ACT-R based human digital twin to enhance operators' performance in process industries

Area of Science:

• Human-computer interaction
• Cognitive science
• Process control

Background:

• Operator performance assessment in process industries is challenging due to subjective methods and lack of cognitive behavior analysis.
• Existing methods fail to predict operator responses in novel or abnormal operational scenarios.
• Timely decision-making based on complex, time-varying data is crucial for safe and efficient industrial operations.

Purpose of the Study:

• To develop a human digital twin (HDT) capable of simulating control room operator behavior, including during abnormal situations.
• To utilize the Adaptive Control of Thought-Rational (ACT-R) cognitive architecture for creating a realistic operator simulation.
• To assess the HDT's ability to mimic human cognitive processes and performance in process control tasks.

Main Methods:

• Developed a human digital twin (HDT) based on the ACT-R cognitive architecture.
• Conducted 426 simulation trials to test the HDT's performance in disturbance rejection tasks.
• Validated the HDT by comparing its eye gaze behavior with that of 10 human subjects performing similar tasks.

Main Results:

• The HDT demonstrated comparable cognitive capabilities to human operators.
• HDT's simulated eye gaze behavior closely matched that of human subjects, even in abnormal situations.
• The developed HDT successfully simulated operator monitoring and intervention during process disturbances.

Conclusions:

• The human digital twin (HDT) provides a viable tool for simulating operator behavior in process industries.
• HDT can generate extensive data on human responses during abnormalities, aiding in identifying and correcting operator errors.
• The HDT has the potential to enhance operator training and real-time decision-making support.