processes" New types of displays and interaction principles can be made especially for use in advanced operator training tools. These would be more than mere replic a of the systems used in real time, and could expand the current use of the simulators from just training to developing operators' deeper process understanding by letting operators explore and observe the behavior of a simulated process without the constr aints that exist in the physical plant and its instrumentation. The emphasis in the EID/CWA-framework deep understanding and alternative problem-solving strategies represents a promising starting point for this.