Assessing cogtool time prediction accuracy on control room displays
Document Type
Conference Proceeding
Publication Date
1-1-2014
Abstract
There is a need for a human performance modeling (HPM) tool which accurately estimates skilled task time for any interface design and requires minimal programming and cost. This study assessed the accuracy of an emerging HPM tool, CogTool, on two versions of a petrochemical industry user interface, DeltaV. Twenty-four students and 4 experienced operators' mean speeds to complete 3 tasks were compared to CogTool's prediction of identical tasks. CogTool predictions were expected to fall within 95% confidence interval of the means of the participants' speeds. CogTool's estimates fell outside the 95% CI in 9 of 12 cases. Of the 3 that were within range, 2 belonged to the experienced operator group for tasks performed on the grey interface, signifying CogTool was better at predicting operators' performance than the students' and only in interfaces already expected to perform well. Control room monitoring tasks place great demand on an operator's mental capacity; and sometimes require a user to work on multiple screens, commit information to memory, and make complex decisions. In this regard, this study recommends that the one user mental operator for "think time" (estimated as 1.2s), be revised in CogTool to accommodate this demand on the operator.
Publication Source (Journal or Book title)
IIE Annual Conference and Expo 2014
First Page
3325
Last Page
3334
Recommended Citation
Adio, O., Ikuma, L., Harvey, C., & Nahmens, I. (2014). Assessing cogtool time prediction accuracy on control room displays. IIE Annual Conference and Expo 2014, 3325-3334. Retrieved from https://repository.lsu.edu/mechanical_engineering_pubs/1088