Basic Data Envelopment Analysis Research Articles

A compromise programming approach for cross efficiency measurement in basic two-stage network system

Data envelopment analysis (DEA) is a technique that allows each decision-making unit (DMU) to calculate efficiency using the most favorable weights for inputs and outputs. The resulting efficiency scores are incomparable and difficult to discriminate. Cross-efficiency is a concept for solving the problem of incomparability among the efficiencies of a set of DMUs that are calculated from different weights and is helpful for ranking. In basic DEA, several cross-efficiency evaluation methods with various secondary purposes have been presented, however, the internal mechanism of the DMUs in cross-efficiency evaluation is often neglected. In this study, we use the composition approach for a basic two-stage structure, so that the efficiencies of the divisions are estimated first, and then the overall efficiency of the system is obtained. To solve the linear bi-objective model in the combination approach, we consider a compromise programming problem to find the solution that is as close as possible to the ideal point. Finally, a secondary objective model is proposed to select a weight among the optimal weights of the compromise model, which also guarantees the efficiency of the optimal weight for the bi-objective problem. We use the optimal solution obtained from this model to evaluate cross-efficiency and rank DMUs for a series two-stage system. The proposed approach enables us to rank the DMUs by assessing the performance of each DMU and each of its separate divisions. The validity of the suggested network cross-efficiency evaluation is demonstrated by a numerical example from the literature.

Evaluation of efficiency and ranking of Indian hotels and restaurants: a bootstrap DEA approach

PurposeThe main purpose of this paper is to determine the bias-corrected efficiencies and rankings of the selected hotels and restaurants (H&Rs) in India.Design/methodology/approachThe data for the Indian H&R sector are collected from the Prowess database. The bootstrap data envelopment analysis (DEA) based on a constant return to scale (CRS), variable return to scale-input oriented (VRS-IP) and variable return to scale-output oriented (VRS-OP) are applied on H&Rs to obtain the bias-corrected efficiencies.FindingsIt is found that relative efficiencies using basic DEA methods of all the 45 H&Rs of India are overestimated. These efficiencies are corrected using bias correction through bootstrap DEA methods. The bounds for the efficiencies of each H&R are computed using all the adopted methods. All H&Rs are ranked using bias-corrected efficiencies, and the linear trend between ranks suggests that the H&Rs are ranked almost similarly by all the adopted methods.Practical implicationsTo improve efficiency, Indian H&R companies must rethink their personnel needs by enhancing their workforce management capabilities. The government needs to extend more support to this sector by introducing a liberal legislation framework and supporting infrastructure policies.Originality/valueThere is a paucity of studies on H&Rs in India. The current study focused on measuring bias-corrected efficiencies of the selected H&Rs of India. This study is one of the few initiatives to explore bias-corrected efficiencies extensively using the bootstrap DEA method.

Assessing the efficiency of airports by considering health and safety issues: A pilot study for Turkey

Purpose: Airport management is influenced by several related critical decisions on financial and technical operations. Optimum utilization of resources, including assets and personnel, is critical to achieve better service quality for the passengers and an efficient airport. Due to its importance, this paper aims to assess the performance of airports based on security, safety, and possible work-related health problems by considering the uncertain and unclear number of passengers and their baggage. Design/methodology/approach: The efficiency of 30 airports in Turkey is evaluated with a basic Data Envelopment Analysis (DEA) model with two inputs and five outputs. Then, the model is redefined with an undesired output. High, medium, and low seasons are defined in the second DEA model to estimate the number of passenger baggage. The third model utilizes the principles of Fuzzy DEA (F-DEA) that aims to handle the uncertainty for the undesired output data.Findings: The results of three models confirm that the number and weight of baggage and consequently health and safety issues in airports should not be overlooked when optimizing airport efficiency. Utilizing the fuzzy theory has the potential to help managers to improve the operational efficiency of airports when dealing with an uncertain number of passengers and estimating the workload of baggage handlers.Research limitations/implications: No permission was given to make interviews with the ground handling personnel and gather real-life data to analyze task durations and workers’ body movements.Practical implications: Inputs, outputs, and undesired output defined in this study can be used to assess the airports in any other country.Social implications: The importance of health and safety issues for passengers, airport personnel, baggage handlers, and the residents who live close to the airports is considered.Originality/value: This study contributes to the airport performance assessment literature by considering the uncertain and dynamic data related to health and safety issues. This pioneering study, up-to-best knowledge, is the first to assess the airports in Turkey by DEA with the defined undesired output (baggage handler workload) and also utilizing the fuzzy model for the uncertain data.

