存在保证域的模糊非径向偏好DEA模型——基于中科院24个研究所的实证分析

数据包络分析（DEA）是一种非参数化的方法,用于评价具有类似输入和输出的决策单元的效率。传统的非径向DEA模型假设输入和输出数据均为准确值,且对权重变量不加以限制,本文构建了存在保证域的模糊非径向偏好DEA模型,并给出了一种基于模糊数截集的模型求解方法,有效地解决了输入和输出全部或部分为模糊数的决策单元评价问题。最后给出了一个中科院研究所效率评价的实例说明了方法的有效性。     

A slack-based measure of efficiency in context-dependent data envelopment analysis

Data envelopment analysis (DEA) has been proven as an excellent data-oriented performance evaluation method when multiple inputs and outputs are present in a set of peer decision-making units (DMUs). In the DEA literature, a context-dependent DEA is developed to provide finer evaluation results by examining the efficiency of DMUs in specific performance levels based upon radial DEA efficiency scores. In DEA, non-zero input and output slacks are very likely to present after the radial efficiency score improvement. Often, these non-zero slack values represent a substantial amount of inefficiency. Therefore, in order to fully measure the inefficiency in DMU's performance, it is very important to also consider the inefficiency represented by the non-zero slacks in the context-dependent DEA. This study proposes a slack-based context-dependent DEA which allows a full evaluation of inefficiency in a DMUs performance. By using slack-based efficiency measure, we obtain different frontier levels and more appropriate performance benchmarks for inefficient DMUs.     

Measuring effectiveness of production in the public sector

Key concepts in efficiency analyses are efficiency and effectiveness. Efficiency is popularly connected to ‘doing the things right’ and effectiveness to ‘doing the right things’. The paper elaborates upon the latter concept within a setting where resources are transformed into outputs under the control of a public provider, while outcomes with outputs as inputs represent higher social goals, but this production is outside the public provider׳s direct control. A new measure of overall preference effectiveness is introduced and its decomposition into output-oriented efficiency and output-mix efficiency is shown. The monumental task of getting the necessary information for calculating effectiveness is highlighted.     

Stochastic efficiency analysis with a reliability consideration

Stochastic Data Envelopment Analysis (DEA) models have been introduced in the literature to assess the performance of operating entities with random input and output data. A stochastic DEA model with a reliability constraint is proposed in this study that maximizes the lower bound of an entity׳s efficiency score with some pre-selected probability. We define the concept of stochastic efficiency and develop a solution procedure. The economic interpretations of the stochastic efficiency index are presented when the inputs and outputs of each entity follow a multivariate joint normal distribution.     

Application of a novel PROMETHEE-based method for construction of a group compromise ranking to prioritization of green suppliers in food supply chain

The food sector has a prodigious focus and is constantly gaining in importance in today’s global economic marketplace. Due to an increasing global population, society faces a greater challenge for sustainable food production, quality, distribution, and food safety in the food supply chain. Adopting green supply chain management (GSCM) elements is essential for utilizing the food supply chain in an environmentally benign way. As a solution to the above challenge, the economic and green characteristics for supplier selection in green purchasing are studied in this paper. For an organization, the evaluation and selection of the green supplier is a vital issue due to several tangible and intangible criteria involved. Accordingly, we apply multiple criteria decision aiding techniques.We propose a hybrid approach that combines the revised Simos procedure, PROMETHEE methods, algorithms for constructing a group compromise ranking, and robustness analysis. At first, the revised Simos procedure is used to derive the criteria weights. Next, the PROMETHEE method is applied to rank the suppliers according to each Decision Maker׳s (DM׳s) preferences. Then, the compromise ranking is constructed to minimize the distance of the individual׳s rankings from the solution adopted by the whole group. For this purpose, we introduce and apply some original procedures based on Binary Linear Programming. Finally, the results are validated against the outcomes of robustness analysis. The applicability and efficiency of the proposed approach is endorsed with a case study in an Indian food industry.     

