• The Journal of the Korea Contents Association
• /
• v.17 no.1
• /
• pp.201-212
• /
• 2017

This study presented the new evaluation index which can evaluate the discrimination of DEA models. To evaluate the discrimination of DEA models, data were analyzed using importance index as suggested in previous study and the coefficient of variation as suggested in this study for the discrimination evaluation. This study selected the CCR-DEA, BCC-DEA, entropy, bootstrap, super efficiency, and cross efficiency DEA model for the discrimination evaluation and accomplished empirical analysis. In order to grasp the rank correlation of the models, this study implemented the rank correlation analysis between the efficiency of CCR model and BCC model and entropy, bootstrap, super efficiency, and efficiency of the cross efficiency model. The obtained results of this study are as follows. First, the discrimination rank of models using the importance index and the coefficient of variation was shown to be identical. Therefore, the coefficient of variation can be used the discrimination evaluation index of DEA model. Second, the discrimination of the super efficiency model was found to be the highest rank among 4 models according to the analysis of this present study. Third, the highest rank correlation with CCR model was the super efficiency model. In addition, the super efficiency model was found to be the highest rank correlation with BCC model.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.05a
• /
• pp.298-299
• /
• 2018

In this study, the discriminant power of the DEA models which do not require the prior information of decision makers was evaluated. Entropy model, Bootstrap model, Benevolent Cross Efficiency model, Aggressive Cross Efficiency model and Game Cross Efficiency model were selected as the DEA model for discriminant power evaluation. The discriminant power of five DEA models was evaluated using coefficient of variation and degree of importance. According to the evaluation results, the rank of discrimination power was evaluated in the order of Entropy model, Aggressive CE model, Benevolent CE model, Game CE model, and Bootstrap model in both evaluation indexes.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.5B
• /
• pp.364-378
• /
• 2008

Data Envelopment Analysis(DEA) models can be used for performance evaluation on governmental funding projects for IT small and medium-sized enterprises associated with multiple-outputs/multiple-inputs. In order to enhance the accuracy of DEA efficiency scores, DEA models with exogenously fixed variables are required where the corporate competency is taken into account. Additionally, it is necessary to use multiple DEA basic as well as extended models so as to relax the restriction on the performance evaluation to relying on a single DEA model. In this study; 1)a DEA data structure is designed including exogenously fixed variables representing corporate asset, revenue and the number of employees at the point in time that the governmental funding project concerned is initiated; 2)DEA basic as well as extended models are established according to the DEA data structure presented abovementioned; and 3)a case study is illustrated with an empirical testbed dataset. As for the DEA basic models, CCR, BCC, Super-efficiency model are adopted. The DEA extended models are developed based on the models associated with noncontrollable and nondiscretionary variables. In the case study, it is explained a comparison of DEA models and also major numerical outcomes such as efficiency scores, ranks derived from each DEA model are integrated using Analytic Hierarchy Process(AHP) weights. Performance significance with DEA efficiency scores between technical categories are tested based not only on parametric but also nonparametric single-factor analysis of variance method.

• Journal of Korea Port Economic Association
• /
• v.28 no.2
• /
• pp.179-194
• /
• 2012

In this study, we compared the efficiency of container terminals using DEA. To do this, we designed an AHP/DEA hybrid model using AHP and DEA, and evaluated the efficiency by comparing the container terminal operation company in Gwangyang(KEC, KIT, GICT) and Busan(HBCT, DPCT, KBCT, UPT, Gamman, PNC, PNIT, HJNC, HPNT). The proposed model can control the number of selected promising container terminal by applying DEA-AR model. This model can also improve the credibility of analysis by using objective weights through the AHP application to efficiency evaluation data and normalizing the evaluation data to apply AHP and DEA. The model assumes inputs to be container crane, transfer crane, yard tractor, and reach stacker and output as container traffic. The result shows that DPCT was an efficient DMU.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2012.10a
• /
• pp.109-111
• /
• 2012

한-EU FTA 발효 등은 교역 확대가 예상되며, 항만의 역할은 더욱더 중요해 질 것이다. 항만의 성과를 측정할 수 있는 방안으로 DEA를 활용한 효율성 측정을 들 수 있으며, 본 연구에서는 DEA 측정 시 모형 선정의 어려움과 동태적 효율성을 변화를 분석하기 위해 DEA와 Shannon's Entropy 결합모형으로 효율성을 분석하고, DEA-Window 모형으로 동태적 효율변화를 분석하였다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.1 no.1
• /
• pp.83-90
• /
• 2000