Operational performance model for banks: a dynamic data envelopment approach

PurposeOperational performance is critical for the banking sector for both managers and other stakeholders as it strongly affects the overall performance of the banking system. Traditional performance measures such as ratio analysis encountered certain shortcomings. At this juncture, data envelopment analysis (DEA) approaches are increasingly applied in bank efficiency studies. However, basic DEA models ignored the interactions between consecutive terms and focused primarily on measuring performance independently for each study period. All this is required to develop an operational performance model that can enable the long-term decision model.Design/methodology/approachAn attempt has been made to develop a dynamic DEA within a non-radial category to measure interconnection activities considering non-performing loans as an undesirable link. This study uses the Indian banking dataset from 2015 to 2019. The study's research design directs three directions: ‘comparison of the dynamic DEA with the traditional static DEA model, areas of inefficiencies that are investigated for each factor using the factor efficiency index and the robustness results highlighting the performance difference between bank categories.'FindingsComparing with static DEA results, the study confirms that the dynamic model best measures long-term operational performance due to the linkage between consecutive terms. The efficiency analysis concludes that the input factor that requires the most improvement is ‘fixed assets' and ‘deposits'. The output factor that needs the most progress is ‘non-interest income'. The robustness of the developed model is proven by ownership categories present within the Indian banking system. At a significance level of 10%, the result of both the separate and dynamic model for privately owned banks is significantly better than that of publicly owned banks.Originality/valueThis paper proposes an operational efficiency model for Indian banks in line with undesirable output. The mean factor efficiency analysis related to non-radial DEA modelling enhances managerial flexibilities in determining improvement initiatives.

Combined Shannon's entropy and DEA: case of Indian hotel and restaurant sector

A non-parametric linear programming-based data envelopment analysis (DEA) has been broadly applied in many sectors to measure the performance of decision-making units (DMUs). Basic DEA models such as CCR and BCC can bifurcate the DMUs into efficient and inefficient sets but cannot further identify the most efficient DMUs. Numerous methods have been proposed to address this issue and allotted ranks to DMUs. The main objective of the current paper is to implement an integrated DEA with Shannon's entropy to measure efficiencies and to rank hotels and restaurants (H&R) operating in India. The present study is carried out by integrating an orientation independent CCR, orientation dependent BCC, and non-oriented SBM with Shannon's entropy. The combined DEA-Shannon's entropy method provides comprehensive efficiency scores (CES) to all DMUs. The adopted methods have been applied to the selected 45 H&R companies in India for the year 2019-2020.

FairDEA—Removing disparate impact from efficiency scores

Achieving fairness in algorithmic decision-making tools is an issue constantly gaining in need and popularity. Today, unfair decisions made by such tools can even be subject to legal consequences. We propose a new constraint that integrates fairness into data envelopment analysis (DEA). This allows the calculation of relative efficiency scores of decision-making units (DMUs) with fairness included. The proposed fairness constraint restricts disparate impact to occur in efficiency scores, and enables the creation of a single data envelopment analysis for both privileged and unprivileged groups of DMUs simultaneously. We show that the proposed method - FairDEA - produces an interpretable model that was tested on a synthetic dataset and two real-world examples, namely the ranking between hybrid and conventional car designs, and the Latin American and Caribbean economies. We provide the interpretation of the FairDEA method by comparing it to the basic DEA and the balanced fairness and efficiency method (BFE DEA). Along with calculating the disparate impact of the model, we performed a Wilcoxon rank-sum test to inspect for fairness in rankings. The results show that the FairDEA method achieves similar efficiency scores as other methods, but without disparate impact. Statistical analysis indicates that the differences in ranking between the groups are not statistically different, which means that the ranking is fair. This method contributes both to the development of data envelopment analysis, and the inclusion of fairness in efficiency analysis.

Fuzzy preference programming formulation in data envelopment analysis for university department evaluation

PurposeUniversity department efficiency evaluation is a performance assessment on how departments use their resources to attain their goals. The most widely used tool in measuring the efficiency of academic departments in data envelopment analysis (DEA) deals with crisp data, which may be, often, imprecise, vague, missing or predicted. Current literature offers various approaches to addressing these uncertainties by introducing fuzzy set theory within the basic DEA framework. However, current fuzzy DEA approaches fail to handle missing data, particularly in output values, which are prevalent in real-life evaluation. Thus, this study aims to augment these limitations by offering a fuzzy DEA variation.Design/methodology/approachThis paper proposes a more flexible approach by introducing the fuzzy preference programming – DEA (FPP-DEA), where the outputs are expressed as fuzzy numbers and the inputs are conveyed in their actual crisp values. A case study in one of the top higher education institutions in the Philippines was conducted to elucidate the proposed FPP-DEA with fuzzy outputs.FindingsDue to its high discriminating power, the proposed model is more constricted in reporting the efficiency scores such that there are lesser reported efficient departments. Although the proposed model can still calculate efficiency no matter how much missing and unavailable, and uncertain data, more comprehensive data accessibility would return an accurate and precise efficiency score.Originality/valueThis study offers a fuzzy DEA formulation via FPP, which can handle missing, unavailable and imprecise data for output values.

