Data Envelopment Analysis Objective Research Articles

A Comparison of Ensemble and Dimensionality Reduction DEA Models Based on Entropy Criterion

Dimensionality reduction research in data envelopment analysis (DEA) has focused on subjective approaches to reduce dimensionality. Such approaches are less useful or attractive in practice because a subjective selection of variables introduces bias. A competing unbiased approach would be to use ensemble DEA scores. This paper illustrates that in addition to unbiased evaluations, the ensemble DEA scores result in unique rankings that have high entropy. Under restrictive assumptions, it is also shown that the ensemble DEA scores are normally distributed. Ensemble models do not require any new modifications to existing DEA objective functions or constraints, and when ensemble scores are normally distributed, returns-to-scale hypothesis testing can be carried out using traditional parametric statistical techniques.

Novel hydroelectricity data envelopment analysis model

In this paper a data envelopment analysis (DEA) was developed using a spreadsheet to inform policy decision making for a clean renewable hydroelectricity plant located in a natural park preserve. The DEA objective function was to make the minimum changes to the proposed policy rates for hydroelectricity generation subject to the known capabilities and risks calculated from a beta distribution. The DEA decision variables were the proposed changes to the policy rates for each month. The simplex linear programming technique was used to implement DEA in a spreadsheet. Screen shots were included to illustrate how to setup the DEA spreadsheet. The results of this paper should generalise to policy makers, practitioners, analysts and researchers in the public utility and specifically in the clean renewable hydroelectricity community.

Performance assessment and definition of improvement paths for microfinance institutions: an application to a network of village banks in Cameroon

AbstractThe objective of this paper is an assessment of the financial and social performance of a network of village banks in Cameroon and the provision of data‐driven guidance to managers helping them in their decision‐making process. An analysis framework in three stages is developed. First, a data envelopment analysis (DEA) is implemented for measuring efficiency, identifying best practices, and setting goals to less efficient microfinance institutions. Then, a DEA operating frontiers (DEA‐OF) model is designed to identify improvement paths setting short‐term goals toward their long‐term DEA objective and providing effective and time‐dependent guidance to managers in their efficiency improvement process. Finally, DEA and DEA‐OF results are translated into financial and nonfinancial indicators daily used by managers through three different and interrelated scorecards. Results of this third stage are used for identifying village banks ready for the new phase of development of the network aiming to fund community projects.

Modelling undesirable outputs in multiple objective data envelopment analysis

Recent empirical and conceptual work in data envelopment analysis (DEA) have emphasised its potential importance in highlighting the environmental performance of economic entities. Initial work in this emerging research area has focused on the separation of output factors into desirable and undesirable ones. In this paper, we describe recent developments in the modelling undesirable outputs. In particular, the modelling of undesirable outputs in the range adjusted measure (RAM) is investigated. We discuss some of the difficulties of RAM in assessing the environmental efficiency of decision-making units (DMUs) and develop a multiple objective DEA model to overcome these difficulties. The proposed multiple objective model is solved as a linear programming and its applicability as a mechanism for assessing environmental efficiency is demonstrated by evaluating the technical, ecological and process environmental quality efficiency scores of China’s provinces.

INTEGRATED METHODOLOGY FOR SUPPLIER SELECTION: THE CASE OF A SPHYGMOMANOMETER MANUFACTURER IN TAIWAN

Supplier selection is a critical multi-criterion decision-making activity for suc- cessful supply chain management. This study involved developing an integrated supplier selection methodology, which is constructed using analytic network process, data envelop- ment analysis, and multiple objective particle swarm optimization. The proposed integrated methodology can account for multiple supplier selection criteria and set boundaries on weight value for multiple objective data envelopment analysis inputs and outputs. To solve the data envelopment analysis model, a new algorithm based on multiple objective particle swarm optimization is introduced, which embeds with tabu list and group mechanisms, and then, it is found to be superior to the compared algorithms in solving performance on three test functions and the illustrative case. In addition, the proposed integrated method- ology was applied to a supplier selection problem of sphygmomanometer manufacturer in Taiwan to verify its applicability of decision-making process. The results show that the methodology can be implemented as an effective decision aid for supplier selection under multiple criteria with weight restrictions.

Multiple objective data envelopment analysis as applied to the Indian Pharmaceutical Industry

The change of Intellectual Property Protection (IPP) from a softer process patenting to a stronger product patenting in Indian Pharmaceutical Industry (IPI) is attracting many global drug majors to source their production from India, which is the fourth largest producer of pharmaceuticals in the world. In this paper, the interests of different stake holders like the buyers (multinational enterprises), who are searching for efficient partners and the vendors (Indian drug producers) that are competing for the contracts, are analysed for a suitable efficiency evaluation criterion. The primary objective of this paper is to study how various firms in the IPI with different business strategies, competing for the same opportunities can find suitable benchmarking peer groups to meet the challenges of a dynamic business environment using data envelopment analysis (DEA). A multiple objective DEA model that determines suitable peer groups for inefficient companies is discussed along with more traditional DEA models. The proposed model has the flexibility to include inputs like R&D expenditure and outputs like Exports that are not homogeneously distributed across the firms and address the interests of various stake holders like buyers and vendors simultaneously.

Subjective judgements and data envelopment analysis in site selection

Using subjective judgements as reflected in an Analytic Hierarchy Process (AHP) of site selection, a comparison is made to an objective Data Envelopment Analysis (DEA) selection procedure. Similarity in the results are discussed and conditions for expectations of parallel and different results are explored. A two step process of integrating subjective and objective decision making in site selection is posited.