Data Envelopment Analyses
Azam Pourhabib Yekta; Mahnaz Maghbouli
Abstract
Purpose: Data Envelopment Analysis (DEA) is a technique used to assess performance and measure the relative efficiency of Decision Making Units (DMUs) through linear programming. In most cases, DEA models evaluate inefficient units on the boundary of the production possibility set using reference points ...
Read More
Purpose: Data Envelopment Analysis (DEA) is a technique used to assess performance and measure the relative efficiency of Decision Making Units (DMUs) through linear programming. In most cases, DEA models evaluate inefficient units on the boundary of the production possibility set using reference points that are not Pareto efficient. Consequently, these models often yield zero weights for multipliers, failing to justify all sources of inefficiency. This paper aims to introduce a model that generates non-zero weights.Methodology: Weight restriction methods have primarily addressed the issue of non-realistic weights. We impose constraints on the weights in the proposed model to achieve our objectives.Findings: This paper presents a one-stage method based on the BCC model, incorporating weight restrictions, to evaluate the relative efficiency of decision-making units. The proposed model ensures non-zero weights and prevents dissimilarity between weights while maintaining feasibility. Notably, the proposed model does not require any prior information on weights or the classification of units, reducing the complexity of the problem.Originality/Value: To highlight the strength of the proposed method, the model is implemented on two case studies and compared with the results obtained from standard BCC models and those of Ramon and colleagues. The results indicate the superior performance of the proposed model.