Efficiency Evaluation of Regional Environmental Management Systems in Russia Using Data Envelopment Analysis

The concept of eco-efficiency has recently become an issue of great importance due to the growing trend of environmental degradation, and many approaches based on Data Envelopment Analysis (DEA) are used in the literature to evaluate the environmental performance of economic systems. However, research to date has paid little attention to the possibility of extending the DEA approach to the problem of measuring the economic feasibility of eco-efficiency improvement. The main aim of this study is to evaluate the efficiency of investments focused on improving the eco-efficiency of the regional economy in Russia using the DEA approach. The various types of costs for environmental protection measures are considered as inputs and the annual decrease in specific environmental impacts of the regional economy are considered as outputs of DEA models. This is different from previous research, which generally focused on environmental efficiency only, omitting the integration of economic aspects in eco-efficiency measures. This study compares three different modifications of basic DEA models in the context of technical complexity and practical feasibility. The results show that the efficiency of regional environmental management in many Russian regions has a great potential for improvement. From a practical point of view, the Slack-Based Measure (SBM) model provides the most accurate results for policy applications. Unlike other ratings, the DEA-SBM model may stimulate an optimization of environmental protection spending and the introduction of technological and organizational eco-innovations.

Selection and Analysis of Input-Output Variables using Data Envelopment Analysis of Decision Making Units - Indian Private Sector Banks

In banking system the evaluation of productivity and performance is the key factor among the fundamental concepts in management. For identify the potential performance of a bank efficiency is the parameter to evaluate effective banking system. To measure the efficiency of a bank selection of appropriate input-output variables is one of the most vital issues. The suitable identification of input-output variables helps to create and identify model in order to evaluate the efficiency and analysis. The Data Envelopment Analysis (DEA) is a mathematical approach used to measure the efficiency of identified Decision Making Units (DMUs). The DEA is a methodology for evaluating the relative efficiency of peer decision making units of identified input/output variables for the financial year 2018-19. In this study the basic DEA CCR, BCC models used for measure the efficiency of DMUs. In addition to these models for minimize the input excess and output shortfall Slack Based Measure (SBM) efficiency used. The SBM is a scalar measure which directly deals with slacks of input, output variables which help in obtain improved efficiency score compare with previous model. The result from the analysis is

A novel DEA model for solving performance measurement problems with flexible measures: An application to Tehran Stock Exchange

In basic data envelopment analysis (DEA) models the status of input and output measures are exactly known and given. However, in certain cases, some factors could play either input or output roles that are referred to as flexible measures. The existing approaches to deal with flexible measures solve two models for each DMU to rank the DMUs, one to classify flexible measures, and another to estimate efficiency scores. Moreover, these approaches ignore the role of non-Archimedean epsilon in the performance evaluation process. In this paper, we develop a novel mixed binary linear DEA model to classify the flexible measures and find the best DMU. Moreover, we propose an approach to find a suitable epsilon value for our developed model and suggest a new algorithm to find and rank all efficient units. An interesting case study in Tehran Stock Exchange is provided to illustrate the potential application of our new approach.

Data envelopment analysis using the binary-data

PurposeThe purpose of this paper is to propose the data envelopment analysis (DEA) model that can be used as binary-valued data. Often the basic DEA models were developed by assuming that all of the data are non-negative. However, there are situations where all data are binary. As an example, the information on many diseases in health care is binary data. The existence of binary data in traditional DEA models may change the behavior of the production possibility set (PPS). This study defines a binary summation operator, expresses the modified principles and introduces the extracted PPS of axioms. Furthermore, this study proposes a binary integer programming of DEA (BIP-DEA) for assessing the efficiency scores to use as an alternate tool in prediction.Design/methodology/approachIn this study, the extracted PPS of modified axioms and the BIP-DEA model for assessing the efficiency score is proposed.FindingsThe binary integer model was proposed to eliminate the challenges of the binary-value data in DEA.Originality/valueThe importance of the proposed model for many fields including the health-care industry is that it can predict the occurrence or non-occurrence of the events, using binary data. This model has been applied to evaluate the most important risk factors for stroke disease and mechanical disorders. The targets set by this model can help to diagnose earlier the disease and increase the patients’ chances of recovery.

