Aggregation (Composition) Schema For Eigenvector Scaling Of Criteria Priorities In Hierarchical Structures.

Area of Science:

• Decision Science
• Multi-attribute Decision Making
• Psychometrics

Background:

• Traditional models for scaling choice alternatives using paired comparisons and multiple attributes often fail to capture individual utility accurately.
• Both compensatory and noncompensatory aggregation models may lack consistency or appropriateness for subjective evaluations.
• Existing methods struggle with the complexity of simultaneous attribute evaluation.

Purpose of the Study:

• To review the eigenvector approach for scaling choice alternatives evaluated on a relative basis.
• To demonstrate the limitations of compensatory and noncompensatory models in representing individual utility.
• To propose an alternative subjective aggregation method for multi-attribute decision making.

Main Methods:

• Review of eigenvector scaling techniques applied to paired comparison data.
• Analysis of the consistency and appropriateness of compensatory and noncompensatory aggregation models.
• Development of a novel approach where individuals perform subjective aggregation across all attributes simultaneously.

Main Results:

• Eigenvector scaling is a viable method for prioritizing choice alternatives based on paired comparisons.
• Neither compensatory nor noncompensatory models consistently align with individual utility assessments.
• The proposed subjective aggregation method allows for simultaneous evaluation of multiple attributes.

Conclusions:

• The eigenvector approach offers a robust method for scaling choice alternatives.
• Simultaneous subjective aggregation across attributes provides a more accurate representation of individual utility.
• This alternative approach enhances decision-making models by integrating holistic attribute evaluation.