Facet analysis in data envelopment analysis: some pitfalls of the CRS models

Springer Science and Business Media LLC
Ole Bent Olesen1, Niels Christian Petersen1
1Department of Business and Management, The University of Southern Denmark, Odense, Denmark

Tóm tắt

AbstractThe nonparametric estimator of the Extended Facet reference technology for the Constant Returns to Scale case has attracted some attention, because the associated production frontier by construction does only include strongly efficient faces of maximal dimension, or strongly efficient faces that are part of such strongly efficient faces of maximal dimension. The strongly efficient faces of maximal dimension are denoted Full Dimensional Efficient Facets (FDEFs). The identification of such strongly efficient facets is facilitated by removing all inefficient and all strongly efficient but not extreme efficient DMUs from the estimation procedure of the technology set. Any face that i) is passing through the origin and with (s + m − 1) linear independent extreme efficient observed DMUs positioned on it and ii) with a normal vector with strict positive (strict negative) output (input) components, is a FDEF, where s (m) is the number of outputs (inputs). It is, however, not correct that every face that satisfies only i) is a FDEF. We denote a face (a subface) that satisfies only condition i) but not condition ii) for an AP-face (an AP-subface). It is proved that a radial projection of any output input combination in the estimated EXFA technology set is positioned on the strongly efficient frontier if and only if i) no AP-(sub)faces exist, ii) a regulaty condition RC1 is satisfied and only dual multiplier constraints corresponding to extreme efficient DMUs are included in the estimation. A test for the fulfillment of the condition that no AP-(sub)faces exist is provided.

Từ khóa


Tài liệu tham khảo

Aparicio J, Pastor JT (2014) Closest targets and strong monotonicity on the strongly efficient frontier in DEA. Omega 44:51–57

Aparicio J, Pastor JT (2014) On how to properly calculate the Euclidean distance-based measure in DEA. Optimization 63:421–432

Banker RD, Charnes A, Cooper WW (1984) Some models for estimating technical and scale inefficiencies in data envelopment analysis. Manag Sci 30:1078–1092

Barber CB, Dobkin DP, Huhdanpaa H (1996) The Quickhull Algorithm for convex hulls. ACM Trans Math Softw 22:469–483

Charnes A, Cooper WW, Rhodes E (1978) Measuring the efficiency of decision-making units. Eur J Oper Res 2:429–444

Cooper WW, Park KS, Pastor JT (1999) RAM: a range adjusted measure of inefficiency for use with additive models, and relations to other models and measurements in DEA. J Prod Anal 11:5–42

Cooper WW, Seiford LM & Tone K (2000) Data envelopment analysis. A comprehensive text with models, applications, references and DEA-solver software. Kluwer Academic Publisher

Färe R (1988) Fundamentals of production. Springer

Färe R, Grosskopf S & Lovell, CAK (1985) The measurement of efficiency of production. Kluwer-Nijhoff Publication, Boston

Olesen OB, Petersen NC (1996) Indicators of Ill-conditioned data sets and model misspecification in data envelopment analysis: an extended facet approach. Manag Sci 42:205–219

Olesen O B & Petersen NC (2015) Facet analysis in data envelopment analysis. Handbook on DEA J. Zhu (eds)

Pastor JT, Aparicio J & Zofio JL (2022) Benchmarking economic efficiency technical and allocative fundamentals. international series in operations research and management science, Springer, Switzerland

Rockafellar RT (1970) Convex analysis. Princeton University Press, New Jersey

Thông tin
Thông tin xuất bản

Springer Science and Business Media LLC

Thông tin tác giả