DEA is a newly developed analyzing tool to measure efficiency evaluation of decision making units (DMU). It compares DMU by radial Projection on the efficient frontier. The purpose of this study is to show reference point approach used for searching solution in multiple objective linear Programming can be usefully used to determine flexible efficient frontier of each DMU In reference point approach, the minimization of ASF Produces an efficient points in frontier and enhances the usefulness of DEA by Providing flexibility in DEA and optimally allocating resources to DMU. Various DEA models can be supported by reference point method by changing the projection direction in order to choose the targets units, standards costs and management benching-marking.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.5 no.2
• /
• pp.191-196
• /
• 2004

Allocative efficiency measures the extents to which the technically efficient units falls shorts of achieving minimal cost. By using this measure manager can make decision about how to redistribute organizational resources to improve price efficiency. Allocative and overall efficiency are derived on the basis of budget line and cost minimization concept. The purpose of this study is to introduce the concept of allocative efficiency and propose two modified DEA models. Examples are provided to illustrate the similarities and the application procedure of the two model. By providing example and tracing the data application procedure, we found the same results but some cautions are needed to interpret the valuation.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.12 no.6
• /
• pp.2500-2505
• /
• 2011

This study pursues to compare the two types of cost efficiency model based on DEA. Two types of DEA are derived by the two different approaches by Farrell_Debreu and Tone. Based on two concepts, Two different DEA model are derived. The characteristics and difference of two are looked up. Based on the simple numerical case, The efficient rates, the rankings, the reference sets are different. The model based on Tone's approach shows the more cheap attainable target cost level. DEA model set by Tone is superior in measuring cost efficiency, but Farrell_Debreu type DEA model is better to explain data in technical efficiency. So, it is required to use the results of DEA more carefully.

• Journal of agriculture & life science
• /
• v.45 no.6
• /
• pp.279-289
• /
• 2011

The purpose of this study is to estimate efficiency of item-based agricultural cooperative by using Data Envelopment Analysis. A proposed method employs a bootstrapping approach to generating efficiency estimates through Monte Carlo simulation resampling process. The technical efficiency, pure technical efficiency, and scale efficiency measure of item-based agricultural cooperative is 0.80, 0.87, 0.93 respectively. However the bias-corrected estimates are less than those of DEA. We know that the DEA estimator is an upward biased estimator. In technical efficiency, average lower and upper confidence bounds of 0.726 and 0.8747. According to these results, the DEA bootstrapping model used here provides bias-corrected and confidence intervals for the point estimates, it is more preferable.

• Journal of Korea Port Economic Association
• /
• v.31 no.1
• /
• pp.85-110
• /
• 2015

The purpose of this paper is to show the clustering trend and the empirical comparison and to choose the clustering ports for 3 Korean ports(Busan, Incheon and Gwangyang Ports) by using the Fuzzy(Average Index Transformation) DEA and Cross-efficiency models for 38 Asian ports during 11 years(2001-2011) with 4 input variables(birth length, depth, total area, and number of crane) and 1 output variable(container TEU). The main empirical results of this paper are as follows. First, clustering results by using Fuzzy(AIT)DEA show that 3 Korean ports[Busan(56.29%), Incheon(57.96%), and Gwangyang(66.80%) each]can increase the efficiency. Second, according to Cross-efficiency model, Busan(Hongkong, Kobe, Manila, Singapore, and Kaosiung etc.), Incheon(Aquaba, Dammam, Karachi, Mohammad Byin Oasim and Davao), and Gwangyang(Damman, Yokohama, Nogoya, Keelong, Kaosiung, and Bangkok) should be clustered with those ports in parentheses. Third, when both Fuzzy(AIT)DEA and Cross-efficiency models are mixed, the empirical result shows that 3 Korean ports[Busan(71.38%), Incheon(103.89%), and Gwangyang(168.55%) each]can increase the efficiency. The efficiency ranking comparison among the three models by using Wilcoxon Signed-rank Test was matched with the average level of 66%-67%. The policy implication of this paper is that Korean port policy planner should introduce the Fuzzy(AIT)DEA, and Cross-efficiency models with the mixed two models when clustering is needed among the Asian ports for enhancing the efficiency of inputs and outputs. Also, the results of SWOT analysis among the clustering ports should be considered.