0% found this document useful (0 votes)

4 views111 pages

Dea R

The 'deaR' package, version 1.4.1, provides functions for conducting both conventional and fuzzy Data Envelopment Analysis (DEA). It is authored by Vicente Coll-Serrano, Vicente Bolos, and Rafael Benitez Suarez, and is available on CRAN under the GPL license. The package includes various documented R topics related to DEA methodologies and applications.

Uploaded by

Rogério Ribeiro

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

4 views111 pages

Dea R

Uploaded by

Rogério Ribeiro

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

Package ‘deaR’

May 2, 2023
Type Package
Title Conventional and Fuzzy Data Envelopment Analysis
Version 1.4.1
Author Vicente Coll-Serrano, Vicente Bolos, Rafael Benitez Suarez <rabesua@[Link]>
Maintainer Vicente Bolos <[Link]@[Link]>
Description Set of functions for Data Envelopment Analysis. It runs both classic and fuzzy DEA mod-
els. See: Banker, R.; Charnes, A.; Cooper, W.W. (1984). <doi:10.1287/mnsc.30.9.1078>, Charnes, A.; Cooper, W.W.; Rhode
2217(78)90138-
8> and Charnes, A.; Cooper, W.W.; Rhodes, E. (1981). <doi:10.1287/mnsc.27.6.668>.
License GPL
Encoding UTF-8
LazyData true
RoxygenNote 7.2.3
Depends R (>= 3.5)
Imports lpSolve, ggplot2, tidyr, plotly, igraph, writexl, dplyr,
methods, gridExtra, grDevices
NeedsCompilation no
Repository CRAN
Date/Publication 2023-05-02 09:20:02 UTC

R topics documented:
Airlines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
bootstrap_basic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Coelli_1998 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Coll_Blasco_2006 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
cross_efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
cross_efficiency_fuzzy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Departments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Doyle_Green_1994 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Economy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

1
2 R topics documented:

EconomyLong . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
efficiencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
[Link] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
efficiencies.dea_fuzzy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
eff_dmus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Electric_plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
extreme_efficient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Fortune500 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Fried1993 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
FuzzyExample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Golany_Roll_1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Grifell_Lovell_1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Guo_Tanaka_2001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Hotels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Hua_Bian_2007 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
[Link] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
[Link] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
is.deadata_fuzzy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
is.dea_fuzzy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
[Link] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Kao_Liu_2003 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
lambdas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Leon2003 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Libraries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Lim_Zhu_2015 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
make_deadata . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
make_deadata_fuzzy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
make_malmquist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
malmquist_index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
maximal_friends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
modelfuzzy_guotanaka . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
modelfuzzy_kaoliu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
modelfuzzy_possibilistic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
model_additive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
model_addmin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
model_addsupereff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
model_basic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
model_deaps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
model_fdh . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
model_multiplier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
model_nonradial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
model_profit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
model_rdm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
model_sbmeff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
model_sbmsupereff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
model_supereff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
multipliers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
PFT1981 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Airlines 3

[Link] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
plot.dea_fuzzy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Power_plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
[Link] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
print.deadata_fuzzy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
read_data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
read_data_fuzzy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
read_malmquist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
references . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
rts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Ruggiero2007 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
slacks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
[Link] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
summary.dea_fuzzy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Supply_Chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
targets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Tone2001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
Tone2003 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
undesirable_basic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108

Index 110

Airlines Data: Coelli, Grifell-Tatje, and Perelman (2002).

Description
Data of 28 airlines with 2 outputs and 4 inputs.

Usage
data("Airlines")

Format
Data frame with 28 rows and 7 columns. Definition of outputs (Y) and inputs (X):

y1 = Pass Passenger-kilometers flown

y2 = Cargo Freight tonne-kilometers flown
x1 = Lab Labor (number of employees)
x2 = Fuel Fuel (millions of gallons)
x3 = Matl Other inputs (millions of U.S. dollar equivalent) consisting of operating and mainte-
nance expenses excluding labor and fuel expenses
x4 = Cap Capital (sum of the maximum takeoff weights of all aircraft flown multiplied by the
number of days flown)
4 bootstrap_basic

Author(s)
Vicente Coll-Serrano (<[Link]@[Link]>). Quantitative Methods for Measuring Culture
(MC2). Applied Economics.
Vicente Bolos (<[Link]@[Link]>). Department of Business Mathematics
Rafael Benitez (<[Link]@[Link]>). Department of Business Mathematics
University of Valencia (Spain)

Source
Coelli, T.; Griffel-Tatje, E.; Perelman, S. (2002). "Capacity Utilization and Profitability: A De-
composition of Short-Run Profit Efficiency", International Journal of Production Economics 79,
261–278.

bootstrap_basic Bootstrapping DEA

Description
To bootstrap efficiency scores, deaR uses the algorithm proposed by Simar and Wilson (1998). For
now, the function bootstrap_basic can only be used with basic DEA models.

Usage
bootstrap_basic(datadea,
orientation = c("io", "oo"),
rts = c("crs", "vrs", "nirs", "ndrs", "grs"),
L = 1,
U = 1,
B = 2000,
h = NULL,
alpha = 0.05)
bootstrap_basic 5

Arguments
datadea A deadata object with n DMUs, m inputs and s outputs.
orientation A string, equal to "io" (input oriented) or "oo" (output oriented).
rts A string, determining the type of returns to scale, equal to "crs" (constant), "vrs"
(variable), "nirs" (non-increasing), "ndrs" (non-decreasing) or "grs" (general-
ized).
L Lower bound for the generalized returns to scale (grs).
U Upper bound for the generalized returns to scale (grs).
B Number of bootstrap iterations.
h Bandwidth of smoothing window. By default h = 0.014 (you can set h equal to
any other value). The optimal bandwidth factor can also be calculated following
the proposals of Silverman (1986) and Daraio y Simar (2007). So, h = "h1"
is the optimal h referred as "robust normal-reference rule" (Daraio and Simar,
2007 p.60), h = "h2" is the value of h1 but instead of the factor 1.06 with the
factor 0.9, h = "h3" is the value of h1 adjusted for scale and sample size (Daraio
and Simar, 2007 p.61), and h = "h4" is the bandwidth provided by a Gaussian
kernel density estimate.
alpha Between 0 and 1 (for confidence intervals).

Author(s)
Vicente Coll-Serrano (<[Link]@[Link]>). Quantitative Methods for Measuring Culture
(MC2). Applied Economics.
Vicente Bolós (<[Link]@[Link]>). Department of Business Mathematics
Rafael Benítez (<[Link]@[Link]>). Department of Business Mathematics
University of Valencia (Spain)

References
Behr, A. (2015). Production and Efficiency Analysis with R. Springer.
Bogetoft, P.; Otto, L. (2010). Benchmarking with DEA, SFA, and R. Springer.
Daraio, C.; Simar, L. (2007). Advanced Robust and Nonparametric Methods in Efficiency Analysis:
Methodology and Applications. New York: Springer.
Färe, R.; Grosskopf, S.; Kokkenlenberg, E. (1989). "Measuring Plant Capacity, Utilization and
Technical Change: A Nonparametric Approach". International Economic Review, 30(3), 655-666.
Löthgren, M.; Tambour, M. (1999). "Bootstrapping the Data Envelopment Analysis Malmquist
Productivity Index". Applied Economics, 31, 417-425.
Silverman, B.W. (1986). Density Estimation for Statistics and Data Analysis. London: Chapman
and Hall.
Simar, L.; Wilson, P.W. (1998). "Sensitivity Analysis of Efficiency Scores: How to Bootstrap in
Nonparametric Frontier Models". Management Science, 44(1), 49-61.
Simar, L.; Wilson, P.W. (1999). "Estimating and Bootstrapping Malmquist Indices". European
Journal of Operational Research, 115, 459-471.
6 Coelli_1998

Simar, L.; Wilson, P.W. (2008). Statistical Inference in Nonparametric Frontier Models: Recent
Developments and Perspective. In H.O. Fried; C.A. Knox Lovell and S.S. Schmidt (eds.) The
Measurement of Productive Efficiency and Productivity Growth. New York: Oxford University
Press. doi:10.1093/acprof:oso/9780195183528.001.0001

Examples
# To replicate the results in Simar y Wilson (1998, p. 58) you have to
# set B=2000 (in the example B = 100 to save time)
data("Electric_plants")
data_example <- make_deadata(Electric_plants,
ni = 3,
no = 1)
result <- bootstrap_basic(datadea = data_example,
orientation = "io",
rts = "vrs",
B = 100)
result$score_bc
result$CI

Coelli_1998 Data: Coelli, Rao and Battese (1998).

Description

Data of five DMUs with two inputs and one output. Prices for inputs are available. Price for output
is not from Coelli et al. (1998).

Usage

data("Coelli_1998")

Format

Data frame with 6 rows and 5 columns. Definition of inputs (X) and outputs (Y):

Input1 Input 1
Input2 Input 2
Output Output
Price_input1 Price input 1
Price_input2 Price input 2
Price_output Price output
Coll_Blasco_2006 7

Source
Coelli, T.; Prasada Rao, D.S.; Battese, G.E. An introduction to efficiency and productivity analysis.
Boston: Kluwer Academic Publishers.

data_example <- make_deadata(Coelli_1998,

dmus = 1,
ni = 2,
no = 1)
result <- model_profit(data_example,
price_input = data_prices,
rts = "crs",
restricted_optimal = FALSE)
# notice that the option by default is restricted_optimal=TRUE
efficiencies(result)

Coll_Blasco_2006 Data: Coll and Blasco (2006).

Description
Data of six authorized dealers with two inputs and two outputs.

Usage
data("Coll_Blasco_2006")
8 cross_efficiency

Format
Data frame with 6 rows and 5 columns. Definition of inputs (X) and outputs (Y):

x1 = Employees Number of employees

x2 = Capital Impairment of assets
y1 = Vehicles Number of vehicles sold
y2 = Orders Number of orders received at the garage

Source
Coll-Serrano, V.; Blasco-Blasco, O. (2006). Evaluacion de la Eficiencia mediante el Análisis En-
volvente de Datos. Introduccion a los Modelos Básicos.

cross_efficiency Cross efficiency analysis

Description
Computes arbitrary, benevolent and aggressive formulations of cross-efficiency under any returns-
to-scale. Doyle and Green (1994) present three alternatives ways of formulating the secondary goal
(wich will minimize or maximize the other DMUs’ cross-efficiencies in some way). Methods II and
III are implemented in deaR with any returns-to-scale. The maverick index is also calculated.
cross_efficiency 9

Usage
cross_efficiency(datadea,
dmu_eval = NULL,
dmu_ref = NULL,
epsilon = 0,
orientation = c("io", "oo"),
rts = c("crs", "vrs", "nirs", "ndrs", "grs"),
L = 1,
U = 1,
selfapp = TRUE,
correction = FALSE,
M2 = TRUE,
M3 = TRUE)

Arguments
datadea An object of class dea or deadata. If it is of class dea it must have been obtained
with some of the multiplier DEA models.
dmu_eval A numeric vector. Only the multipliers of DMUs in dmu_eval are computed. If
NULL (default), all DMUs are considered.
dmu_ref A numeric vector containing which DMUs are the evaluation reference set. If
NULL (default), all DMUs are considered.
epsilon Numeric, multipliers must be >= epsilon.
orientation A string, equal to "io" (input-oriented) or "oo" (output-oriented).
rts A string, determining the type of returns to scale, equal to "crs" (constant), "vrs"
(variable), "nirs" (non-increasing), "ndrs" (non-decreasing) or "grs" (general-
ized).
L Lower bound for the generalized returns to scale (grs).
U Upper bound for the generalized returns to scale (grs).
selfapp Logical. If it is TRUE, self-appraisal is included in the average scores of A and e.
correction Logical. If it is TRUE, a correction is applied in the "vrs" input-oriented model
in order to avoid negative cross-efficiencies, according to Lim & Zhu (2015).
M2 Logical. If it is TRUE, it computes Method II for aggresive/benevolent estima-
tions.
M3 Logical. If it is TRUE, it computes Method III for aggresive/benevolent estima-
tions.

Note
(1) We can obtain negative cross-efficiency in the input-oriented DEA model under no constant
returns-to-scale. However, the same does not happen in the case of the output-oriented VRS DEA
model. For this reason, the proposal of Lim and Zhu (2015) is implemented in deaR to calculate the
input-oriented cross-efficiency model under no constant returns-to-scale.
(2) The multiplier model can have alternate optimal solutions (see note 1 in model_multiplier). So,
depending on the optimal weights selected we can obtain different cross-efficiency scores.
10 cross_efficiency

References
Sexton, T.R., Silkman, R.H.; Hogan, A.J. (1986). Data envelopment analysis: critique and exten-
sions. In: Silkman RH (ed) Measuring efficiency: an assessment of data envelopment analysis, vol
32. Jossey-Bass, San Francisco, pp 73–104. doi:10.1002/ev.1441
Doyle, J.; Green, R. (1994). “Efficiency and cross efficiency in DEA: derivations, meanings and the
uses”, Journal of Operational Research Society, 45(5), 567–578. doi:10.2307/2584392
Cook, W.D.; Zhu, J. (2015). DEA Cross Efficiency. In: Zhu, J. (ed) Data Envelopment Analysis.
A Handbook of Models and Methods. International Series in Operations Research & Management
Science, vol 221. Springer, Boston, MA, 23-43. doi:10.1007/9781489975539_2
Lim, S.; Zhu, J. (2015). "DEA Cross-Efficiency Under Variable Returns to Scale". Journal of
Operational Research Society, 66(3), p. 476-487. doi:10.1057/jors.2014.13

See Also
model_multiplier, cross_efficiency_fuzzy

Examples
# Example 1.
# Arbitrary formulation. Input-oriented model under constant returns-to-scale.
data("Golany_Roll_1989")
data_example <- make_deadata(datadea = Golany_Roll_1989,
inputs = 2:4,
outputs = 5:6)
result <- cross_efficiency(data_example,
orientation = "io",
rts = "crs",
selfapp = TRUE)
result$Arbitrary$cross_eff
result$Arbitrary$e

# Example 2.
# Benevolent formulation (method II). Input-oriented.
data("Golany_Roll_1989")
data_example <- make_deadata(datadea = Golany_Roll_1989,
inputs = 2:4,
outputs = 5:6)
result <- cross_efficiency(data_example,
orientation = "io",
selfapp = TRUE)
result$M2_ben$cross_eff
cross_efficiency_fuzzy 11

result$M2_ben$e

# Example 3.
# Benevolent formulation (method III). Input-oriented.
data("Golany_Roll_1989")
data_example <- make_deadata(datadea = Golany_Roll_1989,
inputs = 2:4,
outputs = 5:6)
result <- cross_efficiency(data_example,
orientation = "io",
selfapp = TRUE)
result$M3_ben$cross_eff
result$M3_ben$e

# Example 4.
# Arbitrary formulation. Output-oriented.
data("Golany_Roll_1989")
data_example <- make_deadata(datadea = Golany_Roll_1989,
inputs = 2:4,
outputs = 5:6)
result <- cross_efficiency(data_example,
orientation = "oo",
selfapp = TRUE)
result$Arbitrary$cross_eff
result$Arbitrary$e

# Example 5.
# Arbitrary formulation. Input-oriented model under vrs returns-to-scale.
data("Lim_Zhu_2015")
data_example <- make_deadata(Lim_Zhu_2015,
ni = 1,
no = 5)
cross <- cross_efficiency(data_example,
epsilon = 0,
orientation = "io",
rts = "vrs",
selfapp = TRUE,
M2 = FALSE,
M3 = FALSE)
cross$Arbitrary$e

cross_efficiency_fuzzy
Cross efficiency fuzzy analysis

Description
Computes the cross-efficiency fuzzy tables from DEA fuzzy data or a Guo-Tanaka DEA model
solution. The (crisp) relative efficiencies for the case h = 1 are obtained from the CCR model
(model_multiplier).
12 cross_efficiency_fuzzy

Usage
cross_efficiency_fuzzy(datadea,
orientation = c("io", "oo"),
h = 1,
selfapp = TRUE)

Arguments
datadea An object of class dea_fuzzy or deadata_fuzzy. If it is of class dea_fuzzy it
must have been obtained with modelfuzzy_guotanaka.
orientation A string, equal to "io" (input-oriented) or "oo" (output-oriented).
h A numeric vector with the h-levels (in [0,1]).
selfapp Logical. If it is TRUE, self-appraisal is included in the average scores of A and e.