Measuring departmental and overall regional performance: applying the multi-activity DEA model to Taiwan׳s cities/counties

This study measures the departmental and overall efficiency in Taiwan׳s counties/cities by applying a multi-activity data envelopment analysis (MADEA) model. The model overcomes the problems of panel data, undesirable outputs, shared inputs, and environmental variables and intertemporal efficiency changes (productivity) by applying the Malmquist–Luenberger (ML) index. We include data on the economic development, social welfare, police and security, and education departments for 20 counties/cities in Taiwan for the period 1999–2013. We find that the police security department is the most efficient in most counties/cities in the period 1999–2013, and the economic development department is the second efficient one in 2002–2005 and after 2009. Furthermore, there exist urban–rural gaps in the efficiency scores between counties and cities, between service-type and non-service-type counties/cities, and among different regions. With regard to the efficiencies over time (ML indices and their decompositions), we find that the production frontiers of the social welfare and education departments in Taiwan׳s counties/cities expanded continuously during this period. Finally, we also find that urban–rural gaps and gaps between service-type and non-service-type counties/cities exist in terms of technological changes and ML productivity indices in the social welfare and education departments. The area differences of technological changes exist in 4 departments and in overall. Our results will help the mayors of counties/cities understand the strengths and weaknesses of the regions they govern.     

随机环境下的多投入多产出生产前沿面估计

传统的DEA模型假设观测样本的投入产出都是确定型数据, 这使得DEA在实际应用中受到限制, 本文提出的基于拟似然估计的多投入多产出随机非参数包络数据 (PLE-StoNED) 方法拓展了这个假设, 能够估计随机环境下的生产前沿面.本文证明, 生产可能集假设条件下的前沿面可以用一个有凹凸性和单调性限制的函数来表示.相较之前的StoNED方法, 本文提出的方法可以估计随机环境下多投入多产出决策单元 (DMU) 的前沿面.通过Monte Carlo实验, 多投入多产出PLE-StoNED方法的有效性得到验证, 它可纠正DEA等传统方法产生的偏误.最后, 实证研究部分运用这一新提出的方法估计了中国大陆商业银行的生产前沿面和效率.本文提出的方法弥补了DEA缺乏统计性的不足, 可为决策者在随机环境下对多投入多产出决策单元进行生产力和效率评估提供决策参考.     

Multi-output profit efficiency and directional distance functions

We extend a recently developed DEA methodology for cost efficiency analysis towards profit efficiency settings. This establishes a novel DEA toolkit for profit efficiency assessments in situations with multiple inputs and multiple outputs. A distinguishing feature of our methodology is that it assumes output-specific production technologies. In addition, the methodology accounts for the use of joint inputs, and explicitly includes information on the allocation of inputs to individual outputs. We also establish a dual relationship between our multi-output profit inefficiency measure and a technical inefficiency measure that takes the form of a multi-output directional distance function. Finally, we demonstrate the empirical usefulness of our methodology by an empirical application to a large service company.     

存在非期望产出的决策单元阻塞问题研究

阻塞是指企业的过量投入引起生产效益下降的经济现象,近年来已经受到实业界与学术界的广泛关注。现有的决策单元阻塞问题研究普遍假设产出均是期望指标,然而,在实际生产中很多产出属于非期望指标。本文构建了同时存在期望产出和非期望产出决策单元的阻塞度量模型,该模型不仅可以判别决策单元是否存在阻塞,而且可以测算各项投入的阻塞程度,并给出了决策单元阻塞和技术无效之间的关系。最后,运用所提出的模型对我国上市商业银行的效率和阻塞情况进行了测算与分析,验证了本文模型的实用性与可行性。     

Measuring performance of two-stage network structures by DEA: A review and future perspective

Data envelopment analysis (DEA) is a method for measuring the efficiency of peer decision making units (DMUs). An important area of development in recent years has been devoted to applications wherein DMUs represent two-stage or network processes. One particular subset of such processes is those in which all the outputs from the first stage are the only inputs to the second stage. The current paper reviews these models and establishes relations among various approaches. We show that all the existing approaches can be categorized as using either Stackelberg (leader-follower), or cooperative game concepts. Future perspectives and challenges are discussed.     

A slacks-based inefficiency measure for a two-stage system with bad outputs

We model the performance of DMUs (decision-making units) using a two-stage network model. In the first stage of production DMUs use inputs to produce an intermediate output that becomes an input to a second stage where final outputs are produced. Previous black box DEA models allowed for non-radial scaling of outputs and inputs and accounted for slacks in the constraints that define the technology. We extend these models and build a performance measure that accounts for a network structure of production. We use our method to estimate the performance of Japanese banks, which use labor, physical capital, and financial equity capital in a first stage to produce an intermediate output of deposits. In the second stage, those deposits become an input in the production of loans and securities investments. The network estimates reveal greater bank inefficiency than do the estimates that treat the bank production process as a black box with all production taking place in a single stage.     

