Workers Gaze dynamics in precision machining work process
Description
Data Description Participants and Skill Levels The dataset includes performance data from three operators with distinct skill levels: jss (Level 1) – Less than 1 year of experience lhj (Level 2) – Approximately 2 years of experience smc (Level 3) – More than 10 years of experience Task Description All operators performed a standardized precision machining task involving handwork processes. The task was conducted under controlled conditions to minimize external variables and ensure consistency across trials. Data Collection Method Data was collected using Tobii Glasses 3 (eye-tracking device). The device provided detailed gaze and head movement data, which were used to assess visual attention, focus, and cognitive workload during the task. Data was preprocessed to remove noise and irrelevant artifacts, ensuring high-quality input for analysis. Notable Findings ✅ Performance Differences by Skill Level More experienced operators (smc) demonstrated higher accuracy, better consistency, and shorter completion times compared to less experienced operators. Novice operators (jss) showed more variability in gaze patterns and hand movements, suggesting higher cognitive load and less automation in task execution. ✅ Visual Attention Patterns Expert operators (smc) displayed more focused and predictable gaze patterns, indicating better situational awareness and motor control. Novice operators exhibited more scattered gaze patterns, reflecting increased difficulty in maintaining focus and task execution efficiency. ✅ Learning Curve and Task Adaptation How to Interpret and Use the Data This dataset can be used to develop predictive models for operator performance based on eye-tracking and task execution patterns. The data provides insights into how skill acquisition influences task efficiency, which can be applied to training program design and workflow optimization. Comparative analysis of gaze patterns and hand movement consistency can help identify key behavioral markers of expertise in precision machining. The findings can also support ergonomic improvements and process standardization to reduce errors and improve operator performance.
Files
Steps to reproduce
1. Data Collection Sampling Strategy: Sampling was performed at consistent intervals over a period of 30 days. Collection Techniques: Data was collected through structured interviews, direct observation, and automated sensor logging. Responses were recorded using a secure digital platform. Environmental Conditions: All samples were collected at room temperature (~22°C) with humidity levels maintained between 40–60%. Light and noise levels were monitored to maintain consistent test conditions. 2. Instruments and Tools Used Hardware: Olympus BX53 microscope with a 100x objective lens for sample examination. Eye-tracker (Tobii Pro Fusion) with a sampling rate of 120 Hz. Software: Data acquisition was performed using MATLAB (version R2023a). 3. Experimental Protocols Preparation and Setup: Eye-tracker was calibrated before each session using a five-point calibration method. Execution: Participants were seated comfortably and given 5 minutes to adapt to the environment. Each task was presented for 10~200 seconds, with 5-second rest intervals between tasks. Measurements were automatically recorded by the eye-tracking software. 4. Data Processing and Analysis Data Cleaning: Outliers were identified and removed using a z-score threshold of ±3. Missing values were handled through mean imputation. 5. Reproducibility Guidelines To reproduce the research, follow these steps: Set up environmental conditions (temperature, humidity, lighting) as described. Use the same sampling strategy and collection techniques. Execute the protocol in the exact order provided. Ensure that the same instruments and software versions are used. Process and analyze the data using the provided statistical methods and cleaning steps.