DEA 모형을 이용한 전라남도 지역 축제의 정량적 효율성 평가

The purposes of this paper is to evaluate the relative efficiency of regional festivals and to provide inefficient group members with benchmark models for improving their management efficiency. To do this, a DEA(Data Envelopment Analysis) is taken advantage of as the efficiency evaluation method. DEA, as a non-parametric approach, evaluates relative efficiency of similar inputs and output structure and determines a set of Pareto-efficient decision making units(DMUs) with an objective of calculating a discrete piecewise frontier. In this paper, we use a DEA model to identify efficient and inefficient groups and measured efficiency scores of the Jeonnam regional festivals using eleven data for 2006-2008. With respect to efficiency analyzed by CCR and BCC models, the Jindo Miracle Sea festival stands high in 2006, whereas the Kwangyang Maehwa(Japanese apricot flower) Culture and the Gurye Sansuyoo(Cornus Officinalis) festivals show high in 2007 and 2008. In addition, we propose a policy of improving in management efficiency out of regard for output. We also performed Tier analysis for suggesting benchmark cases that could be as a role models to the members of the inefficient groups. Tier analysis is a technique that can cluster DMUs according to their efficiency levels. In this analysis, the efficiency scores of entire DMUs is obtained. The results reveal the most efficient group by indicating their scores are equal to 1. We call this group ‘tier 1’. Then, the analysis proceeds DEA again only with the inefficient DMUs which are not on tier 1. DMUs whose efficiency scores are equal to 1 are set ‘tier 2.’ We repeat the same procedure while the number of remaining inefficient DMUs is at least two times multiple of that of input plus output variables. As a result of Tier analysis, we find that the great battle of Myeongryang festival may take the Bubsung Dano festival, Damyang Bamboo festival and Gwangyang Maehwa festival as their benchmark cases. And Hampyeong butterfly festival may take Gokseong Simcheong festival, Damyang Bamboo festival and Gwangyang Maehwa festival as their benchmark cases. This result may provide a visible road-map for the local government. The local government can better understand the situation their festival and they prepare better strategies for the future. The results of the DEA model may be different according to the used input/output variables. This study examined the data of local government reports and selected proper elements. However, as the comparative efficiency of DMUs may change according to the input and output variables, it is hoped that there will be a number of studies to analyze efficiency of regional festivals by means of DEA in order to make the accurate evidences to identify optimal input and output combination.