Multi-level DEA for the construction of multi-dimensional indices

Area of Science:

• Operations Research
• Econometrics
• Sustainability Science

Background:

• Data Envelopment Analysis (DEA) is a performance evaluation tool for Decision Making Units.
• Existing DEA variations can incorporate multi-stage processes and composite indicator construction.
• There is a need for advanced methods to build robust multi-dimensional indices.

Purpose of the Study:

• To propose a novel two-stage Data Envelopment Analysis (DEA) model for constructing multi-dimensional indices.
• To integrate sub-indicators using a Benefit-of-the-Doubt (BOD) model.
• To apply the proposed method for evaluating the sustainability of EU-28 countries.

Main Methods:

• A two-stage DEA model is employed for calculating individual sub-indicators.
• The Benefit-of-the-Doubt (BOD) mathematical programming model is utilized for integrating sub-indicators.
• The methodology is applied to assess the sustainability performance of EU-28 countries.

Main Results:

• The proposed two-stage DEA approach provides a robust framework for multi-dimensional index construction.
• The integration of sub-indicators via the BOD model ensures a comprehensive final index.
• The application successfully calculated sustainability indicators for EU-28 countries.

Conclusions:

• The novel two-stage DEA model offers an effective approach for developing sophisticated composite indicators.
• This methodology enhances the ability to measure complex phenomena like national sustainability.
• The study demonstrates the practical utility of the proposed DEA variation in policy-relevant assessments.