References
Doyle, J.; Green, R. (1994). “Efficiency and Cross Efficiency in DEA: Derivations, Meanings and
the Uses”, Journal of Operational Research Society, 45(5), 567–578. doi:10.2307/2584392
Guo, P.; Tanaka, H. (2001). "Fuzzy DEA: A Perceptual Evaluation Method", Fuzzy Sets and Sys-
tems, 119, 149–160. doi:10.1016/S01650114(99)001062
León, T.; Liern, V.; Ruiz, J.L.; Sirvent, I. (2003). "A Fuzzy Mathematical Programming Ap-
proach to the assessment of efficiency with DEA Models", Fuzzy Sets Systems, 139(2), 407–419.
doi:10.1016/S01650114(02)006085
Sexton, T.R., Silkman, R.H.; Hogan, A.J. (1986). Data envelopment analysis: critique and exten-
sions. In: Silkman RH (ed) Measuring efficiency: an assessment of data envelopment analysis, vol
32. Jossey-Bass, San Francisco, pp 73–104. doi:10.1002/ev.1441

Examples
data("Guo_Tanaka_2001")
datadea <- make_deadata_fuzzy(datadea = Guo_Tanaka_2001,
[Link] = 2:3,
[Link] = 4:5,
[Link] = 6:7,
[Link] = 8:9)
result <- cross_efficiency_fuzzy(datadea = datadea,
h = seq(0, 1, 0.2))
Departments 13

Departments Data: Tomkins and Green (1988).

Description
Data from 20 University accounting departments in the UK.

Usage
data("Departments")

Format
Data frame with 20 rows and 11 columns. Definition of inputs (X) and outputs (Y):

x1 = Staff Average Full Time Academic Staff 82/3-84/5)

x2 = Salaries 1984-5 Salaries Academics and Related (in pounds))
x3 = Other_Exp 1984-5 Other Expenses (in pounds)
y1 = Undergrad Average Number Undergraduates 82/3-84/5
y2 = Research_post Research Postgraduates
y3 = Taught_post Taught Postgraduates
y4 = Res_co_income Research council income (in pounds)
y5 = Other_res_income Other research income (in pounds)
y6 = Other_income Other income (in pounds)
y7 = Publications Number of publications

Source
Tomkins, C.; Green, R. (1988). "An Experiment in the Use of Data Envelopment Analysis for
Evaluating the Efficiency of UK University Departments of Accounting", Financial Accountability
and Management, 4(2), 147-164. doi:10.1111/j.14680408.1988.tb00296.x

See Also
make_deadata, model_basic
14 Doyle_Green_1994

Examples
# Example.
# Replication of results DEA1 in Tomkins and Green (1988)
data("Departments")
# Calculate Total income
Departments$Total_income <- Departments[, 5] + Departments[, 6] + Departments[, 7]
data_example <- make_deadata(Departments,
inputs = 9,
outputs = c(2, 3, 4, 12))
result <- model_basic(data_example,
orientation = "io",
rts = "crs")
efficiencies(result) # Table 3 (p.156)
references(result) # Table 3 (p.157)

Doyle_Green_1994 Data: Doyle and Green (1994).

Description
Data adapted from Tomkins and Green (1988). 13 DMUs using 3 inputs to produce 2 outputs.

Usage
data("Doyle_Green_1994")

Format
Data frame with 13 rows and 6 columns. Definition of inputs (X) and outputs (Y):

y1 = Undergraduate Number of undergraduates

y2 = Postgraduates Number of postgraduates (taught and research)
y3 = Research_income Research and other income
y4 = Publications Number of publications
x1 = Salaries Salaries of academic and related staff
x2 = Other_expenses Other expenses

Source
Doyle, J.; Green, R. (1994). “Efficiency and cross efficiency in DEA: derivations, meanings and the
uses”, Journal of Operational Research Society, 45(5), 567–578. doi:10.2307/2584392

See Also
make_deadata, model_multiplier, cross_efficiency

Examples
# Example.
data("Doyle_Green_1994")
data_example <- make_deadata(datadea = Doyle_Green_1994,
dmus = 1,
inputs = 6:7,
outputs = 2:5)
result <- cross_efficiency(data_example,
orientation = "io",
selfapp = TRUE)
result$Arbitrary$cross_eff
result$Arbitrary$e
# Aggressive using method II
result$M2_agg$cross_eff
# Aggressive using method III
result$M3_agg$cross_eff

Economy Data: Wang and Lan (2011).

Description
Data of the industrial economy of China in 2005-2009 (data in wide format).

Usage
data("Economy")

Format
Data frame with 31 rows and 16 columns. Definition of inputs (X) and outputs (Y):

x1 = Capital Total assets (in 100 million RMB)

x2 = Labor Annual average employed persons (in 10000 persons)
y1 = GIOV Gross industrial output value (in 100 million RMB)
16 EconomyLong

Source
Wang, Y.; Lan, Y. (2011). "Measuring Malmquist Productiviy Index: A New Approach Based on
Double Frontiers Data Envelopment Analysis". Mathematical and Computer Modelling, 54, 2760-
2771. doi:10.1016/[Link].2011.06.064

See Also
make_malmquist, malmquist_index

Examples
# Example . Data in wide format.
# Replication of results in Wang and Lan (2011, p. 2768)
data("Economy")
data_example <- make_malmquist(Economy,
nper = 5,
arrangement = "horizontal",
ni = 2,
no = 1)
result <- malmquist_index(data_example)

EconomyLong Data: Wang and Lan (2011).

Description
Data of the industrial economy of China in 2005-2009 (data in long format).

Usage
data("EconomyLong")

Format
Data frame with 155 rows and 5 columns. Definition of inputs (X) and outputs (Y):
x1 = Capital Total assets (in 100 million RMB)
x2 = Labor Annual average employed persons (in 10000 persons)
y1 = GIOV Gross industrial output value (in 100 million RMB)
efficiencies 17

See Also
make_malmquist, malmquist_index

Examples
# Example. Data in long format.
# Replication of results in Wang and Lan (2011, p. 2768)
data("EconomyLong")
data_example <- make_malmquist(EconomyLong,
percol = 2,
arrangement = "vertical",
ni = 2,
no = 1)
result <- malmquist_index(data_example)

efficiencies Objective scores

Description
Extract the scores (optimal objective values) of the evaluated DMUs from a conventional, fuzzy or
stochastic DEA solution. Note that these scores may not always be interpreted as efficiencies.

Usage
efficiencies(x, ...)

Arguments
x An object of class dea, dea_fuzzy or dea_stoch.
... ignored.
18 [Link]

[Link] Objective scores

Description
Extract the scores (optimal objective values) of the evaluated DMUs from a conventional DEA
solution. Note that these scores may not always be interpreted as efficiencies.

Usage
## S3 method for class 'dea'
efficiencies(x, ...)

Arguments
x Object of class dea obtained with some of the conventional DEA model_* func-
tions.
... Other options (for compatibility reasons).

References
Tomkins, C.; Green, R. (1988). “An Experiment in the Use of Data Envelopment Analysis for
Evaluating the Efficiency of UK University Departments of Accounting”. Financial Accountability
and Management 4(2): 147.

Examples
# Replication results model DEA1 in Tomkins and Green (1988)
data("Departments")
# Calculate Total income
Departments$Total_income <- Departments[, 5] + Departments[, 6] + Departments[, 7]
data_DEA1 <- make_deadata(Departments,
inputs = 9,
outputs = c(2, 3, 4, 12))
result <- model_basic(data_DEA1,
orientation = "io",
rts = "crs")
efficiencies(result) # Table 3 (p.156)
efficiencies.dea_fuzzy 19

efficiencies.dea_fuzzy
Efficiencies

Description
Extract the scores (optimal objective values) of the evaluated DMUs from a fuzzy DEA solution.
Note that these scores may not always be interpreted as efficiencies.

Usage
## S3 method for class 'dea_fuzzy'
efficiencies(x, ...)

Arguments
x Object of class dea_fuzzy obtained with some of the fuzzy DEA modelfuzzy_*
functions.
... Other options (for compatibility).

Examples
# Replication of results in Boscá, Liern, Sala and Martínez (2011, p.125)
data("Leon2003")
data_example <- make_deadata_fuzzy(datadea = Leon2003,
[Link] = 2,
[Link] = 3,
[Link] = 4,
[Link] = 5)
result <- modelfuzzy_kaoliu(data_example,
kaoliu_modelname = "basic",
alpha = seq(0, 1, by = 0.1),
orientation = "io",
20 eff_dmus

rts = "vrs")
efficiencies(result)

eff_dmus Efficient DMUs.

Description
Returns the efficient DMUs evaluated in a dea class object.

Usage
eff_dmus(deasol, tol = 1e-04)

Arguments
deasol An object of class "dea" obtained by a DEA model function.
tol Numeric. Absolute tolerance for numeric comparisons in efficiency scores. By
default, it is 1e-4.

Value
A numeric vector containing which DMUs has been evaluated as efficient. This vector is empty if
there is not any efficient DMU.

Note
If maxslack is FALSE, the slacks computed in the first stage are supposed to be the max slacks.

Examples
dataFortune <- make_deadata(Fortune500,
ni = 3,
no = 2)
ccrFortune <- model_basic(dataFortune)
eff_dmus(ccrFortune)
Electric_plants 21

Electric_plants Data: Färe, Grosskopf and Kokkelenberg (1989).

Description

Data of 19 coal-fired steam-electric generating plants operating in Illinois in 1978. Each plant uses
3 inputs to produce 1 output.

Usage

data("Electric_plants")

Format

Data frame with 18 rows and 5 columns. Definition of inputs (X) and outputs (Y):

x1 = Labor Labor average annual employment

x2 = Fuel Fuel 1010 Btu
x3 = Capital Capital MW (fixed input)
y1 = Output Output 106 Kwh

Author(s)

Vicente Coll-Serrano (<[Link]@[Link]>). Quantitative Methods for Measuring Culture

(MC2). Applied Economics.
Vicente Bolos (<[Link]@[Link]>). Department of Business Mathematics
Rafael Benitez (<[Link]@[Link]>). Department of Business Mathematics
University of Valencia (Spain)

Source

Färe, R.; Grosskopf, S.; Kokkenlenberg, E. (1989). "Measuring Plant Capacity, Utilization and
Technical Change: A Nonparametric Approach". International Economic Review, 30(3), 655-666.
Simar, L.; Wilson, P.W. (1998). "Sensitivity Analysis of Efficiency Scores: How to Bootstrap in
Nonparametric Frontier Models". Management Science, 44(1), 49-61.

See Also

make_deadata, model_basic
22 extreme_efficient

Examples
# Example. Replication of results in Simar and Wilson (1998, p.59)
data("Electric_plants")
data_example <- make_deadata(Electric_plants,
dmus = 1,
ni = 3,
no = 1)
result <- model_basic(data_example,
orientation = "io",
rts = "vrs")
efficiencies(result)

extreme_efficient Extreme efficient DMUs

Description
Find a set of extreme efficient DMUs from a deadata object.

Usage
extreme_efficient(datadea,
dmu_ref = NULL,
rts = c("crs", "vrs", "nirs", "ndrs"),
tol = 1e-6)

Arguments
datadea A deadata object with n DMUs, m inputs and s outputs.
dmu_ref A numeric vector containing which DMUs are the evaluation reference set, i.e.
the cluster of DMUs from which we want to find a extreme efficient DMUs
subset. If NULL (default), all DMUs are considered.
rts A string, determining the type of returns to scale, equal to "crs" (constant), "vrs"
(variable), "nirs" (non-increasing) or "ndrs" (non-decreasing).
tol Numeric, a tolerance margin for checking efficiency. It is 1e-6 by default.

Value
A numeric vector representing a extreme efficient subset of DMUs.

References
Charnes, A.; Cooper, W.W.; Thrall, R.M. (1991) "A structure for classifying and characterizing effi-
ciency and inefficiency in data envelopment analysis", Journal of Productivity Analisys, 2, 197–237.

Examples
data("PFT1981")
datadea <- make_deadata(PFT1981,
ni = 5,
no = 3)
# We find a extreme efficient subset from a cluster formed by the first 20 DMUs
result <- extreme_efficient(datadea = datadea,
dmu_ref = 1:20)

Fortune500 Data: Zhu (2014).

Description
This dataset consists of 15 firms from the Fortune 500 list 1995 ([Link]
fortune500/) with 3 inputs and 2 outputs.

Usage
data("Fortune500")

Format
Data frame with 15 rows and 6 columns. Definition of inputs (X) and outputs (Y):

x1 = Assets Assets (millions of dollars)

x2 = Equity Equity (millions of dollars)
x3 = Employees Number of employees
y1 = Revenue Revenue (millions of dollars)
y2 = Profit Profit (millions of dollars)

Source

Zhu, J. (2014). Quantitative Models for Performance Evaluation and Benchmarking. Data Envel-
opment Analysis with Spreadsheets. 3rd Edition Springer, New York. doi:10.1007/9783319066479

See Also

make_deadata, model_multiplier

Examples
data("Fortune500")
data_Fortune <- make_deadata(datadea = Fortune500,
dmus = 1,
inputs = 2:4,
outputs = 5:6)
result <- model_multiplier(data_Fortune,
epsilon = 1e-6,
orientation = "io",
rts = "crs")
# results for General Motors and Ford Motor are not shown
# by deaR because the solution is infeasible
efficiencies(result)
multipliers(result)

Fried1993 Data: Fried, Knox Lovell and Schmidt (1993).

Description

Data of 11 DMUs with two inputs and one output.

Usage

data("Fried1993")

Format

Data frame with 11 rows and 4 columns. Definition of inputs (X) and outputs (Y):

x1 Input 1
x2 Input 2
y1 Output 1
FuzzyExample 25

Author(s)

Vicente Coll-Serrano (<[Link]@[Link]>). Quantitative Methods for Measuring Culture

(MC2). Applied Economics.
Vicente Bolos (<[Link]@[Link]>). Department of Business Mathematics
Rafael Benitez (<[Link]@[Link]>). Department of Business Mathematics
University of Valencia (Spain)

Source

Ali, A.I.; Seiford, L.M. (1993). The Mathematical Programming Approach to Efficiency Analysis.
In Fried, H.O.; Knox Lovell, C.A.; Schmidt, S.S.(eds.), The Measurement of Productive Efficiency.
Techniques and Applications. New York: Oxford University Press.

See Also

make_deadata, model_basic

Examples
# Example. Replication of results in Ali and (1993, p.143).
data("Fried1993")
data_example <- make_deadata(Fried1993,
ni = 2,
no = 1)
result <- model_basic(data_example,
orientation = "oo",
rts = "vrs")
efficiencies(result)
targets(result)

FuzzyExample Data: Fuzzy data reading example.

Description

Synthetic dataset of 5 DMUs with 3 inputs and 3 outputs containing fuzzy and crisp data.

Usage

data("FuzzyExample")
26 FuzzyExample

Format
Data frame with 5 rows and 15 columns.
DMU DMU names
[Link] First Input (crisp numbers)
[Link] Second Input (left centers)
[Link] Second Input (right centers)
[Link] Second Input (left radii)
[Link] Second Input (right radii)
[Link] Third Input (centers)
[Link] Third Input (radii)
[Link] First Output (crisp numbers)
[Link] Second Output (left centers)
[Link] Second Output (right centers)
[Link] Second Output (radii)
[Link] Third Output (centers)
[Link] Third Output (left radii)
[Link] Third Output (right radii)

# Example. Reading the data.

data("FuzzyExample")
datafuzzy <- make_deadata_fuzzy(FuzzyExample,
[Link] = c(2, 3, 7),
[Link] = c(NA, 4, NA),
[Link] = c(NA, 5, 8),
[Link] = c(NA, 6, NA),
[Link] = c(9, 10 , 13),
[Link] = c(NA, 11, NA),
[Link] = c(NA, 12, 14),
[Link] = c(NA, NA, 15))
Golany_Roll_1989 27

Golany_Roll_1989 Data: Golany and Roll (1989).

Description
Data of 13 DMUs using 3 inputs to produce 2 outputs.

