The Impact of Instrumentation Location and Reliability on the Performance of Operators Using an Ecological Interface for Process Control

Ecological Interface Design (EID) is a recent philosophy for designing visual displays for human-machine interfaces that advocates displaying the higher-order relations and properties of a work domain so that adaptive operator problem solving is better supported for both normal and abnormal system conditions. Moreover, EID argues that this information should be represented through graphical visualizations. However, EID has assumed that the raw data needed to derive the higher-order information could be measured and that the graphics would not be susceptible to instrumentation failures. In fact, recent EID research indicates that current instrumentation technology and specification policies may adversely impact the implementation of such visual displays. The current research empirically evaluates the impact of such limitations on the robustness and effectiveness of an EID-based interface for a pasteurization microworld simulation representative of industrial process control systems. The results of this research were the following. First, the Abstraction Hierarchy was uniquely extended to indicate instrumentation requirements to support human-machine interfaces. Second, an original EID interface was designed and the iterative process of design was cataloged and characterized---a process often ignored in EID literature. Third, the empirical evaluation suggested that steady-state control performance of operators was not negatively impacted by sensor failures when using an EID interface or the conventional mimic display. Furthermore, displaying higher-order information appeared to support general diagnostic performance when dedicated sensors were used to derive this information. But as hypothesized, this diagnostic performance was compromised when higher-order information was derived from distally measured data. When compared to the mimic display that digitally displayed the measured data, detection and general diagnostic reasoning was supported by an EID interface with distally derived higher-order information even though diagnostic accuracy suffered. Finally, empirical investigation also indicated that there was not a superiority between the EID interface and the conventional interface in terms of average performance for diagnosis and control. This finding suggests that the EID interface might be improved upon by including the results of a formal Activity Analysis in the EID process.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Industrial Engineering

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2015

Author dc:creator

• Reising, Dal Vernon C.

Contributors dc:contributor

• Sanderson, Penelope M.

• Agriculture, Food Science and Technology

Language dc:language

eng

Identifier

(MiAaPQ)AAI9944982

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/87103