Study of Human Tacit Knowledge Based on Electroencephalogram Signal Characteristics

Area of Science:

• Neuroscience
• Cognitive Science
• Industrial Engineering

Background:

• Operator proficiency in mineral grinding relies heavily on tacit knowledge, which is difficult to transfer through traditional means.
• Existing methods for assessing operator skill may not fully capture the cognitive aspects influenced by experience and training.

Purpose of the Study:

• To develop a method for detecting operator proficiency in mineral grinding using electroencephalogram (EEG) signals.
• To investigate the influence of tacit knowledge on functional cortical connectivity during industrial processes.
• To differentiate between trained (High Proficiency - Hps) and non-trained (Low Proficiency - Lps) operators based on brain network characteristics.

Main Methods:

• Combined electroencephalogram (EEG) signal analysis with mineral grinding industrial process data.
• Established functional brain networks (FBNs) using partial direct coherence and directed transfer function from EEG data.
• Employed multi-classifiers with graph-theoretic indexes of FBNs to distinguish between Hps and Lps.

Main Results:

• Functional brain networks of trained operators (Hps) demonstrated significantly better connectivity compared to non-trained operators (Lps) (p < 0.01).
• The classification accuracy for distinguishing Hps from Lps reached up to 94.2%.
• EEG features, combined with industrial operation and cognitive processes, effectively indicated operator proficiency.

Conclusions:

• Tacit knowledge significantly impacts functional cortical connectivity in mineral grinding operators.
• EEG-based brain network analysis is a viable method for objectively detecting operator proficiency.
• This approach offers a novel way to assess and potentially enhance skill transfer in industrial settings.