Usage
data("Golany_Roll_1989")

Format
Data frame with 13 rows and 6 columns. Definition of inputs (X) and outputs (Y):

x1 Input 1
x2 Input 2
x3 Input 3
y1 Output 1
y1 Output 2

Source
Golany, B.; Roll, Y. (1989). "An Application Procedure for DEA". Omega, International Journal of
Management Science, 17(3), 237-250. doi:10.1016/03050483(89)900297

See Also
make_deadata, model_multiplier, cross_efficiency

Examples
# Example.
data("Golany_Roll_1989")
data_example <- make_deadata(datadea = Golany_Roll_1989,
dmus = 1,
inputs = 2:4,
outputs = 5:6)
28 Grifell_Lovell_1999

result <- cross_efficiency(data_example,

orientation = "io",
selfapp = TRUE)
result$Arbitrary$cross_eff
result$Arbitrary$e

Grifell_Lovell_1999 Data: Grifell-Tatjé and Lovell (1999).

Description

Data of 8 DMUs producing 1 output (Y) by using 1 input (X) for two periods of time.

Usage

data("Grifell_Lovell_1999")

Format

Data frame with 16 rows and 4 columns. Definition of inputs (X) and outputs (Y):

X Input
Y Output

Author(s)

Vicente Coll-Serrano (<[Link]@[Link]>). Quantitative Methods for Measuring Culture

(MC2). Applied Economics.
Vicente Bolos (<[Link]@[Link]>). Department of Business Mathematics
Rafael Benitez (<[Link]@[Link]>). Department of Business Mathematics
University of Valencia (Spain)

Source

Grifell-Tatjé, E.; Lovel, C.A.K. (1999). "A Generalized Malmquist productivity index". Top, 7(1),
81-101.

See Also

make_malmquist, malmquist_index
Guo_Tanaka_2001 29

Examples

# Example. Replication of results in Grifell-Tatjé and Lovell (1999, p. 100).

data("Grifell_Lovell_1999")
data_example <- make_malmquist(Grifell_Lovell_1999,
percol = 1,
dmus = 2,
inputs = 3,
outputs = 4,
arrangement = "vertical")

result_fgnz <- malmquist_index(data_example,

orientation = "oo",
rts = "vrs",
type1 = "cont",
type2 = "fgnz")

result_fgnz$mi

Guo_Tanaka_2001 Data: Guo and Tanaka (2001).

Description
Data of 5 DMUs with two symmetric triangular fuzzy inputs, Xj = (xj, alphaj), and two symmetric
triangular fuzzy outputs, Yj = (yj, betaj).

Usage
data("Guo_Tanaka_2001")

Format
Data frame with 5 rows and 9 columns. Definition of fuzzy inputs (X) and fuzzy outputs (Y):

x1 Input 1
x2 Input 2
alpha1 spread vector Input 1
alpha2 spread vector Input 2
y1 Output 1
y2 Output 2
beta1 spread vector Output 1
beta2 spread vector Output 2
30 Hotels

Author(s)

Vicente Coll-Serrano (<[Link]@[Link]>). Quantitative Methods for Measuring Culture

(MC2). Applied Economics.
Vicente Bolos (<[Link]@[Link]>). Department of Business Mathematics
Rafael Benitez (<[Link]@[Link]>). Department of Business Mathematics
University of Valencia (Spain)

Source

Guo, P.; Tanaka, H. (2001). "Fuzzy DEA: A Perceptual Evaluation Method", Fuzzy Sets and Sys-
tems, 119, 149–160. doi:10.1016/S01650114(99)001062

See Also

make_deadata_fuzzy, modelfuzzy_guotanaka, cross_efficiency_fuzzy

Examples
data("Guo_Tanaka_2001")
data_example <- make_deadata_fuzzy(Guo_Tanaka_2001,
dmus = 1,
[Link] = 2:3,
[Link] = 4:5,
[Link] = 6:7,
[Link] = 8:9)
result <- modelfuzzy_guotanaka(data_example,
h = seq(0, 1, by = 0.1),
orientation = "io")
efficiencies(result)

Hotels Data: Wu, Tsai and Zhou (2011).

Description

This dataset consists of 23 four- and five-plum ITHs in Taipei in 2006. Authors consider 4 inputs
and 3 outputs.

Usage

data("Hotels")
Hotels 31

Format

Data frame with 23 rows and 8 columns. Definition of inputs (X) and outputs (Y):

x1 = Employees Total number of employees)

x2 = Guest_rooms Total number of guest rooms)
x3 = Area_F&B Total area of F&B departments (in 36 square-feet)
x4 = Operating_cost Total operating cost (in NT$)
y1 = Room_revenue Room revenues (in NT$)
y2 = F&B_revenue F&B revenues (in NT$)
y3 = Other_revenue Other revenues (in NT$)

Author(s)

Vicente Coll-Serrano (<[Link]@[Link]>). Quantitative Methods for Measuring Culture

(MC2). Applied Economics.
Vicente Bolos (<[Link]@[Link]>). Department of Business Mathematics
Rafael Benitez (<[Link]@[Link]>). Department of Business Mathematics
University of Valencia (Spain)

Source

Wu, J.; Tsai, H. and Zhou, Z. (2011). "Improving efficiency in International tourist hotels in Taipei
using a non-radial DEA mode", Internationl Journal of Contemporary Hospitality Management,
23(1), 66-83. doi:10.1108/09596111111101670

See Also

make_deadata, model_nonradial

Examples
# Example. Replication of results in Wu,Tsai and Zhou (2011)
data("Hotels")
data_hotels <- make_deadata(Hotels,
dmus = 1,
inputs = 2:5,
outputs = 6:8)
result <- model_nonradial(data_hotels,
orientation = "oo",
rts = "vrs")
efficiencies(result)
32 Hua_Bian_2007

Hua_Bian_2007 Data: Hua and Bian (2007).

Description

Data of 30 DMUs with two desirable inputs, two desirable outputs and one udesirable output.

Usage

data("Hua_Bian_2007")

Format

Data frame with 30 rows and 6 columns. Definition of inputs (X) and outputs (Y):

x1 = D-Input1 Desirable Input 1

x2 = D-Input2 Desirable Input 2
y1 = D-Output1 Desirable Output 1
y2 = D-Output2 Desirable Output 2
y3 = UD-Output1 Undesirable Output 1

Author(s)

Vicente Coll-Serrano (<[Link]@[Link]>). Quantitative Methods for Measuring Culture

(MC2). Applied Economics.
Vicente Bolos (<[Link]@[Link]>). Department of Business Mathematics
Rafael Benitez (<[Link]@[Link]>). Department of Business Mathematics
University of Valencia (Spain)

Source

Hua Z.; Bian Y. (2007). DEA with Undesirable Factors. In: Zhu J., Cook W.D. (eds) Modeling
Data Irregularities and Structural Complexities in Data Envelopment Analysis. Springer, Boston,
MA. doi:10.1007/9780387716077_6

See Also

make_deadata, model_basic
[Link] 33

Examples
# Example. Replication of results in Hua and Bian (2007).
data("Hua_Bian_2007")
# The third output is an undesirable output
data_example <- make_deadata(Hua_Bian_2007,
ni = 2,
no = 3,
ud_outputs = 3)

# Translation parameter (vtrans_o) is set to 1500

result <- model_basic(data_example,
orientation = "oo",
rts = "vrs",
vtrans_o = 1500)
eff <- efficiencies(result)
1 / eff # results M5 in Table 6-5 (p.119)

[Link] dea class check.

Description
Checks whether an R object is of dea class or not.

Usage
[Link](x)

Arguments
x Any R object.

Value
Returns TRUE if its argument is a dea object (that is, has "dea" amongst its classes) and FALSE
otherwise.

[Link] deadata class check.

Description
Checks whether an R object is of deadata class or not.

Usage
[Link](x)

Arguments
x Any R object.

Value
Returns TRUE if its argument is a deadata object (that is, has "deadata" amongst its classes) and
FALSE otherwise.

is.deadata_fuzzy deadata_fuzzy class check.

Description
Checks whether an R object is of deadata_fuzzy class or not.

Usage
is.deadata_fuzzy(x)

Arguments
x Any R object.
is.dea_fuzzy 35

Value
Returns TRUE if its argument is a deadata_fuzzy object (that is, has "deadata_fuzzy" amongst its
classes) and FALSE otherwise.

is.dea_fuzzy dea_fuzzy class check.

Description
Checks whether an R object is of dea_fuzzy class or not.

Usage
is.dea_fuzzy(x)

Arguments
x Any R object.

Value
Returns TRUE if its argument is a dea_fuzzy object (that is, has "dea_fuzzy" amongst its classes) and
FALSE otherwise.

[Link] Friends check.

Description
Checks whether a subset of DMUs is friends or not, according to Tone (2010).

Usage
[Link](datadea,
dmu_eval = NULL,
dmu_ref = NULL,
rts = c("crs", "vrs", "nirs", "ndrs"),
tol = 1e-6)

Arguments
datadea The data, including n DMUs, m inputs and s outputs.
dmu_eval A numeric vector containing the subset of DMUs to be checked. If NULL (de-
fault), all DMUs are considered.
dmu_ref A numeric vector containing which DMUs are the evaluation reference set. If
NULL (default), all DMUs are considered.
rts A string, determining the type of returns to scale, equal to "crs" (constant), "vrs"
(variable), "nirs" (non-increasing) or "ndrs" (non-decreasing).
tol Numeric, a tolerance margin for checking efficiency. It is 1e-6 by default.

Value
Returns TRUE if dmu_eval is friends of dmu_ref, and FALSE otherwise.

References
Tone, K. (2010). "Variations on the theme of slacks-based measure of efficiency in DEA", European
Journal of Operational Research, 200, 901-907. doi:10.1016/[Link].2009.01.027

See Also
maximal_friends, model_sbmeff
Kao_Liu_2003 37

Examples
data("PFT1981")
datadea <- make_deadata(PFT1981,
ni = 5,
no = 3)
subset1 <- c(15, 16, 17, 19) # Subset of DMUs to be checked
result1 <- [Link](datadea = datadea,
dmu_eval = subset1,
dmu_ref = 1:20) # We only consider a cluster formed by the first 20 DMUs
subset2 <- c(15, 16, 17, 20) # Another subset of DMUs to be checked
result2 <- [Link](datadea = datadea,
dmu_eval = subset2,
dmu_ref = 1:20) # We only consider a cluster formed by the first 20 DMUs

Kao_Liu_2003 Data: Kao and Liu (2003).

Description
Data of 24 university libraries in Taiwan with one input and five outputs.

Usage
data("Kao_Liu_2003")

Format
Data frame with 24 rows and 11 columns. Definition of fuzzy inputs (X) and fuzzy outputs (Y):

x1 = Patronage It is a weighted sum of the standardized scores of faculty, graduate students, un-
dergraduate students, and extension students in the range of 0 and 1.
y1 = Collections Books, serials, microforms, audiovisual works, and database.
y2 = Personnel Classified staff, unclassified staff, and student assistants.
y3 = Expenditures Capital expenditure, operating expenditure, and special expenditure.
y4 = Buildings Area and seats
y5 = Services Operating hours, attendance, circulation, communication channels, range of ser-
vices, amount of services, etc.
beta3_l lower spread vector Expenditures
beta3_u upper spread vector Expenditures
beta5_l lower spread vector Services
beta5_u upper spread vector Services
38 lambdas

Note
There are three observations that are missing: expenditures of Library 24 and services of Library
22 and Library 23. Kao and Liu (2000b) represent the expenditures of Library 24 by the triangular
fuzzy number Y = (0.11; 0.41; 1.0). The services of Library 22 and Library 23 are expressed by a
same triangular fuzzy number Y = (0.41; 0.69; 1.0).

Source
Kao, C., Liu, S.T. (2003). “A mathematical programming approach to fuzzy efficiency ranking”,
International Journal of Production Economics, 85. doi:10.1016/S09255273(03)000264

See Also
make_deadata_fuzzy, model_basic

Examples
# Example. Replication of results in Kao and Liu (2003, p.152)
data_example <- make_deadata_fuzzy(Kao_Liu_2003,
dmus = 1,
[Link] = 2,
[Link] = 3:7,
[Link] = c(NA, NA, 8, NA, 10),
[Link] = c(NA, NA, 9, NA, 11))
result <- modelfuzzy_kaoliu(data_example,
kaoliu_modelname = "basic",
orientation = "oo",
rts = "vrs",
alpha = 0)
eff <- efficiencies(result)
eff

lambdas Lambdas

Description
Extract the lambdas of the DMUs from a dea or dea_fuzzy solution.
Leon2003 39

Usage
lambdas(deasol)

Arguments
deasol Object of class dea or dea_fuzzy obtained with some of the DEA model func-
tions.

Examples
data("Coll_Blasco_2006")
data_example <- make_deadata(Coll_Blasco_2006,
ni = 2,
no = 2)
result <- model_multiplier(data_example,
orientation = "io",
rts = "crs")
lambdas(result)

Leon2003 Data: Leon, Liern, Ruiz and Sirvent (2003).

Description
Data of 8 DMUs with one symmetric triangular fuzzy inputs: Xj = (xj, alphaj), and one symmetric
triangular fuzzy outputs: Yj = (yj, betaj).

Usage
data("Leon2003")

Format
Data frame with 8 rows and 5 columns. Definition of fuzzy inputs (X) and fuzzy outputs (Y):
x1 Input 1
alpha1 spread vector Input 1
y1 Output 1
beta1 spread vector Output 1
40 Libraries

Source
Leon, T.; Liern, V. Ruiz, J.; Sirvent, I. (2003). "A Possibilistic Programming Approach to the
Assessment of Efficiency with DEA Models", Fuzzy Sets and Systems, 139, 407–419. doi:10.1016/
S01650114(02)006085

See Also
make_deadata_fuzzy, modelfuzzy_possibilistic, cross_efficiency_fuzzy, modelfuzzy_guotanaka

Examples
# Example. Replication of results in Leon et. al (2003, p. 416)
data("Leon2003")
data_example <- make_deadata_fuzzy(Leon2003,
dmus = 1,
[Link] = 2,
[Link] = 3,
[Link] = 4,
[Link] = 5)
result <- modelfuzzy_possibilistic(data_example,
h = seq(0, 1, by = 0.1),
orientation = "io",
rts = "vrs")
efficiencies(result)

Libraries Data: Cooper, Seiford and Tone (2007).

Description
Data for 23 public libraries of the Tokyo Metropolitan Area in 1986.

Usage
data("Libraries")
Libraries 41

Format
Data frame with 23 rows and 7 columns. Definition of inputs (X) and outputs (Y):

x1 = AREA Floor area (unit=1000 m2)

x2 = BOOKS Number of books (unit=1000)
x3 = STAFF Staff
x4 = POPULATION Population (unit=1000)
y1 = REGISTERED Registered residents (unit=1000)
y2 = BORROWED Borrowed books (unit=1000)

Source
Cooper, W.W.; Seiford, L.M. and Tone, K. (2007). Data Envelopment Analysis. A Comprehensive
Text with Models, Applications, References and DEA-Solver Software. Springer.

# Example 2. Non-discretionary input (POPULATION).

data(Libraries)
# POPULATION (non-controllable input) is the 4th input.
data_example2 <- make_deadata(Libraries,
42 Lim_Zhu_2015

dmus=1,
inputs=2:5,
nd_inputs=4,
outputs=6:7)
result2 <- model_basic(data_example2,
orientation="io",
rts="crs")
efficiencies(result2)
targets(result2)

Lim_Zhu_2015 Data: Lim and Zhu (2015).

Description
Data of 37 R&D project proposal relating to the Turkish iron and steel industry. Authors consider
one input and five outputs.

