Theory and Methodology Super-eciency and DEA sensitivity analysis
نویسنده
چکیده
This paper discusses and reviews the use of super-eciency approach in data envelopment analysis (DEA) sensitivity analyses. It is shown that super-eciency score can be decomposed into two data perturbation components of a particular test frontier decision making unit (DMU) and the remaining DMUs. As a result, DEA sensitivity analysis can be done in (1) a general situation where data for a test DMU and data for the remaining DMUs are allowed to vary simultaneously and unequally and (2) the worst-case scenario where the eciency of the test DMU is deteriorating while the eciencies of the other DMUs are improving. The sensitivity analysis approach developed in this paper can be applied to DMUs on the entire frontier and to all basic DEA models. Necessary and sucient conditions for preserving a DMUÕs eciency classi®cation are developed when various data changes are applied to all DMUs. Possible infeasibility of super-eciency DEA models is only associated with extreme-ecient DMUs and indicates eciency stability to data perturbations in all DMUs. Ó 2001 Elsevier Science B.V. All rights reserved.
منابع مشابه
Prioritization method for non-extreme ecient unitsin data envelopment analysis
Super eciency data envelopment analysis(DEA) model can be used in ranking the performanceof ecient decision making units(DMUs). In DEA, non-extreme ecient unitshave a super eciency score one and the existing super eciency DEA models do notprovide a complete ranking about these units. In this paper, we will propose a methodfor ranking the performance of the extreme and non-extreme ecient units.
متن کاملSENSITIVITY ANALYSIS OF EFFICIENT AND INEFFICIENT UNITS IN INTEGER-VALUED DATA ENVELOPMENT ANALYSIS
One of the issues in Data Envelopment Analysis (DEA) is sensitivity and stability region of the speci c decision making unit (DMU), included ecient and inecient DMUs. In sensitivity analysis of ecient DMUs,the largest region should be found namely stability region thatdata variations are only for ecient DMU under evaluation and the data for the remainingDMUs are assumed xed. Also ecient DMU u...
متن کاملA Chance-Constrained DEA model with random input and output data:Considering maintenance groups of Iranian Aluminum Company
In this paper, we use an input oriented chance-constrained DEA model withrandom inputs and outputs. A super-eciency model with chance constraintsis used for ranking. However, for convenience in calculations a non-linear deterministicequivalent model is obtained to solve the models. The non-linearmodel is converted into a model with quadratic constraints to solve the nonlineardeterministic model...
متن کاملUsnig LR-Fuzzy Numbers Data to Measure the Eciency and the Malmquist Productivity Index in Data Envelopment Analysis , and Its Application in Insurance Organizations
In many real applications, the data of production processes can't be precisely measured.We develop some fuzzy versions of the classical DEA models (in particular, the CCRmodel) by using some ranking methods based on the comparison of cuts. Our approachescan be seen as an extension of the DEA methodology. The provides users and practitionerswith models which represent some real life processes mo...
متن کاملData Envelopment Analysis with Sensitive Analysis and Super-efficiency in Indian Banking Sector
Data envelopment analysis (DEA) is non-parametric linear programming (LP) based technique for estimating the relative efficiency of different decision making units (DMUs) assessing the homogeneous type of multiple-inputs and multiple-outputs. The procedure does not require a priori knowledge of weights, while the main concern of this non-parametric technique is to estimate the optimal weights o...
متن کامل