Evaluating the administrative efficiency of courts

In addition to being held accountable for judicial decision, courts, like other public agencies, can and should be evaluated in terms of their administrative efficiency. This paper illustrates how courts can be evaluated in terms of their relative administrative efficiency, using a new approach—data envelopment analysis (DEA)—first proposed by Charnes et al. [1]. The DEA is based upon the economic notion of Pareto optimality which states that a given decision making unit (DMU) is inefficient if some other DMU, or some combination of other DMUs, can produce at least the same amounts of all outputs with less of some resource input and not more of any other resource. Conversely a DMU is said to be efficient if the above is not possible. Charnes et al. [1] generalized the usual input/output ratio measure of efficiency for a given unit in terms of a fractional linear program with fractional constraints. In the case of courts, the efficiency of any particular court is calculated by forming the ratio of a weighted sum of outputs to a weighted sum of inputs, where the weights for both outputs and inputs are to be selected in a manner that calculates the Pareto-Koopmans efficiency of the court. This paper reviews the DEA method and illustrates its application to a data base for 100 criminal superior courts in North Carolina.     

Eco-efficiency analysis of paper mills along the Huai River: An extended DEA approach

The objective of this paper is to estimate the ecological efficiency of paper mills along the Huai River in China. The main characteristic of the ecological efficiency evaluation problem is that an undesirable output of biochemical oxygen demand (BOD) and a non-discretionary input (BOD emission quota) should be considered simultaneously. By analyzing the impacts of the non-discretionary input on decision-making units’ (DMUs) desirable and undesirable outputs, a non-radial output-oriented DEA model is proposed. In the proposed model, we describe a new approach of defining reference set that requires reference units operate in a similar environment on average. We employ the model to provide efficient inputs/outputs targets for DMU managers to improve DMUs’ efficiencies. Based on the developed model, impacts of the non-discretionary input on DMUs’ returns are also analyzed. We illustrate the proposed model, using real data, for 32 paper mills along the Huai River in China.     

Non-convex value efficiency analysis and its application to bank branch sales evaluation

We have observed when applying Value Efficiency Analysis [21] that decision makers wish to provide preference information related to existing rather than virtual (efficient) units. This observation motivated us to develop an approach based on the preference comparisons of existing units. The Free Disposal Hull model provides the requisite framework. We assume that a Decision Maker compares units using an implicitly known value function that reaches its maximum at his/her most preferred (efficient) unit. The unknown value function is assumed to be quasi-concave in outputs and quasi-convex in inputs. The main purpose – as in the original Value Efficiency Analysis – is to approximate the distance of each unit from the contour of the value function passing through the most preferred unit. We use examples to illustrate the approach. Finally, we describe a real application in which Value Efficiency Analysis was used to produce information for bank managers wishing to evaluate the performance of bank branches.     

Data envelopment analysis: Prior to choosing a model

In this paper, we address several issues related to the use of data envelopment analysis (DEA). These issues include model orientation, input and output selection/definition, the use of mixed and raw data, and the number of inputs and outputs to use versus the number of decision making units (DMUs). We believe that within the DEA community, researchers, practitioners, and reviewers may have concerns and, in many cases, incorrect views about these issues. Some of the concerns stem from what is perceived as being the purpose of the DEA exercise. While the DEA frontier can rightly be viewed as a production frontier, it must be remembered that ultimately DEA is a method for performance evaluation and benchmarking against best-practice. DEA can be viewed as a tool for multiple-criteria evaluation problems where DMUs are alternatives and each DMU is represented by its performance in multiple criteria which are coined/classified as DEA inputs and outputs. The purpose of this paper is to offer some clarification and direction on these matters.     

Pitfalls of decomposition weights in the additive multi-stage DEA model

This paper examines limitations of the multi-stage DEA (data envelopment analysis) model in the literature. We focus on the DEA model with additive efficiency decomposition. We create taxonomy for the multi-stage DEA models and show when the decomposition weights can be non-increasing. When the decomposition weight for a stage is deemed reflective of the stage׳s relative importance, this property then implies that upstream stages (regardless the stage efficiency scores) in the model will obtain higher priority in efficiency decomposition. We also find that the non-increasing weights can affect the evaluation of overall and stage efficiency scores. We illustrate our findings through an empirical data set.     