Usage
data("Lim_Zhu_2015")

Format
Data frame with 37 rows and 7 columns. Definition of inputs (X) and outputs (Y):

x1 = Budget Budget
y1 = Indirect_economic Indirect economic contribution
y2 = Direct_economic Direct economic contribution
y3 = Technical Technical contribution
y4 = Social Social contribution
y5 = Scientific Scientific contribution

Source
Lim, S.; Zhu, J. (2015). "DEA Cross-Efficiency Under Variable Returns to Scale". Journal of
Operational Research Society, 66(3), p. 476-487. doi:10.1057/jors.2014.13
make_deadata 43

See Also

make_deadata, model_multiplier, cross_efficiency

Examples
# Example. Arbitrary formulation.
# Input-oriented model under variable returns-to-scale.
data("Lim_Zhu_2015")
data_example <- make_deadata(Lim_Zhu_2015,
dmus = 1,
ni = 1,
no = 5)
cross <- cross_efficiency(data_example,
epsilon = 0,
orientation = "io",
rts = "vrs",
selfapp = TRUE,
M2 = FALSE,
M3 = FALSE)
cross$Arbitrary$e

make_deadata make_deadata

Description

This function creates, from a data frame, a deadata structure, which is as list with fields input,
output, dmunames, nc_inputs, nc_outputs, nd_inputs, nd_outputs.

Usage

make_deadata(datadea = NULL,
ni = NULL,
no = NULL,
dmus = 1,
inputs = NULL,
outputs = NULL,
nc_inputs = NULL,
nc_outputs = NULL,
nd_inputs = NULL,
nd_outputs = NULL,
ud_inputs = NULL,
ud_outputs = NULL)
44 make_deadata

Arguments
datadea Data frame with DEA data.
ni Number of inputs, if inputs are in columns 2:(ni + 1) (if DMUs are in the first
column) or 1:ni (no DMUs column).
no Number of outputs, if outputs are in columns (ni + 2):(ni + no + 1) (if DMUs are
in the first column) or (ni + 1):(ni + no) (no DMUs column). If not specified,
DMUs are in the first column.
dmus Column (number or name) of DMUs (optional). By default, it is the first column.
If there is not any DMU column, then it must be NULL.
inputs Columns (numbers or names) of inputs (optional). It prevails over ni. Alterna-
tively to datadea, it can be a matrix with the inputs (DMUs in columns). In this
case, DMUs names are taken from the columns names.
outputs Columns (numbers or names) of outputs (optional). It prevails over no. Alter-
natively to datadea, it can be a matrix with the outputs (DMUs in columns).
nc_inputs A numeric vector containing the indices of non-controllable inputs.
nc_outputs A numeric vector containing the indices of non-controllable outputs.
nd_inputs A numeric vector containing the indices of non-discretionary inputs.
nd_outputs A numeric vector containing the indices of non-discretionary outputs.
ud_inputs A numeric vector containing the indices of undesirable (good) inputs.
ud_outputs A numeric vector containing the indices of undesirable (bad) outputs.

Value
An object of class deadata

Examples
data("Coll_Blasco_2006")
data_example <- make_deadata(datadea = Coll_Blasco_2006,
ni = 2,
no = 2)
# This is the same as:
data_example <- make_deadata(Coll_Blasco_2006,
inputs = 2:3,
outputs = 4:5)
# And the same as:
dmunames <- c("A", "B", "C", "D", "E", "F")
make_deadata_fuzzy 45

nd <- length(dmunames) # Number of DMUs

inputnames <- c("Employees", "Capital")
ni <- length(inputnames) # Number of Inputs
outputnames <- c("Vehicles", "Orders")
no <- length(outputnames) # Number of Outputs
inputs <- matrix(c(8, 8, 11, 15, 14, 12, 12, 13, 11, 18, 18, 20),
nrow = ni, ncol = nd, dimnames = list(inputnames, dmunames))
outputs <- matrix(c(14, 20, 25, 42, 8, 30, 25, 8, 40, 22, 24, 30),
nrow = no, ncol = nd, dimnames = list(outputnames, dmunames))
data_example <- make_deadata(inputs = inputs,
outputs = outputs)
# If the first input is a non-controllable input:
data_example <- make_deadata(Coll_Blasco_2006,
inputs = 2:3,
outputs = 4:5,
nc_inputs = 1)
# If the second output is a non-discretionary output:
data_example <- make_deadata(Coll_Blasco_2006,
inputs = 2:3,
outputs = 4:5,
nd_outputs = 2)
# If the second input is a non-discretionary input and the second output is an undesirable:
data_example <- make_deadata(Coll_Blasco_2006,
inputs = 2:3,
outputs = 4:5,
nd_inputs = 2,
ud_outputs = 2)

make_deadata_fuzzy make_deadata_fuzzy

Description
This function creates, from a data frame, a deadata_fuzzy structure, which is as list with fields
input, output and dmunames. At the same time, input and output are lists with fields mL, mR, dL
and dR.
" alt" alt

Usage
make_deadata_fuzzy(datadea,
dmus = 1,
[Link] = NULL,
[Link] = NULL,
[Link] = NULL,
[Link] = NULL,
[Link] = NULL,
[Link] = NULL,
46 make_deadata_fuzzy

[Link] = NULL,
[Link] = NULL,
nc_inputs = NULL,
nc_outputs = NULL,
nd_inputs = NULL,
nd_outputs = NULL,
ud_inputs = NULL,
ud_outputs = NULL)

Arguments
datadea Data frame with DEA data.
dmus Column (number or name) of DMUs (optional). By default, it is the first column.
If there is not any DMU column, then it must be NULL.
[Link] Where are (columns) the mL (left centers) of trapezoidal fuzzy inputs in datadea.
If an input is triangular or crisp, we put the column where the centers or the crisp
values are, respectively.
Alternatively to datadea, [Link] can be a matrix of size (number of inputs
x number of DMUs) with the mL of trapezoidal fuzzy inputs, the centers of tri-
angular inputs, and the crisp values of crisp inputs. In this case, DMUs names
are taken from the columns names.
[Link] Where are (columns) the mR (right centers) of trapezoidal fuzzy inputs in datadea.
If an input is triangular or crisp, we put NA.
Alternatively to datadea, [Link] can be a matrix of size (number of inputs
x number of DMUs) with the mR of trapezoidal fuzzy inputs, the centers of tri-
angular inputs, and the crisp values of crisp inputs. If all inputs are triangular or
crisp, then [Link] must be NULL (default) or equal to [Link].
[Link] Where are (columns) the dL (left radii) of trapezoidal and triangular fuzzy inputs
in datadea. If an input is symmetric, we put the column where the radii are. If
an input is rectangular or crisp, we put NA.
Alternatively to datadea, [Link] can be a matrix of size (number of inputs
x number of DMUs) with the dL of trapezoidal and triangular fuzzy inputs. If
an input is rectangular or crisp, its radius is zero. If all inputs are rectangular or
crisp, then [Link] must be NULL (default) or a zero matrix.
[Link] Where are (columns) the dR (right radii) of trapezoidal and triangular fuzzy in-
puts in datadea. If an input is symmetric, rectangular or crisp, we put NA.
Alternatively to datadea, [Link] can be a matrix of size (number of in-
puts x number of DMUs) with the dR of trapezoidal and triangular fuzzy inputs.
If an input is rectangular or crisp, its radius is zero. If all inputs are symmet-
ric, rectangular or crisp, then [Link] must be NULL (default) or equal to
[Link].
[Link] Analogous to [Link], but relating to outputs.
[Link] Analogous to [Link], but relating to outputs.
[Link] Analogous to [Link], but relating to outputs.
[Link] Analogous to [Link], but relating to outputs.
make_malmquist 47

nc_inputs A numeric vector containing the indices of non-controllable inputs.

nc_outputs A numeric vector containing the indices of non-controllable outputs.
nd_inputs A numeric vector containing the indices of non-discretionary inputs.
nd_outputs A numeric vector containing the indices of non-discretionary outputs.
ud_inputs A numeric vector containing the indices of undesirable (good) inputs.
ud_outputs A numeric vector containing the indices of undesirable (bad) outputs.

Value

An object of class deadata_fuzzy.

Examples
# Example 1. If inputs and/or outputs are symmetric triangular fuzzy numbers
data("Leon2003")
data_example <- make_deadata_fuzzy(datadea = Leon2003,
[Link] = 2,
[Link] = 3,
[Link] = 4,
[Link] = 5)
# Example 2. If inputs and/or outputs are non-symmetric triangular fuzzy numbers
data("Kao_Liu_2003")
data_example <- make_deadata_fuzzy(Kao_Liu_2003,
[Link] = 2,
[Link] = 3:7,
[Link] = c(NA, NA, 8, NA, 10),
[Link] = c(NA, NA, 9, NA, 11))

make_malmquist make_malmquist

Description

This function creates, from a data frame, a list of deadata objects.

Usage

make_malmquist(datadea,
nper = NULL,
percol = NULL,
arrangement = c("horizontal", "vertical"),
...)
48 make_malmquist

Arguments

datadea Data frame with DEA data.

nper Number of time periods (with dataset in wide format).
percol Column of time period (with dataset in long format).
arrangement Horizontal with data in wide format. Vertical with data in long format.
... Other options to be passed to the make_deadata function.

Value

An object of class deadata

Author(s)

Vicente Coll-Serrano (<[Link]@[Link]>). Quantitative Methods for Measuring Culture

(MC2). Applied Economics.
Vicente Bolós (<[Link]@[Link]>). Department of Business Mathematics
Rafael Benítez (<[Link]@[Link]>). Department of Business Mathematics
University of Valencia (Spain)

Examples

# Example 1. If you have a dataset in wide format.

data("Economy")
data_example <- make_malmquist(datadea = Economy,
nper = 5,
arrangement = "horizontal",
ni = 2,
no = 1)
# This is the same as:
data_example <- make_malmquist(datadea = Economy,
nper = 5,
arrangement = "horizontal",
inputs = 2:3,
outputs = 4)
# Example 2. If you have a dataset in long format.
data("EconomyLong")
data_example2 <- make_malmquist(EconomyLong,
percol = 2,
arrangement = "vertical",
inputs = 3:4,
outputs = 5)
malmquist_index 49

malmquist_index Malmquist index

Description
This function calculates the input/output oriented Malmquist productivity index under constant or
variable returns-to-scale.

Usage
malmquist_index(datadealist,
dmu_eval = NULL,
dmu_ref = NULL,
orientation = c("io", "oo"),
rts = c("crs", "vrs"),
type1 = c("cont", "seq", "glob"),
type2 = c("fgnz", "rd", "gl", "bias"),
tc_vrs = FALSE,
vtrans_i = NULL,
vtrans_o = NULL)

Arguments
datadealist A list with the data (deadata objects) at different times, including DMUs, inputs
and outputs.
dmu_eval A numeric vector containing which DMUs have to be evaluated. If NULL (de-
fault), all DMUs are considered.
dmu_ref A numeric vector containing which DMUs are the evaluation reference set. If
NULL (default), all DMUs are considered.
orientation A string, equal to "io" (input oriented) or "oo" (output oriented).
rts A string, determining the type of returns to scale, equal to "crs" (constant) or
"vrs" (variable).
type1 A string, equal to "cont" (contemporary), "seq" (sequential) or "glob" (global).
type2 A string, equal to "fgnz" (Fare et al. 1994), "rd" (Ray and Desli 1997), "gl"
(generalized) or "bias" (biased).
tc_vrs Logical. If it is FALSE, it computes the vrs bias malmquist index by using the
technical change under crs (Fare and Grosskopf 1996). Otherwise, it uses the
technical change under vrs.
vtrans_i Numeric vector of translation for undesirable inputs in non-directional basic
models. If vtrans_i[i] is NA, then it applies the "max + 1" translation to the i-th
undesirable input. If vtrans_i is a constant, then it applies the same translation
to all undesirable inputs. If vtrans_i is NULL, then it applies the "max + 1"
translation to all undesirable inputs.
vtrans_o Numeric vector of translation for undesirable outputs in non-directional basic
models, analogous to vtrans_i, but applied to outputs.
50 malmquist_index

Value

A numeric list with Malmquist index and other parameters.

Note

In the results: EC = Efficiency Change, PTEC = Pure Technical Efficiency Change, SEC = Scale
Efficiency Change, TC = Technological Change, MI = Malmquist Index

Author(s)

Vicente Coll-Serrano (<[Link]@[Link]>). Quantitative Methods for Measuring Culture

(MC2). Applied Economics.
Vicente Bolos (<[Link]@[Link]>). Department of Business Mathematics
Rafael Benitez (<[Link]@[Link]>). Department of Business Mathematics
University of Valencia (Spain)

References

Caves, D.W.; Christensen, L. R.; Diewert, W.E. (1982). “The Economic Theory of Index Numbers
and the Measurement of Input, Output, and Productivity”. Econometrica, 50(6), 1393-1414.
Fare, R.; Grifell-Tatje, E.; Grosskopf, S.; Lovell, C.A.K. (1997). "Biased Technical Change and the
Malmquist Productivity Index". Scandinavian Journal of Economics, 99(1), 119-127.
Fare, R.; Grosskopf, S.; Lindgren, B.; Roos, P. (1989). “Productivity Developments in Swedish
Hospitals: A Malmquist Output Index Approach”. Discussion paper n. 89-3. Southern Illinois
University. Illinois.
Fare, R.; Grosskopf, S.; Lindgren, B.; Roos, P. (1992). “Productivity changes in Swedish Phar-
macies 1980-89: A nonparametric Malmquist Approach”. Journal of productivity Analysis, 3(3),
85-101.
Fare, R.; Grosskopf, S.; Norris, M.; Zhang, Z. (1994). “Productivity Growth, Technical Progress,
and Efficiency Change in Industrialized Countries”. American Economic Review, 84(1), 66-83.
Fare, R.; Grosskopf, S.; Roos, P. (1998), Malmquist Productivity Indexes: A Survey of Theory and
Practice. In: Fare R., Grosskopf S., Russell R.R. (eds) Index Numbers: Essays in Honour of Sten
Malmquist. Springer.
Grifell-Tatje, E.; Lovell, C.A.K. (1999). "A Generalized Malmquist productivity index". Top, 7(1),
81-101.
Pastor, J.T.; Lovell, C.A.k. (2005). "A global Malmquist productiviyt index". Economics Letters,
88, 266-271.
Ray, S.C.; Desli, E. (1997). "Productivity Growth, Technical Progress, and Efficiency Change in
Industrialized Countries: Comment". The American Economic Review, 87(5), 1033-1039.
Shestalova, V. (2003). "Sequential Malmquist Indices of Productivity Growth: An Application to
OECD Industrial Activities". Journal of Productivity Analysis, 19, 211-226.
malmquist_index 51

Examples

# Example 1. With dataset in wide format.

# Replication of results in Wang and Lan (2011, p. 2768)
data("Economy")
data_example <- make_malmquist(datadea = Economy,
nper = 5,
arrangement = "horizontal",
ni = 2,
no = 1)
result <- malmquist_index(data_example, orientation = "io")
mi <- result$mi
effch <- result$ec
tech <- result$tc

# Example 2. With dataset in long format.

# Replication of results in Wang and Lan (2011, p. 2768)
data("EconomyLong")
data_example2 <- make_malmquist(EconomyLong,
percol = 2,
arrangement = "vertical",
inputs = 3:4,
outputs = 5)
result2 <- malmquist_index(data_example2, orientation = "io")
mi2 <- result2$mi
effch2 <- result2$ec
tech2 <- result2$tc

# Example 3. Replication of results in Grifell-Tatje and Lovell (1999, p. 100).

data("Grifell_Lovell_1999")
data_example <- make_malmquist(Grifell_Lovell_1999,
percol = 1,
dmus = 2,
inputs = 3,
outputs = 4,
arrangement = "vertical")
result_fgnz <- malmquist_index(data_example,
orientation = "oo",
rts = "vrs",
type1 = "cont",
type2 = "fgnz")
mi_fgnz <- result_fgnz$mi

result_rd <- malmquist_index(data_example,

orientation = "oo",
rts = "vrs",
type1 = "cont",
type2 = "rd")
mi_rd <- result_rd$mi

result_gl <- malmquist_index(data_example,

orientation = "oo",
rts = "vrs",
52 maximal_friends

type1 = "cont",
type2 = "gl")
mi_gl <- result_gl$mi

maximal_friends Maximal friends of a set of DMUs.

Description
Finds the maximal friends subsets of a given set of DMUs, according to Tone (2010). It uses an
ascending algorithm in order to find directly maximal subsets.

