Artificial intelligence-based human-centric decision support framework: an application to predictive maintenance in...

Area of Science:

• Operations Research
• Artificial Intelligence
• Industry 4.0

Background:

• Pandemic events necessitate operational reformulation for businesses, especially Small and Medium Enterprises (SMEs).
• Digital transformation and Industry 4.0 technologies are crucial for SMEs to minimize disruptions and enhance pandemic preparedness.
• Asset management digitization and predictive maintenance are key areas for operational optimization.

Purpose of the Study:

• To propose an AI-based, human-centric decision support framework for predictive maintenance in asset management.
• To facilitate prompt and informed decision-making for SMEs operating in pandemic environments.
• To enhance business sustainability through optimized asset predictive maintenance.

Main Methods:

• Development of an AI-based human-centric decision support framework.
• Introduction of an enhanced trust-based ensemble model to address imbalanced data in predictive maintenance.
• Incorporation of a human-in-the-loop mechanism to integrate subject matter expert knowledge.

Main Results:

• The proposed framework effectively addresses imbalanced data issues in predictive maintenance tasks.
• Evaluations using benchmark and real-world databases demonstrate the framework's effectiveness.
• A real-world case study achieved an accuracy rate of 82% in asset predictive maintenance.

Conclusions:

• The AI-based framework supports business sustainability for SMEs in pandemic environments.
• Human-centric approaches combined with AI are vital for effective predictive maintenance.
• The framework shows significant potential for optimizing asset management and minimizing disruptions during crises.