A Sensor-based Multi-modal Evaluation Framework for Optimizing Redesigned Interdisciplinary Human–Machine Interfaces by Webcam-based Eye-tracking in Intelligent Manufacturing

In this study, a sensor-based evaluation framework is developed to validate a redesigned interdisciplinary human-machine interface (HMI) by webcam-based eye-tracking in intelligent manufacturing systems.Traditional industrial HMIs often contain excessive and mixed information for different user roles, which reduces operational clarity and increases cognitive burden.To improve interface performance, a role-oriented redesign strategy is implemented on the basis of user interface/user experience design principles.A sensor-enhanced evaluation approach is adopted by integrating webcam-based eye-tracking measurement, behavioral performance data, and System Usability Scale assessment, where the eye-tracking sensor captures visual attention distribution during task execution.Experimental results showed that the redesigned interface produces a more concentrated visual attention pattern and reduces unnecessary visual scanning.The task error rate decreases from 66.7% in the original interface to 13.3% in the redesigned interface, while usability perception also improves after redesign.Overall, the results demonstrated that sensor-based visual attention measurement combined with quantitative usability assessment provides a practical and reproducible approach for evaluating HMI design optimization in industrial applications, highlighting its relevance to sensor-based sensing systems and intelligent human-centered interface technologies.