Usage
maximal_friends(datadea,
dmu_ref = NULL,
rts = c("crs", "vrs", "nirs", "ndrs"),
tol = 1e-6,
silent = FALSE)

Arguments
datadea A deadata object with n DMUs, m inputs and s outputs.
dmu_ref A numeric vector containing which DMUs are the evaluation reference set, i.e.
the cluster of DMUs from which we want to find maximal friends. If NULL
(default), all DMUs are considered.
rts A string, determining the type of returns to scale, equal to "crs" (constant), "vrs"
(variable), "nirs" (non-increasing) or "ndrs" (non-decreasing).
tol Numeric, a tolerance margin for checking efficiency. It is 1e-6 by default.
silent Logical, if FALSE (default) steps are printed.

Value
A list with numeric vectors representing maximal friends subsets of DMUs.

References
Tone, K. (2010). "Variations on the theme of slacks-based measure of efficiency in DEA", European
Journal of Operational Research, 200, 901-907. doi:10.1016/[Link].2009.01.027

modelfuzzy_guotanaka Fuzzy DEA model

Description
Solve the Fuzzy input-oriented and output-oriented DEA model proposed by Guo and Tanaka
(2001) under constant returns to scale. In deaR is implemented the LP poblem given by the model
(16) in Guo and Tanaka (2001, p.155). The fuzzy efficiencies are calculated according to equations
in (17) (Guo and Tanaka, 2001, p.155). The (crisp) relative efficiencies and multipliers for the case
h = 1 are obtained from the CCR model (model_multiplier).

Usage
modelfuzzy_guotanaka(datadea,
dmu_eval = NULL,
dmu_ref = NULL,
orientation = c("io", "oo"),
h = 1)

Arguments
datadea A deadata_fuzzy object, including DMUs, inputs and outputs.
dmu_eval A numeric vector containing which DMUs have to be evaluated. If NULL (de-
fault), all DMUs are considered.
dmu_ref A numeric vector containing which DMUs are the evaluation reference set. If
NULL (default), all DMUs are considered.
54 modelfuzzy_guotanaka

orientation A string, equal to "io" (input oriented) or "oo" (output oriented).

h A numeric vector with the h-levels (in [0,1]).

Value
An object of class deadata_fuzzy.

Note
The optimal solution of model (16) is not unique.

References
Emrouznejad, A.; Tavana, M.; Hatami-Marbini, A. (2014). “The State of the Art in Fuzzy Data
Envelopment Analysis”, in A. Emrouznejad and M. Tavana (eds.), Performance Measurement with
Fuzzy Data Envelopment Analysis. Studies in Fuzziness and Soft Computing 309. Springer, Berlin.
doi:10.1007/9783642413728_1
Guo, P.; Tanaka, H. (2001). "Fuzzy DEA: A Perceptual Evaluation Method", Fuzzy Sets and Sys-
tems, 119, 149–160. doi:10.1016/S01650114(99)001062
Hatami-Marbini, A.; Emrouznejad, A.; Tavana, M. (2011). "A Taxonomy and Review of the Fuzzy
Data Envelopment Analysis Literature: Two Decades in the Making", European Journal of Opera-
tional Research, 214, 457–472. doi:10.1016/[Link].2011.02.001

See Also
model_basic, model_multiplier, modelfuzzy_kaoliu, modelfuzzy_possibilistic, cross_efficiency_fuzzy

Examples
# Example 1.
# Replication results in Guo and Tanaka (2001, p. 159).
# In deaR is implemented the LP poblem given by the model 16 in Guo and Tanaka (2001, p. 155).
# The fuzzy efficiencies are calculated according to equations in (17) (Guo and Tanaka, 2001,p.155).
data("Guo_Tanaka_2001")
data_example <- make_deadata_fuzzy(Guo_Tanaka_2001,
[Link] = 2:3,
[Link] = 4:5,
[Link] = 6:7,
[Link] = 8:9)
result <- modelfuzzy_guotanaka(data_example,
h = c(0, 0.5, 0.75, 1),
modelfuzzy_kaoliu 55

orientation = "io")
efficiencies(result)

# Example 2.
data("Guo_Tanaka_2001")
data_example <- make_deadata_fuzzy(Guo_Tanaka_2001,
[Link] = 2:3,
[Link] = 4:5,
[Link] = 6:7,
[Link] = 8:9)
result2 <- modelfuzzy_guotanaka(data_example,
h = seq(0, 1, by = 0.1),
orientation = "io")
efficiencies(result2)

modelfuzzy_kaoliu Fuzzy DEA model.

Description
Solve the fuzzy DEA model by Kao and Liu (2000)

Usage
modelfuzzy_kaoliu(datadea,
dmu_eval = NULL,
kaoliu_modelname = c("basic", "additive", "addsupereff",
"deaps", "fdh", "multiplier", "nonradial", "profit",
"rdm", "sbmeff", "sbmsupereff", "supereff"),
alpha = 1,
...)

Arguments
datadea A deadata_fuzzy object, including DMUs, inputs and outputs.
dmu_eval A numeric vector containing which DMUs have to be evaluated. If NULL (de-
fault), all DMUs are considered.
kaoliu_modelname
a string containing the name of the model.
alpha A numeric vector with the alpha-cuts (in [0,1]). If alpha>1, it determines the
number of alpha-cuts, equispatially distributed in [0,1].
... dmu_ref, orientation, rts and other model parameters.

Value
An object of class deadata_fuzzy.
56 modelfuzzy_kaoliu

References
Boscá, J.E.; Liern, V.; Sala, R.; Martínez, A. (2011). "Ranking Decision Making Units by Means
of Soft Computing DEA Models". International Journal of Uncertainty, Fuzziness and Knowledge-
Based Systems, 19(1), p.115-134.
Emrouznejad, A.; Tavana, M.; Hatami-Marbini, A. (2014). “The State of the Art in Fuzzy Data
Envelopment Analysis”, in A. Emrouznejad and M. Tavana (eds.), Performance Measurement with
Fuzzy Data Envelopment Analysis. Studies in Fuzziness and Soft Computing 309. Springer, Berlin.
Hatami-Marbini, A.; Emrouznejad, A.; Tavana, M. (2011). "A Taxonomy and Review of the Fuzzy
Data Envelopment Analysis Literature: Two Decades in the Making", European Journal of Opera-
tional Research, 214, 457–472.
Kao, C.; Liu, S.T. (2000). “Fuzzy efficiency measures in data envelopment analysis, Fuzzy Sets and
Systems”, 119, 149–160.
Kao, C., Liu, S.T., (2000). “Data envelopment analysis with missing data: An application to uni-
versity libraries in Taiwan”, Journal of the Operational Research Society, 51, 897–905.
Kao, C., Liu, S.T. (2003). “A mathematical programming approach to fuzzy efficiency ranking”,
International Journal of Production Economics, 85.

See Also
model_basic, model_multiplier, modelfuzzy_possibilistic, modelfuzzy_guotanaka

Examples
# Example 1.
# Replication of results in Boscá, Liern, Sala and Martínez (2011, p.125)
data("Leon2003")
data_example <- make_deadata_fuzzy(datadea = Leon2003,
[Link] = 2,
[Link] = 3,
[Link] = 4,
[Link] = 5)
result <- modelfuzzy_kaoliu(data_example,
kaoliu_modelname = "basic",
alpha = seq(0, 1, by = 0.1),
orientation = "io",
rts = "vrs")
efficiencies(result)

# Example 2.
modelfuzzy_possibilistic 57

# Replication of results in Kao and Liu (2003, p.152)

data("Kao_Liu_2003")
data_example <- make_deadata_fuzzy(Kao_Liu_2003,
[Link] = 2,
[Link] = 3:7,
[Link] = c(NA, NA, 8, NA, 10),
[Link] = c(NA, NA, 9, NA, 11))
result <- modelfuzzy_kaoliu(data_example,
kaoliu_modelname = "basic",
orientation = "oo",
rts = "vrs",
alpha = 0)
sol <- efficiencies(result)
eff <- [Link](1 / sol$Worst, 1 / sol$Best)
names(eff) <- c("eff_lower", "eff_upper")
eff

modelfuzzy_possibilistic
Possibilistic Fuzzy DEA model.

Description
Solve the possibilistic fuzzy DEA model proposed by León et. al (2003).

Usage
modelfuzzy_possibilistic(datadea,
dmu_eval = NULL,
poss_modelname = c("basic"),
h = 1,
...)

Arguments
datadea A deadata_fuzzy object, including DMUs, inputs and outputs.
dmu_eval A numeric vector containing which DMUs have to be evaluated. If NULL (de-
fault), all DMUs are considered.
poss_modelname a string containing the name of the model.
h A numeric vector with the h-levels (in [0,1]).
... dmu_ref, orientation, rts and other model parameters.

Value
An object of class deadata_fuzzy.
58 model_additive

Author(s)

Vicente Coll-Serrano (<[Link]@[Link]>). Quantitative Methods for Measuring Culture

(MC2). Applied Economics.
Vicente Bolós (<[Link]@[Link]>). Department of Business Mathematics
Rafael Benítez (<[Link]@[Link]>). Department of Business Mathematics
University of Valencia (Spain)

References

Emrouznejad, A.; Tavana, M.; Hatami-Marbini, A. (2014). “The State of the Art in Fuzzy Data
Envelopment Analysis”, in A. Emrouznejad and M. Tavana (eds.), Performance Measurement with
Fuzzy Data Envelopment Analysis. Studies in Fuzziness and Soft Computing 309. Springer, Berlin.
doi:10.1007/9783642413728_1
Hatami-Marbini, A.; Emrouznejad, A.; Tavana, M. (2011). "A Taxonomy and Review of the Fuzzy
Data Envelopment Analysis Literature: Two Decades in the Making", European Journal of Opera-
tional Research, 214, 457–472. doi:10.1016/[Link].2011.02.001
Leon, T.; Liern, V. Ruiz, J.; Sirvent, I. (2003). "A Possibilistic Programming Approach to the
Assessment of Efficiency with DEA Models", Fuzzy Sets and Systems, 139, 407–419. doi:10.1016/
S01650114(02)006085

See Also

model_basic, modelfuzzy_kaoliu, modelfuzzy_guotanaka

Examples
# Replication of results in Leon et. al (2003, p. 416)
data("Leon2003")
data_example <- make_deadata_fuzzy(Leon2003,
[Link] = 2,
[Link] = 3,
[Link] = 4,
[Link] = 5)
result <- modelfuzzy_possibilistic(data_example,
h = seq(0, 1, by = 0.1),
orientation = "io",
rts = "vrs")
efficiencies(result)

model_additive Additive DEA model.

model_additive 59

Description
Solve the additive model of Charnes et. al (1985). With the current version of deaR, it is possible
to solve input-oriented, output-oriented, and non-oriented additive model under constant and non-
constant returns to scale.
Besides, the user can set weights for the input slacks and/or output slacks. So, it is also possible to
solve weighted additive models. For example: Measure of Inefficiency Proportions (MIP), Range
Adjusted Measure (RAM), etc.

Usage
model_additive(datadea,
dmu_eval = NULL,
dmu_ref = NULL,
orientation = NULL,
weight_slack_i = 1,
weight_slack_o = 1,
rts = c("crs", "vrs", "nirs", "ndrs", "grs"),
L = 1,
U = 1,
compute_target = TRUE,
returnlp = FALSE,
...)

returnlp Logical. If it is TRUE, it returns the linear problems (objective function and
constraints).
... Ignored, for compatibility issues.

Note
In this model, the efficiency score is the sum of the slacks. Therefore, a DMU is efficient when the
objective value (objval) is zero.

References
Charnes, A.; Cooper, W.W.; Golany, B.; Seiford, L.; Stuz, J. (1985) "Foundations of Data En-
velopment Analysis for Pareto-Koopmans Efficient Empirical Production Functions", Journal of
Econometrics, 30(1-2), 91-107. doi:10.1016/03044076(85)901332
Charnes, A.; Cooper, W.W.; Lewin, A.Y.; Seiford, L.M. (1994). Data Envelopment Analysis: The-
ory, Methology, and Application. Boston: Kluwer Academic Publishers. doi:10.1007/97894011-
06375
Cooper, W.W.; Park, K.S.; Pastor, J.T. (1999). "RAM: A Range Adjusted Measure of Inefficiencies
for Use with Additive Models, and Relations to Other Models and Measures in DEA". Journal of
Productivity Analysis, 11, p. 5-42. doi:10.1023/A:1007701304281

model_addmin Additive-min DEA model.

Description
Solve the weighted version of the additive-min (mADD) model of Aparicio et. al (2007) with
different returns to scale. For non constant returns to scale, a modification given by Zhu et al.
(2018) is done.

Usage
model_addmin(datadea,
dmu_eval = NULL,
62 model_addmin

dmu_ref = NULL,
orientation = NULL,
weight_slack_i = 1,
weight_slack_o = 1,
rts = c("crs", "vrs", "nirs", "ndrs"),
method = c("mf", "milp"),
extreff = NULL,
M_d = NULL,
M_lambda = 1e3,
maxfr = NULL,
tol = 1e-6,
silent = TRUE,
compute_target = TRUE,
check_target = FALSE,
returnlp = FALSE,
...)

Arguments
datadea A deadata object with n DMUs, m inputs and s outputs.
dmu_eval A numeric vector containing which DMUs have to be evaluated. If NULL (de-
fault), all DMUs are considered.
dmu_ref A numeric vector containing which DMUs are the evaluation reference set. If
NULL (default), all DMUs are considered.
orientation This parameter is either NULL (default) or a string, equal to "io" (input-oriented)
or "oo" (output-oriented). It is used to modify the weight slacks. If input-
oriented, weight_slack_o are taken 0. If output-oriented, weight_slack_i
are taken 0.
weight_slack_i A value, vector of length m, or matrix m x ne (where ne is the length of dmu_eval)
with the weights of the input slacks. If 0, output-oriented.
weight_slack_o A value, vector of length s, or matrix s x ne (where ne is the length of dmu_eval)
with the weights of the output slacks. If 0, input-oriented.
rts A string, determining the type of returns to scale, equal to "crs" (constant),
"vrs" (variable), "nirs" (non-increasing) or "ndrs" (non-decreasing). Under non-
increasing or non-decreasing returns to scale, you may set check_target =
TRUE because methods are not reliable. Generalized returns to scale are not
available.
method A string with the method: "mf" (default) for maximal friends, or "milp" for the
mixed integer linear program of Aparicio et al. (2007). MILP method is faster
but very problematic numerically.
extreff A vector with the extreme efficient DMUs for "milp" method, as it is returned
by function extreme_efficient. If NULL (default) this vector is computed in-
ternally.
M_d Numeric, a big positive quantity for "milp" method. It is an upper bound for
auxiliary variables named "d" in Aparicio (2007). If NULL (default), it is esti-
mated automatically. A very big value can produce catastrophic cancellations.
If the results are not correct or the solver hangs, try to change its value.
model_addmin 63

M_lambda Numeric, a big positive quantity for "milp" method. It is an upper bound for
lambda variables. A very big value can produce catastrophic cancellations. If
the results are not correct or the solver hangs, try to change its value (1e3 by
default).
maxfr A list with the maximal friends sets for "mf" method, as it is returned by function
maximal_friends. If NULL (default) this list is computed internally.
tol Numeric, a tolerance margin for checking efficiency in extreme_efficient or
maximal_friends functions, and for checking targets.
silent Logical. If FALSE, it prints all the messages from function maximal_friends.
compute_target Logical. If it is TRUE (default), it computes targets.
check_target Logical. If it is TRUE, it checks the efficiency of targets. If a target is not efficient,
the method has failed.
returnlp Logical. If it is TRUE, it returns the linear problems (objective function and
constraints).
... For compatibility issues.

Note

In this model, the efficiency score is the sum of the slacks. Therefore, a DMU is efficient when the
objective value (objval) is zero.

