¹York University, Toronto, ON, Canada

Correspondence: Z Yang, School of Information Technology, Atkinson Faculty of Liberal & Professional Studies, York University, 4700 Keele Street, Toronto, Ontario M3J 1P3, Canada. E-mail: zyang@mathstat.yorku.ca

Received May 2007; Accepted March 2008; Published online 21 May 2008.

Abstract

This paper examines the performance of 758 branches of one big Canadian bank nationwide using Data Envelopment Analysis (DEA) from several perspectives. The system-differentiated DEA models used in the paper consider the systematic difference among different geographical areas. Five alternatives to express outputs are presented in order to provide complementary information to the bank management and further investigation of the cause of inefficiency when assessing bank branches is given. Moreover, the correlation analysis of different models is provided. The potential management uses of DEA results are also presented.