On the Monte Carlo weights in multiple criteria decision analysis

Area of Science:

• Decision Sciences
• Operations Research
• Engineering Management

Background:

• Accurate criteria weights are essential for effective multiple-criteria decision making (MCDM/MCDA).
• Existing methods for weight derivation face challenges, particularly with a high number of criteria.
• The problem of setting correct criteria weights remains a significant issue in MCDM/MCDA.

Purpose of the Study:

• To propose a novel approach for determining criteria weights in MCDM/MCDA.
• To investigate the weight values required for a specific alternative to be ranked highest.
• To introduce concepts of central weights and multi-dominance for robust decision analysis.

Main Methods:

• Randomly drawing a large number of feasible weights using the Monte Carlo method.
• Employing predefined dominance relations for comparing and ranking alternatives based on generated cases.
• Estimating sample sizes for robust results and introducing central weights and multi-dominance measures.

Main Results:

• The study provides a new perspective on criteria weight determination by focusing on achieving the best rank for a given alternative.
• It establishes methods for estimating sample sizes to ensure the robustness and reliability of the results.
• Introduces central weights and multi-dominance as key concepts for analyzing and comparing alternatives.

Conclusions:

• The proposed method offers a robust approach to criteria weight determination in MCDM/MCDA, especially for complex scenarios.
• The concepts of central weights and multi-dominance provide valuable tools for decision-making and analysis.
• The application to wind turbine selection demonstrates the practical utility of the developed methodology.