Author(s)

Vicente Coll-Serrano (<[Link]@[Link]>). Quantitative Methods for Measuring Culture

(MC2). Applied Economics.
Vicente Bolós (<[Link]@[Link]>). Department of Business Mathematics
Rafael Benítez (<[Link]@[Link]>). Department of Business Mathematics
University of Valencia (Spain)

References

Aparicio, J.; Ruiz, J.L.; Sirvent, I. (2007) "Closest targets and minimum distance to the Pareto-
efficient frontier in DEA", Journal of Productivity Analysis, 28, 209-218. doi:10.1007/s11123007-
00395
Zhu, Q.; Wu, J.; Ji, X.; Li, F. (2018) "A simple MILP to determine closest targets in non-oriented
DEA model satisfying strong monotonicity", Omega, 79, 1-8. doi:10.1016/[Link].2017.07.003

See Also

model_additive, extreme_efficient, maximal_friends

64 model_addsupereff

Examples
# Example 1.
data("Airlines")
datadea <- make_deadata(Airlines,
inputs = 4:7,
outputs = 2:3)
result <- model_addmin(datadea = datadea,
method = "milp")
targets(result)

## Not run:
# Example 2. Directional model with Additive-min model in second stage
data("Airlines")
datadea <- make_deadata(Airlines,
inputs = 4:7,
outputs = 2:3)
resdir <- model_basic(datadea = datadea,
orientation = "dir",
maxslack = FALSE)
proj_input <- targets(resdir)[[1]] + slacks(resdir)[[1]]
proj_output <- targets(resdir)[[2]] - slacks(resdir)[[2]]
nd <- ncol(datadea$dmunames) # Number of DMUs
maxfr <- maximal_friends(datadea = datadea)
for (i in 1:nd) {
datadea2 <- datadea
datadea2$input[, i] <- proj_input[i, ]
datadea2$output[, i] <- proj_output[i, ]
DMUaux <- model_addmin(datadea = datadea2,
method = "mf",
maxfr = maxfr,
dmu_eval = i)$DMU[[1]]
resdir$DMU[[i]]$slack_input <- DMUaux$slack_input
resdir$DMU[[i]]$slack_output <- DMUaux$slack_output
resdir$DMU[[i]]$target_input <- DMUaux$target_input
resdir$DMU[[i]]$target_output <- DMUaux$target_output
}
targets(resdir)

## End(Not run)

model_addsupereff Additive super-efficiency DEA model.

Description
Solve the additive super-efficiency model proposed by Du, Liang and Zhu (2010). It is an extension
of the SBM super-efficiency to the additive DEA model.
model_addsupereff 65

Usage

model_addsupereff(datadea,
dmu_eval = NULL,
dmu_ref = NULL,
orientation = NULL,
weight_slack_i = NULL,
weight_slack_o = NULL,
rts = c("crs", "vrs", "nirs", "ndrs", "grs"),
L = 1,
U = 1,
compute_target = TRUE,
returnlp = FALSE,
...)

Arguments

datadea A deadata object with n DMUs, m inputs and s outputs.

dmu_eval A numeric vector containing which DMUs have to be evaluated. If NULL (de-
fault), all DMUs are considered.
dmu_ref A numeric vector containing which DMUs are the evaluation reference set. If
NULL (default), all DMUs are considered.
orientation This parameter is either NULL (default) or a string, equal to "io" (input-oriented)
or "oo" (output-oriented). It is used to modify the weight slacks. If input-
oriented, weight_slack_o are taken 0. If output-oriented, weight_slack_i
are taken 0.
weight_slack_i A value, vector of length m, or matrix m x ne (where ne is the length of dmu_eval)
with the weights of the input super-slacks (t_input). If 0, output-oriented. If
weight_slack_i is the matrix of the inverses of inputs of DMUS in dmu_eval
(default), the model is unit invariant.
weight_slack_o A value, vector of length s, or matrix s x ne (where ne is the length of dmu_eval)
with the weights of the output super-slacks (t_output). If 0, input-oriented. If
weight_slack_o is the matrix of the inverses of outputs of DMUS in dmu_eval
(default), the model is unit invariant.
rts A string, determining the type of returns to scale, equal to "crs" (constant), "vrs"
(variable), "nirs" (non-increasing), "ndrs" (non-decreasing) or "grs" (general-
ized).
L Lower bound for the generalized returns to scale (grs).
U Upper bound for the generalized returns to scale (grs).
compute_target Logical. If it is TRUE, it computes targets, projections and slacks.
returnlp Logical. If it is TRUE, it returns the linear problems (objective function and
constraints).
... Ignored, for compatibility issues.
66 model_basic

References
Du, J.; Liang, L.; Zhu, J. (2010). "A Slacks-based Measure of Super-efficiency in Data Envelopment
Analysis. A Comment", European Journal of Operational Research, 204, 694-697. doi:10.1016/
[Link].2009.12.007
Zhu, J. (2014). Quantitative Models for Performance Evaluation and Benchmarking. Data Envel-
opment Analysis with Spreadsheets. 3rd Edition Springer, New York. doi:10.1007/9783319066479

See Also
model_additive, model_supereff, model_sbmsupereff

Examples
# Replication of results in Du, Liang and Zhu (2010, Table 6, p.696)
data("Power_plants")
Power_plants <- make_deadata(Power_plants,
ni = 4,
no = 2)
result <- model_addsupereff(Power_plants,
rts = "crs")
efficiencies(result)

model_basic Basic (radial and directional) DEA model.

Description
It solves input and output oriented, along with directional, basic DEA models (envelopment form)
under constant (CCR model), variable (BCC model), non-increasing, non-decreasing or generalized
returns to scale. By default, models are solved in a two-stage process (slacks are maximized).
You can use the model_basic function to solve directional DEA models by choosing orientation
= "dir".
The model_basic function allows to treat with non-discretional, non-controllable and undesirable
inputs/outputs.
model_basic 67

Usage
model_basic(datadea,
dmu_eval = NULL,
dmu_ref = NULL,
orientation = c("io", "oo", "dir"),
dir_input = NULL,
dir_output = NULL,
rts = c("crs", "vrs", "nirs", "ndrs", "grs"),
L = 1,
U = 1,
maxslack = TRUE,
weight_slack_i = 1,
weight_slack_o = 1,
vtrans_i = NULL,
vtrans_o = NULL,
compute_target = TRUE,
compute_multiplier = FALSE,
returnlp = FALSE,
silent_ud = FALSE,
...)

Arguments
datadea A deadata object with n DMUs, m inputs and s outputs.
dmu_eval A numeric vector containing which DMUs have to be evaluated. If NULL (de-
fault), all DMUs are considered.
dmu_ref A numeric vector containing which DMUs are the evaluation reference set. If
NULL (default), all DMUs are considered.
orientation A string, equal to "io" (input oriented), "oo" (output oriented), or "dir" (direc-
tional).
dir_input A value, vector of length m, or matrix m x ne (where ne is the length of dmu_eval)
with the input directions. If dir_input == input matrix (of DMUS in dmu_eval)
and dir_output == 0, it is equivalent to input oriented (beta = 1 - efficiency).
If dir_input is omitted, input matrix (of DMUS in dmu_eval) is assigned.
dir_output A value, vector of length s, or matrix s x ne (where ne is the length of dmu_eval)
with the output directions. If dir_input == 0 and dir_output == output matrix
(of DMUS in dmu_eval), it is equivalent to output oriented (beta = efficiency
- 1). If dir_output is omitted, output matrix (of DMUS in dmu_eval) is as-
signed.
rts A string, determining the type of returns to scale, equal to "crs" (constant), "vrs"
(variable), "nirs" (non-increasing), "ndrs" (non-decreasing) or "grs" (general-
ized).
L Lower bound for the generalized returns to scale (grs).
U Upper bound for the generalized returns to scale (grs).
maxslack Logical. If it is TRUE, it computes the max slack solution.
68 model_basic

weight_slack_i A value, vector of length m, or matrix m x ne (where ne is the length of dmu_eval)

with the weights of the input slacks for the max slack solution.
weight_slack_o A value, vector of length s, or matrix s x ne (where ne is the length of dmu_eval)
with the weights of the output slacks for the max slack solution.
vtrans_i Numeric vector of translation for undesirable inputs with non-directional orien-
tation. If vtrans_i[i] is NA, then it applies the "max + 1" translation to the i-th
undesirable input. If vtrans_i is a constant, then it applies the same translation
to all undesirable inputs. If vtrans_i is NULL, then it applies the "max + 1"
translation to all undesirable inputs.
vtrans_o Numeric vector of translation for undesirable outputs with non-directional ori-
entation, analogous to vtrans_i, but applied to outputs.
compute_target Logical. If it is TRUE, it computes targets of the max slack solution.
compute_multiplier
Logical. If it is TRUE, it computes multipliers (dual solution) when orientation
is "io" or "oo".
returnlp Logical. If it is TRUE, it returns the linear problems (objective function and
constraints) of stage 1.
silent_ud Logical. For internal use, to avoid multiple warnings in the execution of malmquist_index
function with undesirable variables.
... Ignored, for compatibility issues.

Note
(1) Model proposed by Seiford and Zhu (2002) is applied for undesirable inputs/outputs and non-
directional orientation (i.e., input or output oriented). You should select "vrs" returns to scale (BCC
model) in order to maintain translation invariance. If deaR detects that you are not specifying rts
= "vrs", it makes the change to "vrs" automatically.
(2) With undesirable inputs and non-directional orientation use input-oriented BCC model, and with
undesirable outputs and non-directional orientation use output-oriented BCC model. Alternatively,
you can also treat the undesirable outputs as inputs and then apply the input-oriented BCC model
(similarly with undesirable inputs).
(3) Model proposed by Fare and Grosskopf (2004) is applied for undesirable inputs/outputs and
directional orientation.
(4) With orientation = "dir" (directional distance function model), efficient DMUs are those for
which beta = 0.

References
Charnes, A.; Cooper, W.W.; Rhodes, E. (1978). “Measuring the efficiency of decision making
units”, European Journal of Operational Research 2, 429–444.
Charnes, A.; Cooper, W.W.; Rhodes, E. (1979). “Short communication: Measuring the efficiency
of decision making units”, European Journal of Operational Research 3, 339.
Charnes, A.; Cooper, W.W.; Rhodes, E. (1981). "Evaluating Program and Managerial Efficiency:
An Application of Data Envelopment Analysis to Program Follow Through", Management Science,
27(6), 668-697.
Banker, R.; Charnes, A.; Cooper, W.W. (1984). “Some Models for Estimating Technical and Scale
Inefficiencies in Data Envelopment Analysis”, Management Science; 30; 1078-1092.
Undesirable inputs/outputs:
Pastor, J.T. (1996). "Translation Invariance in Data Envelopment Analysis: a Generalization", An-
nals of Operations Research, 66(2), 91-102.
Seiford, L.M.; Zhu, J. (2002). “Modeling undesirable factors in efficiency evaluation”, European
Journal of Operational Research 142, 16-20.
Färe, R. ; Grosskopf, S. (2004). “Modeling undesirable factors in efficiency evaluation: Comment”,
European Journal of Operational Research 157, 242-245.
Hua Z.; Bian Y. (2007). DEA with Undesirable Factors. In: Zhu J., Cook W.D. (eds) Modeling
Data Irregularities and Structural Complexities in Data Envelopment Analysis. Springer, Boston,
MA.
Non-discretionary/Non-controllable inputs/outputs:
Banker, R.; Morey, R. (1986). “Efficiency Analysis for Exogenously Fixed Inputs and Outputs”,
Operations Research; 34; 513-521.
Ruggiero J. (2007). Non-Discretionary Inputs. In: Zhu J., Cook W.D. (eds) Modeling Data Irregu-
larities and Structural Complexities in Data Envelopment Analysis. Springer, Boston, MA.
Directional DEA model:
Chambers, R.G.; Chung, Y.; Färe, R. (1996). "Benefit and Distance Functions", Journal of Eco-
nomic Theory, 70(2), 407-419.
Chambers, R.G.; Chung, Y.; Färe, R. (1998). "Profit Directional Distance Functions and Nerlovian
Efficiency", Journal of Optimization Theory and Applications, 95, 351-354.

See Also
model_multiplier, model_supereff

Examples
# Example 1. Basic DEA model with desirable inputs/outputs.
# Replication of results in Charnes, Cooper and Rhodes (1981).
data("PFT1981")
# Selecting DMUs in Program Follow Through (PFT)
PFT <- PFT1981[1:49, ]
PFT <- make_deadata(PFT,
inputs = 2:6,
70 model_deaps

outputs = 7:9 )
eval_pft <- model_basic(PFT,
orientation = "io",
rts = "crs")
eff <- efficiencies(eval_pft)
s <- slacks(eval_pft)
lamb <- lambdas(eval_pft)
tar <- targets(eval_pft)
ref <- references(eval_pft)
returns <- rts(eval_pft)

# Example 2. Basic DEA model with undesirable outputs.

# Replication of results in Hua and Bian (2007).
data("Hua_Bian_2007")
# The third output is an undesirable output.
data_example <- make_deadata(Hua_Bian_2007,
ni = 2,
no = 3,
ud_outputs = 3)
# Translation parameter (vtrans_o) is set to 1500
result <- model_basic(data_example,
orientation = "oo",
rts = "vrs",
vtrans_o = 1500)
eff <- efficiencies(result)
1 / eff # results M5 in Table 6-5 (p.119)

# Example 3. Basic DEA model with non-discretionary (fixed) inputs.

# Replication of results in Ruggiero (2007).
data("Ruggiero2007")
# The second input is a non-discretionary input.
datadea <- make_deadata(Ruggiero2007,
ni = 2,
no = 1,
nd_inputs = 2)
result <- model_basic(datadea,
orientation = "io",
rts = "crs")
efficiencies(result)

model_deaps Preference Structure DEA model.

Description
With this non-radial DEA model (Zhu, 1996), the user can specify the preference input (or output)
weigths that reflect the relative degree of desirability of the adjustments of the current input (or
output) levels.
model_deaps 71

Usage
model_deaps(datadea,
dmu_eval = NULL,
dmu_ref = NULL,
weight_eff = 1,
orientation = c("io", "oo"),
rts = c("crs", "vrs", "nirs", "ndrs", "grs"),
L = 1,
U = 1,
restricted_eff = TRUE,
maxslack = TRUE,
weight_slack = 1,
compute_target = TRUE,
returnlp = FALSE,
...)

Arguments
datadea A deadata object, including n DMUs, m inputs and s outputs.
dmu_eval A numeric vector containing which DMUs have to be evaluated. If NULL (de-
fault), all DMUs are considered.
dmu_ref A numeric vector containing which DMUs are the evaluation reference set. If
NULL (default), all DMUs are considered.
weight_eff Preference weights. If input-oriented, it is a value, vector of length m, or matrix
m x ne (where ne is the lenght of dmu_eval) with the weights applied to the input
efficiencies. If output-oriented, it is a value, vector of length s, or matrix s x ne
with the weights applied to the output efficiencies.
orientation A string, equal to "io" (input-oriented) or "oo" (output-oriented).
rts A string, determining the type of returns to scale, equal to "crs" (constant), "vrs"
(variable), "nirs" (non-increasing), "ndrs" (non-decreasing) or "grs" (general-
ized).
L Lower bound for the generalized returns to scale (grs).
U Upper bound for the generalized returns to scale (grs).
restricted_eff Logical. If it is TRUE, the efficiencies are restricted to be <=1 (input-oriented) or
>=1 (output-oriented).
maxslack Logical. If it is TRUE, it computes the max slack solution.
weight_slack If input-oriented, it is a value, vector of length s, or matrix s x ne with the
weights of the output slacks for the max slack solution. If output-oriented, it is
a value, vector of length m, or matrix m x ne with the weights of the input slacks
for the max slack solution.
compute_target Logical. If it is TRUE, it computes targets of the max slack solution.
returnlp Logical. If it is TRUE, it returns the linear problems (objective function and
constraints) of stage 1.
... Ignored, for compatibility issues.
72 model_fdh

References
Zhu, J. (1996). “Data Envelopment Analysis with Preference Structure”, The Journal of the Opera-
tional Research Society, 47(1), 136. doi:10.2307/2584258
Zhu, J. (2014). Quantitative Models for Performance Evaluation and Benchmarking. Data Envel-
opment Analysis with Spreadsheets. 3rd Edition Springer, New York. doi:10.1007/9783319066479

See Also
model_nonradial, model_profit, model_sbmeff

Examples
data("Fortune500")
data_deaps <- make_deadata(datadea = Fortune500,
ni = 3,
no = 2)
result <- model_deaps(data_deaps,
weight_eff = c(1, 2, 3),
orientation = "io",
rts = "vrs")
efficiencies(result)

model_fdh Free disposal hull (FDH) model.

Description
FDH model allows the free disposability to construct the production possibility set. The central
feature of the FDH model is the lack of convexity for its production possibility set (Thrall, 1999).

