Identifier

etd-09292016-190404

Degree

Master of Science in Industrial Engineering (MSIE)

Department

Mechanical Engineering

Document Type

Thesis

Abstract

In the petrochemical industry, schematic interfaces have been traditionally used as the main interface for console operators to monitor activities. There is limited research in this industry investigating alternative interface types to better support console operator’s decisions during alarm management. Furthermore, even less of that research includes eye-tracking as a measure for console operator situation awareness (SA). This research aimed to investigate an alternative interface, called a functional interface, in its level of support of console operator situation awareness, accuracy, subjective workload, and average response time. Additionally, eye-tracking was incorporated to explore its value as measure for situation awareness on interfaces in petrochemical control rooms. This research used a 2x3 factorial design to explore the effects of interface type (schematic vs. functional) and complexity level (easy, medium, and hard) in engineering students at Louisiana State University (LSU). The experiment involved three 30 minute simulations on either the schematic or the functional interface design of a main overview display that is typically seen in a refinery. The dependent variables included SA, subjective workload, accuracy, average response time, and eye fixation percentages for certain areas of interest (AOI). The mixed model analyses showed that there were no significant differences between interface types for any dependent variables except for the eye fixations in non-AOIs during non-alarm times. Participants spent significantly less time looking at non-AOIs during non-alarm times for the functional interface than the schematic. For complexity levels, there were no significant differences except for average response times. Average response times were were significantly higher for the medium level then the easy or hard levels. Also, the eye-tracking results showed that participants spent significantly less time in the intended AOIs and non-intended areas on the easy complexity level than the medium or hard. There was a significant positive correlation between the fixation percentages of the intended AOI during alarm times and SA1, indicating that eye-tracking was able to capture participants noticing process deviations during the simulation. Eye-tracking appears to be a good measure of SA1 among console operators. Overall, this research does not provide evidence that functional interfaces provide more support of console operator SA, workload, or performance.

Date

2016

Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

Harvey, Craig

Share

COinS