Toward a Methodology for Evaluating Instrument-Panel Controls

Abstract

Although there is a relatively large body of literature which documents evaluations of driver performance with instrument-panel (IP) controls, there does not exist a standard methodology which can be applied to obtain dynamic and objective performance data. Because of this, it is difficult to compare and integrate the findings from different evaluations. To address this problem, the G.M. Systems Engineering Human Factors Department has developed two implementations of a methodology which show promise — the “Video Method” and the “Automated Method.” Both implementations are identical with respect to the experimental tasks performed by the subjects and the types of objective data which can be obtained. The implementations differ, however, with respect to the method by which task completion times and errors are obtained. The Video Method relies upon frame-by-frame video analysis to obtain task completion times, while the Automated Method employs instrumented controls. A computer records driver control inputs, which are later reduced through the use of a custom software package, to obtain task completion times. Because of the differences in data acquisition techniques, both implementations exhibit unique strengths and weaknesses, and differ in their appropriateness for use with certain types of controls.