TY - GEN
T1 - Adaptations to Leveled-DEA
T2 - 2023 Portland International Conference on Management of Engineering and Technology, PICMET 2023
AU - Tiruvengadam, Naveen
AU - Elizondo-Noriega, Armando
AU - Guemes-Castorena, David
AU - Aviles-Sacoto, Sonia Valeria
N1 - Publisher Copyright:
© 2023 PICMET (Portland International Center for Management of Engineering and Technology(.
PY - 2023
Y1 - 2023
N2 - Leveled-Data Envelopment Analysis (L-DEA) is an adaptation of traditional Data Envelopment Analysis (DEA) in that it affords a limited comparison of heterogeneous Decision-Making Units (DMUs). In other words, while traditional DEA is intended for comparison and benchmarking of homogeneous DMUs (e.g. comparing hospitals only against other hospitals), LDEA allows for a rather narrow-in-scope comparison of DMUs across industries (hospitals against banks). In theory, L-DEA affords technology managers the ability to build external context against which to evaluate internal efficiencies. The L-DEA method as originally proposed was computed using a parametric method that was computationally slow. This paper provides an applied computational alternative that remains true to mathematics undergirding L-DEA whilst speeding up its computation considerably. In addition, and importantly, L-DEA's ability to remain functional with adaptations and perform against another dataset are both evaluated, in order to explore its capacity for further use.
AB - Leveled-Data Envelopment Analysis (L-DEA) is an adaptation of traditional Data Envelopment Analysis (DEA) in that it affords a limited comparison of heterogeneous Decision-Making Units (DMUs). In other words, while traditional DEA is intended for comparison and benchmarking of homogeneous DMUs (e.g. comparing hospitals only against other hospitals), LDEA allows for a rather narrow-in-scope comparison of DMUs across industries (hospitals against banks). In theory, L-DEA affords technology managers the ability to build external context against which to evaluate internal efficiencies. The L-DEA method as originally proposed was computed using a parametric method that was computationally slow. This paper provides an applied computational alternative that remains true to mathematics undergirding L-DEA whilst speeding up its computation considerably. In addition, and importantly, L-DEA's ability to remain functional with adaptations and perform against another dataset are both evaluated, in order to explore its capacity for further use.
UR - http://www.scopus.com/inward/record.url?scp=85170357882&partnerID=8YFLogxK
U2 - 10.23919/PICMET59654.2023.10216794
DO - 10.23919/PICMET59654.2023.10216794
M3 - Contribución a la conferencia
AN - SCOPUS:85170357882
T3 - 2023 Portland International Conference on Management of Engineering and Technology (PICMET)
BT - PICMET 2023 - Portland International Conference on Management of Engineering and Technology
A2 - Kocaoglu, Dundar F.
A2 - Anderson, Timothy R.
A2 - Kozanoglu, Dilek Cetindamar
A2 - Niwa, Kiyoshi
A2 - Steenhuis, Harm-Jan
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 23 July 2023 through 27 July 2023
ER -