A Scheduling Method for Maintenance Tasks of Damaged Equipment Based on Digital Twin and Robust Optimization

Area of Science:

• Engineering
• Operations Research
• Computer Science

Background:

• Traditional maintenance scheduling for damaged equipment struggles with adaptability to uncertain factors and emergency scenarios.
• Existing methods lack dynamic synchronization between physical assets and virtual models.
• There is a need for robust and efficient scheduling solutions for battle-damaged equipment.

Purpose of the Study:

• To propose a novel maintenance task scheduling method for battle-damaged equipment.
• To leverage digital twin (DT) technology for dynamic synchronization and robust optimization for anti-interference capabilities.
• To improve the adaptability and efficiency of maintenance scheduling in emergency situations.

Main Methods:

• Constructed a multi-objective optimization model to maximize equipment importance and minimize maintenance time.
• Utilized discrete particle swarm optimization (DPSO) algorithm integrated with digital twin (DT) concepts.
• Compared the proposed DT-DPSO method against DPSO and DT-NSGAII algorithms.

Main Results:

• The DT-DPSO method demonstrated superior performance in convergence speed, robustness, and dynamic reallocation speed compared to DPSO and DT-NSGAII.
• Convergence speed was 33.3% faster than DPSO and 20% faster than DT-NSGAII.
• Robustness improved by 16.3% over DPSO and 10.7% over DT-NSGAII, with dynamic reallocation speeds over 40% and 30% faster, respectively.

Conclusions:

• The proposed DT-DPSO method significantly enhances maintenance scheduling efficiency and anti-interference capabilities for battle-damaged equipment.
• This approach is well-suited for high-speed, stable, and interference-resistant maintenance scheduling requirements.
• Digital twin and robust optimization integration offers a powerful solution for complex equipment maintenance challenges.