Dynamic environmental efficiency analysis of China's power generation enterprises: a game cross-Malmquist index approach.

Since the power unbundling reform in 2002, China's power industry has been a typical sector moving towards marketization. The supply-side structural reform that began in 2015 has intensified the competition among China's power generation enterprises. In this context, this paper combines a data envelopment analysis (DEA) game cross-efficiency model with the Malmquist index approach to investigate the dynamic environmental efficiency of China's power generation enterprises in a competitive market. Furthermore, the system generalized method of moments (SYS-GMM) is employed to analyze the influences of factors on the dynamic environmental efficiency. The results show that: (1) Compared with the basic DEA Charnes-Cooper-Rhodes (CCR)-Malmquist index model, the DEA game cross-Malmquist index approach can enhance the discrimination among the decision making units (DMUs). (2) The change in the overall dynamic environmental efficiency of China's power generation enterprises mainly comes from the catch-up effect, while technological progress has not played a role as significant as expected. (3) The installed capacity growth rate, implementation of carbon emission trading pilots, economic growth rate, and enterprise ownership all affect the dynamic environmental performance of China's power generation enterprises during the study period.

Combined Role of Manpower Ethics and Physical Space Facilities of Bank Branches in Productivity Analysis

Productivity growth and efficiency improvements are the major sources of economic development. Pure efficiency, scale efficiency, and technology are basic factors, and rules and regulations and balance are recently known factors affecting the productivity growth. This paper focuses on the effect of manpower ethics and physical space facilities of bank branches as two factors affecting the Malmquist productivity index in the bank branches evaluation. The proposed model uses assurance region weight restrictions to increase discrimination power of basic data envelopment analysis models for the constant return to scale technologies. The validity of the proposed model and the extended Malmquist index is confirmed with an empirical case study of 74 branches of aspecializedbank in thehousing sector of Iran in two time periods of 2017 and 2018. The results for both traditional and extended indicesand the effects of manpower ethics and physical space facilities are analyzed.

Efficiency measurement to identify the best efficient unit in the presence of dual-role factors

Efficiency measurement is crucial to management for performance evaluation of several decision-making units (DMUs) in order to find the most efficient unit. Data envelopment analysis (DEA) is a practical and popular method that has been used widely for efficiency measurement in many applications. Thus, in the literature of DEA, several mixed integer DEA models are developed to efficiency measure and find the best efficient unit. In some situations, some factors named dual-role factors may be simultaneously considered as both inputs and outputs. However, a little attention has been paid in the literature of DEA to these types of factors. The current paper develops a new mixed integer linear DEA model to identify the best efficient DMU in the presence of dual-role factors. The basic DEA models consider different roles for each dual-role factor, and also solve at least one model for each DMU to determine the best unit. However, the developed approach in this paper considers a unique role for each dual-role factor, and also determines the best unit through by solving just one mixed integer DEA model. Two numerical examples are used to illustrate the profits and applicability of the developed approach.

Application of Bootstrap-DEA With Fuzzy Computing in Performance Evaluation of Forklift Leasing Supplier

The study uses data envelopment analysis (DEA) combined with fuzzy computing to construct an improved evaluation model of service providers. From the perspective of a forklift company's scale and fleet level, the performance appraisal system of forklift leasing service providers is established, where the two fuzzy variables, namely team level, and service support, are embodied and calculated to be more specific. Bootstrap-DEA method is used to analyze the performance of the service provider in combination with qualitative and quantitative analysis, improving the consistency of the estimator of the basic DEA method. This research proposed an effective performance evaluation system for the assessment of forklift rental service providers. The effectiveness of this evaluation based on bootstrap-DEA with fuzzy computing can help enhance the management of logistic companies.

New approach for ranking efficient DMUs based on Euclidean norm in data envelopment analysis

Data envelopment analysis (DEA) is a widely used technique for measuring the relative efficiencies of decision-making unit (DMUs) with multiple inputs and multiple outputs. In many applications, ranking of DMUs is an important and essential procedure to decision makers in DEA, especially when there are extremely efficient DMUs. Basic DEA models usually give the same efficiency score for some DMUs. Hence, it is necessary to rank of all extreme DMUs. The main purpose of this study is to propose an appropriate method in order to overcome the drawbacks in several methods for ranking DMUs based on the DEA concept. In the present paper, we propose a model for ranking extreme efficient DMUs in DEA by super efficiency technique and Euclidean norm (l2-norm). The presented method in this paper is able to overcome the existing obstacles in some methods. As regards, the proposed model is into nonlinear programming form, a linear model is suggested to approximate the nonlinear model.

