Concept of operations for data fusion visualization [chapter]

T McJunkin, R Boring, M McQueen, L Shunn, J Wright, D Gertman, O Linda, K McCarty, M Manic
2011 Advances in Safety, Reliability and Risk Management  
Data fusion for process control involves the presentation of synthesized sensor data in a manner that highlights the most important system states to an operator. The design of a data fusion interface must strike a balance between providing a process overview to the operator while still helping the operator pinpoint anomalies as needed. With the inclusion of a predictor system in the process control interface, additional design requirements must be considered, including the need to convey
more » ... inty regarding the prediction and to minimize nuisance alarms. This paper reviews these issues and establishes a design process for data fusion interfaces centered on creating a concept of operations as the basis for a design style guide. The statement has been made that the human operator should not have to do mental data fusion. A dashboard concept is appropriate until the process becomes large, and the contents expand to a large screen and then to a mural size display. Then the repetition and coverage of the operator scan of the display must be considered. The possibility of coercing the operator with cues about what has not scanned recently is an option. The system and process design must consider as many known attributes as possible. Detailed analyses can be used to prioritize and make available the information. Over time the process operation becomes mature, and risks become well qualified (if not quantified). A resilient architecture allows for refinement to take place with reasonably priced retrofits to the system, instead of living with the existing system or eventually cannibalizing it for newer system designs. An adaptive approach, as identified in this paper, is being examined as a possible step towards a resilient architecture. STRUCTURE OF AN INDUSTRY SYSTEM The conceptual HMI is designed around a plant for generating synthetic fuels. The plant is captured in a complicated model. The model framework was conceived to design and test scenarios for anomaly detection, and can handle situations ranging from malfunctions of system components to sophisticated attacks via cyber or infiltrated firmware (Chen and Abu-Nimeh, 2011). The model represents a not yet-built part of a larger
doi:10.1201/b11433-89 fatcat:5uh66fynybb7firemzfnmzlf7m