Premium
Using the nonhomogeneous frontier two‐stage DEA model to assess the efficiencies of expense utilization and operation of the Taiwanese hotel industry
Author(s) -
Huang Chinwei,
Chiu Yungho,
Tu Chienheng,
Luo Zhengying,
Wang Zebin
Publication year - 2016
Publication title -
international transactions in operational research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.032
H-Index - 52
eISSN - 1475-3995
pISSN - 0969-6016
DOI - 10.1111/itor.12180
Subject(s) - data envelopment analysis , computer science , set (abstract data type) , order (exchange) , tourism , operations research , empirical research , frontier , industrial engineering , econometrics , mathematical optimization , business , economics , mathematics , statistics , engineering , archaeology , finance , political science , law , history , programming language
In conventional multistage data envelopment analysis (DEA) studies, different hotel types are mostly assumed to be in a single technology set. However, the assumption that a single technology can have multiple operating types has been criticized by researchers in the past. To overcome this shortcoming, the main propose of this study is to establish a combination of the models of Tone and Chen and Zhu in order to assess the efficiencies of two hotel types. We first provide evidence that there is an existing defect arising from the direct combination of the two models. The nonhomogeneous two‐stage model is established to evaluate the operational and expense utilization efficiencies for the Taiwanese international tourist hotels. The results verify that the defect from the direct combination of these models can be corrected by the new model. Empirical evaluation reveals that independent hotels have higher expense utilization efficiency and chain hotels have higher operational efficiency. The evaluation also indicates that tourist hotels can achieve best practices through a mutual learning strategy.