A PageRank-based heuristic for the minimization of open stacks problem

Area of Science:

• Operations Research
• Industrial Engineering
• Computer Science

Background:

• The minimization of open stacks problem (MOSP) is crucial for optimizing physical space in manufacturing.
• Existing MOSP solutions struggle with large-scale problem instances.
• There is a need for efficient algorithms applicable to complex, large-scale manufacturing models.

Purpose of the Study:

• To develop a scalable heuristic for solving large instances of the minimization of open stacks problem.
• To evaluate the performance of a PageRank-based heuristic against state-of-the-art methods.
• To analyze the impact of instance size and graph density on heuristic performance.

Main Methods:

• A PageRank-based heuristic was developed to model and solve the MOSP on graphs.
• Computational experiments were conducted using literature data and new, larger datasets (up to 25x).
• The proposed heuristic was compared against existing state-of-the-art methods across 1330 instances.

Main Results:

• The PageRank-based heuristic demonstrated competitiveness in solution quality, achieving optimal results in several cases.
• The heuristic achieved significantly shorter run times compared to the fastest existing method.
• Solution quality differences were minimal for specific graph densities, favoring the faster heuristic.

Conclusions:

• The proposed PageRank-based heuristic is a scalable and efficient solution for large MOSP instances.
• The heuristic's performance is more sensitive to graph density than to instance size.
• This method offers a practical approach for optimizing production sequencing in manufacturing settings.