Shapley value-based multi-objective data envelopment analysis application for assessing academic efficiency of university departments

Subject Areas : Mathematical Optimization

Stephen Lloyd N. Abing ¹ , Mercie Grace L. Barton ² , Michael Gerard M. Dumdum ³ , Miriam F. Bongo ⁴ , Lanndon A. Ocampo ⁵

1 - Department of Industrial Engineering, School of Engineering, University of San Carlos, Nasipit, Talamban, Cebu, 6000, Philippines
2 - Department of Industrial Engineering, School of Engineering, University of San Carlos, Nasipit, Talamban, Cebu, 6000, Philippines
3 - Department of Industrial Engineering, School of Engineering, University of San Carlos, Nasipit, Talamban, Cebu, 6000, Philippines
4 - Department of Mechanical and Manufacturing Engineering, School of Engineering, University of San Carlos, Nasipit, Talamban, Cebu, 6000, Philippines
5 - Department of Industrial Engineering, College of Engineering, Cebu Technological University, Corner M.J. Cuenco Ave. & R. Palma St., Cebu City, 6000, Philippines

Received: 2020-10-07 Accepted : 2020-10-07 Published : 2018-12-01

Keywords: Academic efficiency  Data envelopment analysis  Multi, objective data envelopment analysis  University departments,

Abstract :

This paper adopts a modified approach of data envelopment analysis (DEA) to measure the academic efficiency of university departments. In real-world case studies, conventional DEA models often identify too many decision-making units (DMUs) as efficient. This occurs when the number of DMUs under evaluation is not large enough compared to the total number of decision variables. To overcome this limitation and reduce the number of decision variables, multi-objective data envelopment analysis (MODEA) approach previously presented in the literature is applied. The MODEA approach applies Shapley value as a cooperative game to determine the appropriate weights and efficiency score of each category of inputs. To illustrate the performance of the adopted approach, a case study is conducted in a university in the Philippines. The input variables are academic staff, non-academic staff, classrooms, laboratories, research grants, and department expenditures, while the output variables are the number of graduates and publications. The results of the case study revealed that all DMUs are inefficient. DMUs with efficiency scores close to the ideal efficiency score may be emulated by other DMUs with least efficiency scores.