TY - JOUR
T1 - Selective measures in data envelopment analysis
AU - Toloo, Mehdi
AU - Barat, Mona
AU - Masoumzadeh, Atefeh
N1 - Funding Information:
The research was supported by the Czech Science Foundation (GACR project 14-31593S) and through European Social Fund within the project CZ.1.07/2.3.00/20.0296.
Publisher Copyright:
© 2014, Springer Science+Business Media New York.
PY - 2014/3
Y1 - 2014/3
N2 - Data envelopment analysis (DEA) is a data based mathematical approach, which handles large numbers of variables, constraints, and data. Hence, data play an important and critical role in DEA. Given a set of decision making units (DMUs) and identified inputs and outputs (performance measures), DEA evaluates each DMU in comparison with all DMUs. According to some statistical and empirical rules, a balance between the number of DMUs and the number of performance measures should exist. However, in some situations the number of performance measures is relatively large in comparison with the number of DMUs. These cases lead us to choose some inputs and outputs in a way that produces acceptable results. We refer to these selected inputs and outputs as selective measures. This paper presents an approach toward a large number of inputs and outputs. Individual DMU and aggregate models are recommended and expanded separately for developing the idea of selective measures. The practical aspect of the new approach is illustrated by two real data set applications.
AB - Data envelopment analysis (DEA) is a data based mathematical approach, which handles large numbers of variables, constraints, and data. Hence, data play an important and critical role in DEA. Given a set of decision making units (DMUs) and identified inputs and outputs (performance measures), DEA evaluates each DMU in comparison with all DMUs. According to some statistical and empirical rules, a balance between the number of DMUs and the number of performance measures should exist. However, in some situations the number of performance measures is relatively large in comparison with the number of DMUs. These cases lead us to choose some inputs and outputs in a way that produces acceptable results. We refer to these selected inputs and outputs as selective measures. This paper presents an approach toward a large number of inputs and outputs. Individual DMU and aggregate models are recommended and expanded separately for developing the idea of selective measures. The practical aspect of the new approach is illustrated by two real data set applications.
KW - Data envelopment analysis
KW - Decision making unit
KW - Efficiency
KW - Selective measures
UR - http://www.scopus.com/inward/record.url?scp=84922970092&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84922970092&partnerID=8YFLogxK
U2 - 10.1007/s10479-014-1714-3
DO - 10.1007/s10479-014-1714-3
M3 - Article
AN - SCOPUS:84922970092
SN - 0254-5330
VL - 226
SP - 623
EP - 642
JO - Annals of Operations Research
JF - Annals of Operations Research
IS - 1
ER -