Heuristics for prioritizing pair-wise elicitation questions with additive multi-attribute value models

Additive value models are widely used in Multiple Criteria Decision Analysis. Direct elicitation of the value model preference parameters can impose excessive cognitive burden on the decision maker. Indirect techniques that employ pair-wise questions have been proposed for lowering the elicitation effort. In all practically relevant problems, more than a single question needs to be answered for arriving at a sufficiently precise outcome. The selection and ordering of questions affects the number of answers required for ranking the decision alternatives. However, evaluating all possible questions and answers is intractable due to the search space being, in the worst case, of factorial size. This paper develops heuristics for prioritizing pair-wise elicitation questions based on (1) necessary preference relations, (2) extreme ranks attained by the alternatives, (3) pair-wise preference indices, and (4) rank acceptability indices. We also introduce three metrics for assessing quality of a question prioritization heuristic. Numerical results allow us to identify a subset of heuristics that score well on our metrics in a variety of problem settings. This conclusion was validated in a real-world experiment where 101 subjects answered pair-wise questions to rank 10 mobile phone packages evaluated in terms of four criteria.     

Common benchmarking and ranking of units with DEA

This paper develops a common framework for benchmarking and ranking units with DEA. In many DEA applications, decision making units (DMUs) experience similar circumstances, so benchmarking analyses in those situations should identify common best practices in their management plans. We propose a DEA-based approach for the benchmarking to be used when there is no need (nor wish) to allow for individual circumstances of the DMUs. This approach identifies a common best practice frontier as the facet of the DEA efficient frontier spanned by the technically efficient DMUs in a common reference group. The common reference group is selected as that which provides the closest targets. A model is developed which allows us to deal not only with the setting of targets but also with the measurement of efficiency, because we can define efficiency scores of the DMUs by using the common set of weights (CSW) it provides. Since these weights are common to all the DMUs, the resulting efficiency scores can be used to derive a ranking of units. We discuss the existence of alternative optimal solutions for the CSW and find the range of possible rankings for each DMU which would result from considering all these alternate optima. These ranking ranges allow us to gain insight into the robustness of the rankings.     

A modified DEA cross efficiency method with negative data and its application in supplier selection

Basic Data Envelopment Analysis (DEA) models are designed for non-negative data. However, negative data is inevitably used in many real-world issues. Also, multiple units with a maximum relative performance score (equal to one) can be obtained due to the benevolent view of evaluating Decision Making Units (DMUs) consistent performance. Therefore, the researchers proposed ranking models to differentiate efficient units. Cross efficiency is one of the most useful tools for DMUs ranking in the DEA. There are two major drawbacks to implementing this process. First, it gives different results in the presence of other optimal solutions; second, it does not provide a compelling reason to use the arithmetic mean to aggregate the results of the cross efficiency matrix. In this paper, first a new non-radial model is proposed to evaluate the performance of DMUs in the presence of negative data and then based on this model a new secondary goal model is proposed to eliminate the first drawback in the cross efficiency method. Also, to solve the second drawback in this method, a hybrid Multi-Attribute Decision Making (MADM)-DEA process with the help of fuzzy VlseKriterijumska Optimizacija I Kompromisno Resenje method is proposed. Finally, to show the applicability of the proposed methods, the results are used to select the supplier in a real-world problem.

     

Diversification-consistent data envelopment analysis based on directional-distance measures

We propose new diversification-consistent DEA models suitable for assessing efficiency of investment opportunities available on financial markets. The formulations based on directional distance measures enable to use several risk measures as inputs and return measures as outputs, which can take both positive and negative values. We show that various models with different strength can be obtained and strongly, semi-strongly or weakly Pareto–Koopmans efficient investment opportunities can be identified. Moreover, the optimal solutions correspond to efficient investment opportunities and can be used by investors to revise the inefficient ones. If we consider discretely distributed returns, we can prove that under proper choice of the inputs (CVaRs) and outputs (expected return), the strongest model is able to identify efficient investment opportunities with respect to the second-order stochastic dominance. Moreover, the model can be formulated as a linear programming problem. In the numerical study, the proposed DEA models are applied to 48 representative industry portfolios from US stock markets.