A Self-Learning Hyper-Heuristic Algorithm Based on a Genetic Algorithm: A Case Study on Prefabricated Modular Cabin...

Area of Science:

• Operations Research
• Artificial Intelligence
• Logistics Management

Background:

• Engineering optimization problems often involve complex constraints.
• Meta-heuristic algorithms can struggle with evaluating and repairing infeasible solutions.
• Logistics scheduling for prefabricated modular cabin units (PMCUs) in cruise ships presents multi-objective fuzzy challenges.

Purpose of the Study:

• Introduce a self-learning hyper-heuristic algorithm (GA-SLHH) for optimizing PMCUs logistics in cruise ships.
• Enhance the efficiency and stability of solving complex, multi-objective fuzzy logistics scheduling problems.
• Validate the algorithm's effectiveness in real-world manufacturing scenarios.

Main Methods:

• Developed a self-learning hyper-heuristic algorithm (GA-SLHH) using a genetic algorithm as the high-level strategy.
• Optimized low-level heuristics (LLHs) by incorporating a self-learning strategy and classic scheduling rules.
• Conducted multiple sets of numerical experiments and validated with practical enterprise cases.

Main Results:

• The GA-SLHH demonstrated superior comprehensive optimization ability and stability compared to other methods.
• The algorithm effectively addresses the challenges of multi-objective fuzzy logistics collaborative scheduling.
• Practical case studies confirmed the algorithm's applicability in cruise ship manufacturing.

Conclusions:

• The GA-SLHH is a robust and efficient algorithm for complex logistics scheduling problems.
• The proposed method can significantly reduce transportation time, achieving up to 37% reduction in practical applications.
• GA-SLHH offers a viable solution for real-world decision-making in the cruise ship industry.