Investigation of accurate location planning for wind farm establishment: a case study

Purpose Wind as a major source of renewable energy has received tremendous attentions due to its unique features to reduce carbon emission and also to keep the environment safe. Nevertheless, to use wind energy properly, the environmental circumstances and geographical location related to wind intensity should be considered as a priority. Different factors may affect the selection of a suitable location for developments of wind power plants; thus, these factors should be considered concurrently to identify the optimum location of wind plants. Design/methodology/approach In this study, first, basic data envelopment analysis (DEA) was used, then dual DEA was used and, finally, Anderson Petersen (AP) model of dual DEA was selected to prioritize cities or decision-making units (DMUs). Numerical Taxonomy (NT) method was also used to assess the validity of AP dual model in DEA. The prescribed approach was applied for five cities in East Azerbaijan province of Iran. Findings The results indicate that wind power as a renewable energy can be harnessed in few cities, and the ranking by DEA illustrated that the city of Tabriz is the first priority. Practical implications Low environmental degradation effects in comparison to other methods and the ability to utilization at a widespread level include the benefits of using wind energy in the generation of electricity. In this regard, the study of relevant potentials and finding suitable locations for the deployment of wind energy utilization equipment are essential. Using DEA method helps us to choose optimal locations according to different criteria. Social implications Wind energy is justifiable in reducing social costs in comparison with fossil fuel plants, which includes negative effects, and its electricity can be used as a sustainable energy in the country's economic, social and cultural development. Originality/value For identifying the most proper location for development of wind power plants in Iran, DEA is applied for the first time to prioritize the suitable locations for installations of wind turbines among five different cities in the East Azerbaijan region. A number of crucial factors including land price, distance to power, rate of natural hazards, wind speed and topography are considered for location optimization of wind turbines for the first time. Also, to validate the results of DEA method, NT method is used to assess the validity of AP dual model in DEA.

A heuristic method for choosing 'virtual best' DMUs to enhance discrimination power of augmented DEA model

Despite its intrinsic advantageous features as a tool for increasing discrimination power of the basic DEA (data envelopment analysis) model, augmented DEA has two main drawbacks including the presence of unrealistic efficiency scores and the presence of great distance between its efficiency scores and scores obtained by primary model. In this regard, this paper extends a heuristic method for dealing with both issues and improving the power of augmented DEA model in performance evaluation. Since different virtual DMUs lead to different results for ranking, the hierarchical clustering algorithm is applied in this study to select the best virtual DMUs in order to reduce the possibility of having inappropriate efficiency scores. Finally, to demonstrate the superiority of the proposed approach over previous approaches in literature, two numerical examples are provided.

Handling negative data in slacks-based measure data envelopment analysis models

This paper proposes slacks-based measure (SBM) data envelopment analysis (DEA) models that handle negative data. Unlike existing negative data allowable DEA models, the proposed SBM DEA models are consistent with ordinary SBM models and units invariant, they handle various types of returns to scale, and they avoid division by zero. These new SBM DEA models transform original negative inputs and outputs into positive counterparts based on a newly defined “base point”. Hence, these models are referred to as the BP-SBM DEA models. In addition to the basic BP-SBM DEA models, this research further develops data-oriented and application-oriented BP-SBM DEA-type models for different application problems involving negative data. Numerical examples are provided to illustrate various aspects and implementation details of these models.

A data envelopment analysis model for evaluating the performance of taxi operators: a case study in Harbin, China

ABSTRACTThis paper measures the performance of taxi operators using data envelopment analysis (DEA) and the improvement of performance based on benchmarks. This includes basic DEA models and a cross efficiency model. In the present work, a sample set of 20 taxi operators are considered, and an evaluation indicator framework is constructed, which mainly involves the identification of decision-making units (DMUs) and the determination of input-output variables. The combined results of multiple DEA models are used to evaluate the performance of taxi operators in a case study of Harbin. The results highlight the significance of the combined application of multiple DEA models, especially a cross-efficiency model that can obtain more objective efficiency values. Finally, the benchmarking method is used to capture the achievements of best-performing operators and provide valuable improvement recommendations for inefficient operators. Conclusions provide a suitable basis for the operation and management of the taxi market.