Correlation between safety assessments in the driver-car interaction design process

Area of Science:

• Human-Computer Interaction
• Automotive Engineering
• Usability Engineering

Background:

• Modern vehicle functionality necessitates robust evaluation methods throughout driver-car interaction design.
• Early identification and resolution of safety issues are critical for automotive manufacturers.
• A key challenge is correlating formative evaluation methods used during development with summative methods applied post-launch.

Purpose of the Study:

• To investigate the correlation between efficiency metrics from formative and summative evaluations in driver-car interaction.
• To compare results from customer surveys with expert evaluations of vehicle systems.

Main Methods:

• Analysis of the J.D. Power and Associates APEAL survey (approx. 2000 customers).
• Expert evaluation study involving six evaluators assessing task completion time, NASA-Task Load Index (TLX), and Nielsen heuristics.
• Comparison of efficiency metrics across customer ratings and expert assessments for sound and navigation systems.

Main Results:

• High correlation observed between customer ratings and task completion time.
• Significant correlation found between customer ratings and NASA-Task Load Index (TLX).
• No correlation detected between Nielsen heuristics and customer ratings, task completion time, or TLX.

Conclusions:

• A high degree of consistency across methodologies supports developing a unified usability evaluation framework.
• Combining quantitative approaches with expert evaluations like task completion time is beneficial for driver-car interaction design.
• The findings enable more effective and predictive usability assessments in automotive development.