Usage
model_fdh(datadea,
fdh_modelname = c("basic"),
...)
model_fdh 73

Arguments
datadea A deadata object, including DMUs, inputs and outputs.
fdh_modelname A string containing the name of the model to apply FDH. For now, only "basic"
is available.
... dmu_eval, dmu_ref, orientation and other model parameters. Parameters like
rts, max_slack and returnlp are ignored.

References
Cherchye, L.; Kuosmanen, T.; Post, T. (2000). "What Is the Economic Meaning of FDH? A Reply
to Thrall". Journal of Productivity Analysis, 13(3), 263–267.
Deprins, D.; Simar, L. and Tulkens, H. (1984). Measuring Labor-Efficiency in Post Offices. In M.
Marchand, P. Pestieau and H. Tulkens (eds.), The Performance of Public Entreprises: Concepts and
Measurement. Amsterdam: North-Holland.
Sanei, M.; Mamizadeh Chatghayeb, S. (2013). “Free Disposal Hull Models in Supply Chain Man-
agement”, International Journal of Mathematical Modelling and Computations, 3(3), 125-129.
Thrall, R. M. (1999). "What Is the Economic Meaning of FDH?", Journal of Productivity Analysis,
11(3), 243–50.

Examples
# Example 1. FDH input-oriented.
# Replication of results in Sanei and Mamizadeh Chatghayeb (2013)
data("Supply_Chain")
data_fdh1 <- make_deadata(Supply_Chain,
inputs = 2:4,
outputs = 5:6)
result <- model_fdh(data_fdh1) # by default orientation = "io"
efficiencies(result)

# Example 2. FDH output-oriented.

# Replication of results in Sanei and Mamizadeh Chatghayeb (2013)
data("Supply_Chain")
data_fdh2 <- make_deadata(Supply_Chain,
inputs = 5:6,
outputs = 7:8)
result2 <- model_fdh(data_fdh2,
orientation = "oo")
efficiencies(result2)
74 model_multiplier

model_multiplier Multiplier DEA model

Description
Solve input-oriented and output-oriented basic DEA models (multiplicative form) under constant
(CCR DEA model), variable (BCC DEA model), non-increasing, non-decreasing or generalized
returns to scale. It does not take into account non-controllable, non-discretionary or undesirable
inputs/outputs.

Usage
model_multiplier(datadea,
dmu_eval = NULL,
dmu_ref = NULL,
epsilon = 0,
orientation = c("io", "oo"),
rts = c("crs", "vrs", "nirs", "ndrs", "grs"),
L = 1,
U = 1,
returnlp = FALSE,
compute_lambda = TRUE,
...)

Arguments
datadea A deadata object, including DMUs, inputs and outputs.
dmu_eval A numeric vector containing which DMUs have to be evaluated. If NULL (de-
fault), all DMUs are considered.
dmu_ref A numeric vector containing which DMUs are the evaluation reference set. If
NULL (default), all DMUs are considered.
epsilon Numeric, multipliers must be >= epsilon.
orientation A string, equal to "io" (input-oriented) or "oo" (output-oriented).
rts A string, determining the type of returns to scale, equal to "crs" (constant), "vrs"
(variable), "nirs" (non-increasing), "ndrs" (non-decreasing) or "grs" (general-
ized).
L Lower bound for the generalized returns to scale (grs).
U Upper bound for the generalized returns to scale (grs).
returnlp Logical. If it is TRUE, it returns the linear problems (objective function and
constraints).
compute_lambda Logical. If it is TRUE, it computes the dual problem and lambdas.
... Ignored, for compatibility issues.
model_multiplier 75

Note
(1) Very important with the multiplier model: "The optimal weights for an efficient DMU need
not be unique" (Cooper, Seiford and Tone, 2007:31). "Usually, the optimal weights for inefficient
DMUs are unique, the exception being when the line of the DMU is parallel to one of the boundaries
of the feasible region" (Cooper, Seiford and Tone, 2007:32).
(2) The measure of technical input (or output) efficiency obtained by using multiplier DEA models
is better the smaller the value of epsilon.
(3) Epsilon is usually set equal to 10^-6. However, if epsilon is not set correctly, the multiplier
model can be infeasible (Zhu,2014:49).

References
Charnes, A.; Cooper, W.W. (1962). “Programming with Linear Fractional Functionals”, Naval
Research Logistics Quarterly 9, 181-185. doi:10.1002/nav.3800090303
Charnes, A.; Cooper, W.W.; Rhodes, E. (1978). “Measuring the Efficiency of Decision Making
Units”, European Journal of Operational Research 2, 429–444. doi:10.1016/03772217(78)901388
Charnes, A.; Cooper, W.W.; Rhodes, E. (1979). “Short Communication: Measuring the Efficiency
of Decision Making Units”, European Journal of Operational Research 3, 339. doi:10.1016/0377-
2217(79)902297
Golany, B.; Roll, Y. (1989). "An Application Procedure for DEA", OMEGA International Journal
of Management Science, 17(3), 237-250. doi:10.1016/03050483(89)900297
Seiford, L.M.; Thrall, R.M. (1990). “Recent Developments in DEA. The Mathematical Program-
ming Approach to Frontier Analysis”, Journal of Econometrics 46, 7-38. doi:10.1016/03044076(90)90045-
U
Zhu, J. (2014). Quantitative Models for Performance Evaluation and Benchmarking. Data Envel-
opment Analysis with Spreadsheets. 3rd Edition Springer, New York. doi:10.1007/9783319066479

model_nonradial Non-radial DEA model.

Description

Non-radial DEA model allows for non-proportional reductions in each input or augmentations in
each output.

Usage

model_nonradial(datadea,
dmu_eval = NULL,
dmu_ref = NULL,
orientation = c("io", "oo"),
rts = c("crs", "vrs", "nirs", "ndrs", "grs"),
L = 1,
U = 1,
maxslack = TRUE,
weight_slack = 1,
compute_target = TRUE,
returnlp = FALSE,
...)
model_nonradial 77

Arguments
datadea A deadata object, including n DMUs, m inputs and s outputs.
dmu_eval A numeric vector containing which DMUs have to be evaluated. If NULL (de-
fault), all DMUs are considered.
dmu_ref A numeric vector containing which DMUs are the evaluation reference set. If
NULL (default), all DMUs are considered.
orientation A string, equal to "io" (input-oriented) or "oo" (output-oriented).
rts A string, determining the type of returns to scale, equal to "crs" (constant), "vrs"
(variable), "nirs" (non-increasing), "ndrs" (non-decreasing) or "grs" (general-
ized).
L Lower bound for the generalized returns to scale (grs).
U Upper bound for the generalized returns to scale (grs).
maxslack Logical. If it is TRUE, it computes the max slack solution.
weight_slack If input-oriented, it is a value, vector of length s, or matrix s x ne (where ne is
the length of dmu_eval) with the weights of the output slacks for the max slack
solution. If output-oriented, it is a value, vector of length m, or matrix m x ne
with the weights of the input slacks for the max slack solution.
compute_target Logical. If it is TRUE, it computes targets of the max slack solution.
returnlp Logical. If it is TRUE, it returns the linear problems (objective function and
constraints) of stage 1.
... Ignored, for compatibility issues.

References
Banker, R.D.; Morey, R.C. (1986). "Efficiency Analysis for Exogenously Fixed Inputs and Out-
puts", Operations Research, 34, 80-97. doi:10.1287/opre.34.4.513
Färe, R.; Lovell, C.K. (1978). "Measuring the Technical Efficiency of Production", Journal of
Economic Theory, 19(1), 150-162. doi:10.1016/00220531(78)900601
Wu, J.; Tsai, H.; Zhou, Z. (2011). "Improving Efficiency in International Tourist Hotels in Taipei
Using a Non-Radial DEA Model", International Journal of Contemporary Hospitatlity Manage-
ment, 23(1), 66-83. doi:10.1108/09596111111101670
Zhu, J. (1996). “Data Envelopment Analysis with Preference Structure”, The Journal of the Opera-
tional Research Society, 47(1), 136. doi:10.2307/2584258
78 model_profit

See Also
model_deaps, model_profit, model_sbmeff

Examples
# Replication of results in Wu, Tsai and Zhou (2011)
data("Hotels")
data_hotels <- make_deadata(Hotels,
inputs = 2:5,
outputs = 6:8)
result <- model_nonradial(data_hotels,
orientation = "oo",
rts = "vrs")
efficiencies(result)

model_profit Profit efficiency DEA model.

Description
Cost, revenue and profit efficiency DEA models.

Usage
model_profit(datadea,
dmu_eval = NULL,
dmu_ref = NULL,
price_input = NULL,
price_output = NULL,
rts = c("crs", "vrs", "nirs", "ndrs", "grs"),
L = 1,
U = 1,
restricted_optimal = TRUE,
returnlp = FALSE,
...)

price_output Unit prices of outputs for revenue or profit efficiency models. It is a value, vector
of length s, or matrix s x ne.
rts A string, determining the type of returns to scale, equal to "crs" (constant), "vrs"
(variable), "nirs" (non-increasing), "ndrs" (non-decreasing) or "grs" (general-
ized).
L Lower bound for the generalized returns to scale (grs).
U Upper bound for the generalized returns to scale (grs).
restricted_optimal
Logical. If it is TRUE, the optimal inputs are restricted to be <= inputs (for cost
efficiency models) or the optimal outputs are restricted to be >= outputs (for
revenue efficiency models).
returnlp Logical. If it is TRUE, it returns the linear problems (objective function and
constraints) of stage 1.
... Ignored, for compatibility issues.

References
Coelli, T.; Prasada Rao, D.S.; Battese, G.E. (1998). An introduction to efficiency and productivity
analysis. Jossey-Bass, San Francisco, pp 73–104. doi:10.1002/ev.1441

See Also
model_deaps, model_nonradial, model_sbmeff

Examples
# Example 1. Replication of results in Coelli et al. (1998, p.166).
# Cost efficiency model.
data("Coelli_1998")
# Selection of prices: input_prices is the transpose where the prices for inputs are.
input_prices <- t(Coelli_1998[, 5:6])

data_example1 <- make_deadata(Coelli_1998,

ni = 2,
no = 1)
result1 <- model_profit(data_example1,
price_input = input_prices,
rts = "crs",
restricted_optimal = FALSE)
# notice that the option by default is restricted_optimal = TRUE
80 model_rdm

efficiencies(result1)

# Example 2. Revenue efficiency model.

data("Coelli_1998")
# Selection of prices for output: output_prices is the transpose where the prices for outputs are.
output_prices <- t(Coelli_1998[, 7])
data_example2 <- make_deadata(Coelli_1998,
ni = 2,
no = 1)
result2 <- model_profit(data_example2,
price_output = output_prices,
rts = "crs",
restricted_optimal = FALSE)
# notice that the option by default is restricted_optimal = TRUE
efficiencies(result2)

# Example 3. Profit efficiency model.

data("Coelli_1998")
# Selection of prices for inputs and outputs: input_prices and output_prices are
# the transpose where the prices (for inputs and outputs) are.
input_prices <- t(Coelli_1998[, 5:6])
output_prices <- t(Coelli_1998[, 7])
data_example3 <- make_deadata(Coelli_1998,
ni = 2,
no = 1)
result3 <- model_profit(data_example3,
price_input = input_prices,
price_output = output_prices,
rts = "crs",
restricted_optimal = FALSE)
# notice that the option by default is restricted_optimal = TRUE
efficiencies(result3)

model_rdm Range directional model.

Description
Range directional model from Portela et al. (2004).

Usage
model_rdm(datadea,
dmu_eval = NULL,
dmu_ref = NULL,
orientation = c("no", "io", "oo"),
irdm = FALSE,
maxslack = TRUE,
weight_slack_i = 1,
model_rdm 81

weight_slack_o = 1,
compute_target = TRUE,
returnlp = FALSE,
...)

Note
Undesirable inputs/outputs are treated as negative inputs/outputs in this model.

References
Portela, M.; Thanassoulis, E.; Simpson, G. (2004). "Negative data in DEA: a directional distance
approach applied to bank branches", Journal of the Operational Research Society, 55 1111-1121.
82 model_sbmeff

model_sbmeff Slack based measure (SBM) of efficiency model.

Description
Calculate the SBM model proposed by Tone (2001).

Usage
model_sbmeff(datadea,
dmu_eval = NULL,
dmu_ref = NULL,
weight_input = 1,
weight_output = 1,
orientation = c("no", "io", "oo"),
rts = c("crs", "vrs", "nirs", "ndrs", "grs"),
L = 1,
U = 1,
kaizen = FALSE,
maxfr = NULL,
tol = 1e-6,
silent = FALSE,
compute_target = TRUE,
returnlp = FALSE,
...)

kaizen Logical. If TRUE, the kaizen version of SBM (Tone 2010), also known as SBM-
Max, is computed.
maxfr A list with the maximal friends sets, as it is returned by function maximal_friends.
If NULL (default) this list is computed internally.
tol Numeric, a tolerance margin for checking efficiency (only for the kaizen ver-
sion).
silent Logical. If FALSE (default) it prints all the messages from function maximal_friends.
compute_target Logical. If it is TRUE, it computes targets.
returnlp Logical. If it is TRUE, it returns the linear problems (objective function and
constraints). If kaizen is TRUE it is ignored.
... Other options (currently not implemented)

References
Tone, K. (2001). "A Slacks-Based Measure of Efficiency in Data Envelopment Analysis", European
Journal of Operational Research, 130, 498-509. doi:10.1016/S03772217(99)004075
Tone, K. (2010). "Variations on the theme of slacks-based measure of efficiency in DEA", European
Journal of Operational Research, 200, 901-907. doi:10.1016/[Link].2009.01.027
Cooper, W.W.; Seiford, L.M.; Tone, K. (2007). Data Envelopment Analysis. A Comprehensive
Text with Models, Applications, References and DEA-Solver Software. 2nd Edition. Springer,
New York. doi:10.1007/9780387452838
Aparicio, J.; Ruiz, J.L.; Sirvent, I. (2007) "Closest targets and minimum distance to the Pareto-
efficient frontier in DEA", Journal of Productivity Analysis, 28, 209-218. doi:10.1007/s11123007-
00395

See Also
model_nonradial, model_deaps, model_profit, model_sbmsupereff

Examples
# Example 1. Replication of results in Tone (2001, p.505)
data("Tone2001")
data_example <- make_deadata(Tone2001,
ni = 2,
no = 2)
result_SBM <- model_sbmeff(data_example,
orientation = "no",
84 model_sbmsupereff

rts = "crs")
result_CCR <- model_basic(data_example,
orientation = "io",
rts = "crs")
efficiencies(result_SBM)
efficiencies(result_CCR)
slacks(result_SBM)
slacks(result_CCR)

# Example 2. Replication of results in Tone (2003), pp 10-11 case 1:1.

data("Tone2003")
data_example <- make_deadata(Tone2003,
ni = 1,
no = 2,
ud_outputs = 2)
result <- model_sbmeff(data_example,
rts = "vrs")
efficiencies(result)
targets(result)

# Example 3. Replication of results in Aparicio (2007).

data("Airlines")
datadea <- make_deadata(Airlines,
inputs = 4:7,
outputs = 2:3)
result <- model_sbmeff(datadea = datadea, kaizen = TRUE)
efficiencies(result)
targets(result)

model_sbmsupereff Slack based measure of superefficiency model

Description
Slack based measure of superefficiency model (Tone 2002) with n DMUs, m inputs and s outputs.

