Research Article

Dynamic fuzzy data envelopment analysis models: case of bus transport performance assessment

¹ Griffith School of Engineering, Gold Coast Campus, Griffith University, 4222, Australia
² Transportation and Logistics Management, National Chiao Tung University. 4F, 118 Sec. 1, Chung-Hsiao W. Rd., Taipei 10012, Taiwan, ROC

^* Corresponding author: t.yen@griffith.edu.au

Received: 30 September 2016
Accepted: 7 September 2017

Abstract

In the transport field, two characteristics–inter-temporal dependency and fuzziness–need to be considered when assessing transport performance. First, input and output levels are inter-temporal dependent due to heavy capital investment and because quasi-fixed input can influence output levels over multiple periods. Second, conventional Data Envelopment Analysis (DEA) models are, in nature, formulated with quantitative variables. However, qualitative measurements that are characterized with “vagueness” or “fuzziness” are as important as quantitative variables for multi-period transport performance assessment. To rectify these problems, the present study extends previous research by proposing a Dynamic Fuzzy Data Envelopment Analysis (DFDEA) method for assessing the comparative efficiency where inter-temporal dependence exists in operating production processes with some “fuzzy” variables. An case study was conducted to evaluate the performance of city bus transport companies in Taipei, Taiwan. Results showed the superiority of the proposed DFDEA model by comparing the results with static models.