Usage
model_sbmsupereff(datadea,
dmu_eval = NULL,
dmu_ref = NULL,
weight_input = 1,
weight_output = 1,
orientation = c("no", "io", "oo"),
rts = c("crs", "vrs", "nirs", "ndrs", "grs"),
L = 1,
U = 1,
compute_target = TRUE,
model_sbmsupereff 85

compute_rho = FALSE,
kaizen = FALSE,
silent = FALSE,
returnlp = FALSE)

Arguments

datadea A deadata object, including DMUs, inputs and outputs.

dmu_eval A numeric vector containing which DMUs have to be evaluated. If NULL (de-
fault), all DMUs are considered.
dmu_ref A numeric vector containing which DMUs are the evaluation reference set. If
NULL (default), all DMUs are considered.
weight_input A value, vector of length m, or matrix m x ne (where ne is the length of dmu_eval)
with weights to inputs corresponding to the relative importance of items.
weight_output A value, vector of length m, or matrix m x ne (where ne is the length of dmu_eval)
with weights to outputs corresponding to the relative importance of items.
orientation A string, equal to "no" (non-oriented), "io" (input-oriented) or "oo" (output-
oriented).
rts A string, determining the type of returns to scale, equal to "crs" (constant), "vrs"
(variable), "nirs" (non-increasing), "ndrs" (non-decreasing) or "grs" (general-
ized).
L Lower bound for the generalized returns to scale (grs).
U Upper bound for the generalized returns to scale (grs).
compute_target Logical. If it is TRUE, it computes targets, superslacks (t_input and t_output)
and slacks.
compute_rho Logical. If it is TRUE, it computes the SBM efficiency score (applying model_sbmeff)
of the DMU (project_input, project_output).
kaizen Logical. If TRUE, the kaizen version of SBM (Tone 2010), also known as SBM-
Max, is computed for the efficiency score of the DMU (project_input, project_output).
silent Logical. If FALSE (default) it prints all the messages from function maximal_friends.
returnlp Logical. If it is TRUE, it returns the linear problems (objective function and
constraints).

Author(s)

Vicente Coll-Serrano (<[Link]@[Link]>). Quantitative Methods for Measuring Culture

(MC2). Applied Economics.
Vicente Bolós (<[Link]@[Link]>). Department of Business Mathematics
Rafael Benítez (<[Link]@[Link]>). Department of Business Mathematics
University of Valencia (Spain)
86 model_supereff

References

Tone, K. (2002). "A slacks-based measure of super-efficiency in data envelopment analysis", Euro-
pean Journal of Operational Research, 143, 32-41. doi:10.1016/S03772217(01)003241
Tone, K. (2010). "Variations on the theme of slacks-based measure of efficiency in DEA", European
Journal of Operational Research, 200, 901-907. doi:10.1016/[Link].2009.01.027
Cooper, W.W.; Seiford, L.M.; Tone, K. (2007). Data Envelopment Analysis. A Comprehensive
Text with Models, Applications, References and DEA-Solver Software. 2nd Edition. Springer,
New York. doi:10.1007/9780387452838

See Also

model_sbmeff, model_supereff, model_addsupereff

Examples
# Replication of results in Tone(2002, p.39)
data("Power_plants")
data_example <- make_deadata(Power_plants,
ni = 4,
no = 2)
result <- model_sbmsupereff(data_example,
orientation = "io",
rts = "crs")
efficiencies(result)
slacks(result)$slack_input
references(result)

model_supereff Radial super-efficiency basic DEA model

Description

Solve Andersen and Petersen radial Super-efficiency DEA model.

Usage

model_supereff(datadea,
dmu_eval = NULL,
dmu_ref = NULL,
supereff_modelname = c("basic"),
...)
model_supereff 87

Arguments
datadea An object of class deadata.
dmu_eval A numeric vector containing which DMUs have to be evaluated. If NULL (de-
fault), all DMUs are considered.
dmu_ref A numeric vector containing which DMUs are the evaluation reference set. If
NULL (default), all DMUs are considered.
supereff_modelname
A string containing the name of the radial model to apply super-efficiency.
... orientation, rts and other model parameters.

Note
(1) Radial super-efficiency model under variable (vrs, nirs, ndrs, grs) returns to scale can be infea-
sible for certain DMUs. See example 2.
(2) DMUs with infeasible solution are not shown in the results.

References
Andersen, P.; Petersen, N.C. (1993). "A procedure for ranking efficient units in data envelopment
analysis", Management Science, 39, 1261-1264.
Tone, K. (2002). "A slacks-based measure of super-efficiency in data envelopment analysis", Euro-
pean Journal of Operational Research, 143, 32-41.

See Also
model_basic, model_sbmsupereff, model_addsupereff

Examples
# Example 1.
# Replication of results in Tone (2002, p.38)
data("Power_plants")
data_example <- make_deadata(Power_plants,
ni = 4,
no = 2)
result <- model_supereff(data_example,
orientation = "io",
rts = "crs")
eff <- efficiencies(result)
88 multipliers

# Example 2.
# Results of Super-efficiency with vrs returns to scale show infeasibility solutions
# for DMUs D4 and D6 (these DMUs are not shown in deaR results).
data("Power_plants")
data_example2 <- make_deadata(Power_plants,
ni = 4,
no = 2)
result2 <- model_supereff(data_example2,
orientation = "io",
rts = "vrs")
eff2 <- efficiencies(result2)

multipliers Multipliers

Description
Extract the multipliers of the DMUs from a dea or dea_fuzzy solution.

Usage
multipliers(deasol)

Arguments
deasol Object of class dea or dea_fuzzy obtained with some of the DEA model func-
tions.

Examples
data("Coll_Blasco_2006")
data_example <- make_deadata(Coll_Blasco_2006,
ni = 2,
no = 2)
result <- model_multiplier(data_example,
orientation = "io",
rts = "crs")
multipliers(result)
PFT1981 89

PFT1981 Data: Charnes, Cooper and Rhodes (1981).

Description
Data from Project Follow Through (PTF) in public school education. There are 49 DMUs (school
sites) in PFT and 21 DMUs in Non-Follow Through (NFT). Authors consider 3 outputs (Y) and 5
inputs (X).

Usage
data("PFT1981")

Format
Data frame with 70 rows and 10 columns. Definition of inputs (X) and outputs (Y):

Y1 = Reading Total Reading Scores (as measured by the Metropolitan Achievement Test).
Y2 = Math Total Math Scores (total mathematics score by the Metropolitan Achievement Test.
Y3 = Coopersmith Total Coopersmith Scores (Coopersmith self-esteem inventory, intended as a
measure of self-esteem).
X1 = Education Education level of mother (as measured in terms of percentage of high school
graduates among female parents).
X2 = Occupation Occupation Index (highest occupation of a family member according to a pre-
arranged rating scale).
X3 = Parental Parental Visit Index (representing the number of visits to the school site).
X4 = Counseling Counseling Index (parent counselling index calculated from data on time spent
with child on school-related topics such as reading together, etc.).
X5 = Teachers Number of Teachers (number of teachers at a given site).
Program PFT or NFT.

Source
Charnes, A.; Cooper, W.W.; Rhodes, E. (1981). "Evaluating Program and Managerial Efficiency:
An Application of Data Envelopment Analysis to Program Follow Through", Management Science,
27(6), 668-697. doi:10.1287/mnsc.27.6.668
90 [Link]

# Example 2. Replication of results in Charnes, Cooper and Rhodes (1981)

data("PFT1981")
# selecting DMUs in Non-Follow Through (NFT)
NFT <- PFT1981[50:70,]
NFT <- make_deadata(NFT,
dmus = 1,
inputs = 2:6,
outputs = 7:9 )
eval_nft <- model_basic(NFT,
orientation = "io",
rts = "crs")
eff_nft <- efficiencies(eval_nft)

[Link] Plot for DEA models.

Description
Plot some attribute of a DEA model.

Usage
## S3 method for class 'dea'
plot(x, tol = 1e-04, showPlots = TRUE, ...)

Arguments
x An object of class dea obtained by a DEA model function.
tol Numeric. Absolute tolerance for numeric comparisons. By default, it is 1e-4.
plot.dea_fuzzy 91

showPlots Logical. When TRUE (default) the plots are shown one by one. When it is
FALSE the plots are not shown and are returned by the function (invisibly) as a
list.
... Ignored, for compatibility issues.

Value

Depending on the model, it returns some plots.

Author(s)

Vicente Coll-Serrano (<[Link]@[Link]>). Quantitative Methods for Measuring Culture

(MC2). Applied Economics.
Vicente Bolós (<[Link]@[Link]>). Department of Business Mathematics
Rafael Benítez (<[Link]@[Link]>). Department of Business Mathematics
University of Valencia (Spain)

References

Zhu, J. (2014). Quantitative Models for Performance Evaluation and Benchmarking. Data Envel-
opment Analysis with Spreadsheets. 3rd Edition Springer, New York. doi:10.1007/9783319066479

Examples
data_example <- make_deadata(datadea = Fortune500,
inputs = 2:4,
outputs = 5:6)
result <- model_basic(data_example)
plot(result)

plot.dea_fuzzy Plot for fuzzy DEA models.

Description

Plot some attributes of a fuzzy DEA model (Guo-Tanaka, Kao-Liu and possibilistic models).

Usage

## S3 method for class 'dea_fuzzy'

plot(x, showPlots = TRUE, ...)
92 Power_plants

Arguments
x An object of class dea_fuzzy obtained by a fuzzy DEA model function.
showPlots Logical. When TRUE (default) the plots are shown one by one. When it is
FALSE the plots are not shown and are returned by the function (invisibly) as a
list.
... Ignored, for compatibility issues.

Value
Depending on the model, it returns some plots.

References
Zhu, J. (2014). Quantitative Models for Performance Evaluation and Benchmarking. Data Envel-
opment Analysis with Spreadsheets. 3rd Edition Springer, New York. doi:10.1007/9783319066479

Power_plants Data: Tone (2002).

Description
This dataset consists of six power plants with 4 inputs (X) and 2 outputs (Y).

Usage
data("Power_plants")

Format
Data frame with 15 rows and 7 columns. Definition of inputs (X) and outputs (Y):
x1 Manpower requiered
x2 Construction costs in millions of dollars
x3 Annual maintenance costs in millions of dollars
x4 Number of villages to be evacuated
y1 Power generated in megawatts
y2 Safety level
Power_plants 93

Source
Andersen, P.; Petersen, N.C. (1993). "A procedure for ranking efficient units in data envelopment
analysis", Management Science, 39, 1261-1264.
Doyle, J. and Green R. (1993). "Data envelopment analysis and multiple criteria decision making",
Omega, 21 (6), 713-715. doi:10.1016/03050483(93)90013B
Tone, K. (2002). "A slacks-based measure of super-efficiency in data envelopment analysis", Euro-
pean Journal of Operational Research, 143, 32-41. doi:10.1016/S03772217(01)003241

See Also
make_deadata, model_supereff, model_sbmsupereff

Examples
# Example 1. Radial super-efficiency model.
# Replication of results in Tone (2002)
data("Power_plants")
data_example <- make_deadata(Power_plants,
ni = 4,
no = 2)
result <- model_supereff(data_example,
orientation = "io",
rts = "crs")
eff <- efficiencies(result)
eff

# Example 2. SBM super-efficiency model.

data("Power_plants")
data_example <- make_deadata(Power_plants,
ni = 4,
no = 2)
result2 <- model_sbmsupereff(data_example,
orientation = "io",
rts = "crs")
efficiencies(result2)
slacks(result2)$input
references(result2)
94 print.deadata_fuzzy

[Link] deadata class print method

Description
Print method for deadata class.

Usage
## S3 method for class 'deadata'
print(x, ...)

Arguments
x A deadata object (as returned by make_deadata function).
... For compatibility issues.

print.deadata_fuzzy deadata_fuzzy class print method

Description
Print method for deadata_fuzzy class.

Usage
## S3 method for class 'deadata_fuzzy'
print(x, ...)

Arguments
x A deadata_fuzzy object (as returned by make_deadata_fuzzy function).
... For compatibility issues.
read_data 95

read_data read_data

Description
This function is deprecated. Use make_deadata instead.

Usage
read_data(datadea = NULL,
ni = NULL,
no = NULL,
dmus = 1,
inputs = NULL,
outputs = NULL,
nc_inputs = NULL,
nc_outputs = NULL,
nd_inputs = NULL,
nd_outputs = NULL,
ud_inputs = NULL,
ud_outputs = NULL)

nc_inputs A numeric vector containing the indices of non-controllable inputs.

read_data_fuzzy read_data_fuzzy

Description
This function is deprecated. Use make_deadata_fuzzy instead.

Usage
read_data_fuzzy(datadea,
dmus = 1,
[Link] = NULL,
[Link] = NULL,
[Link] = NULL,
[Link] = NULL,
[Link] = NULL,
[Link] = NULL,
[Link] = NULL,
[Link] = NULL,
nc_inputs = NULL,
nc_outputs = NULL,
nd_inputs = NULL,
nd_outputs = NULL,
ud_inputs = NULL,
ud_outputs = NULL)

[Link] Where are (columns) the mR (right centers) of trapezoidal fuzzy inputs in datadea.
If an input is triangular or crisp, we put NA.
Alternatively to datadea, [Link] can be a matrix of size (number of inputs
x number of DMUs) with the mR of trapezoidal fuzzy inputs, the centers of tri-
angular inputs, and the crisp values of crisp inputs. If all inputs are triangular or
crisp, then [Link] must be NULL (default) or equal to [Link].
[Link] Where are (columns) the dL (left radii) of trapezoidal and triangular fuzzy inputs
in datadea. If an input is symmetric, we put the column where the radii are. If
an input is rectangular or crisp, we put NA.
Alternatively to datadea, [Link] can be a matrix of size (number of inputs
x number of DMUs) with the dL of trapezoidal and triangular fuzzy inputs. If
an input is rectangular or crisp, its radius is zero. If all inputs are rectangular or
crisp, then [Link] must be NULL (default) or a zero matrix.
[Link] Where are (columns) the dR (right radii) of trapezoidal and triangular fuzzy in-
puts in datadea. If an input is symmetric, rectangular or crisp, we put NA.
Alternatively to datadea, [Link] can be a matrix of size (number of in-
puts x number of DMUs) with the dR of trapezoidal and triangular fuzzy inputs.
If an input is rectangular or crisp, its radius is zero. If all inputs are symmet-
ric, rectangular or crisp, then [Link] must be NULL (default) or equal to
[Link].
[Link] Analogous to [Link], but relating to outputs.
[Link] Analogous to [Link], but relating to outputs.
[Link] Analogous to [Link], but relating to outputs.
[Link] Analogous to [Link], but relating to outputs.
nc_inputs A numeric vector containing the indices of non-controllable inputs.
nc_outputs A numeric vector containing the indices of non-controllable outputs.
nd_inputs A numeric vector containing the indices of non-discretionary inputs.
nd_outputs A numeric vector containing the indices of non-discretionary outputs.
ud_inputs A numeric vector containing the indices of undesirable (good) inputs.
ud_outputs A numeric vector containing the indices of undesirable (bad) outputs.

read_malmquist read_malmquist

Description
This function is deprecated. Use make_malmquist instead.

Usage
read_malmquist(datadea,
nper = NULL,
percol = NULL,
arrangement = c("horizontal", "vertical"),
...)
98 references

Arguments
datadea Data frame with DEA data.
nper Number of time periods (with dataset in wide format).
percol Column of time period (with dataset in long format).
arrangement Horizontal with data in wide format. Vertical with data in long format.
... Other options to be passed to the make_deadata function.

references References

Description
Extract the reference set for each DMU (inefficient DMUs and efficicent DMUs that are combina-
tion of other efficient DMUs) from a DEA model solution.

Usage
references(deasol,
thr = 1e-4)

Arguments
deasol Object of class dea obtained with some of the DEA model functions.
thr Tolerance threshold (for avoiding miss detection of efficient DMUs due to round
off errors)

Examples
# Replication results model DEA1 in Tomkins and Green (1988).
data("Departments")
# Calculate Total income
Departments$Total_income <- Departments[, 5] + Departments[, 6] + Departments[, 7]
data_DEA1 <- make_deadata(Departments,
inputs = 9,
outputs = c(2, 3, 4, 12))
result <- model_basic(data_DEA1,
orientation = "io",
rts 99

rts = "crs")
references(result) # Table 3 (p.157)

rts Returns to scale

Description

Extract the returns to scale.

Usage

rts(deamodel,
thr = 1e-4)

Arguments

deamodel Object of class dea obtained with some of the DEA model functions.
thr Threshold for the tolerance for considering something equal to 1. Defaults to
1e-4.

Author(s)

Vicente Coll-Serrano (<[Link]@[Link]>). Quantitative Methods for Measuring Culture

(MC2). Applied Economics.
Vicente Bolós (<[Link]@[Link]>). Department of Business Mathematics
Rafael Benítez (<[Link]@[Link]>). Department of Business Mathematics
University of Valencia (Spain)

Examples
data("Coll_Blasco_2006")
data_example <- make_deadata(Coll_Blasco_2006,
ni = 2,
no = 2)
result <- model_basic(data_example,
orientation = "io",
rts ="crs")
rts(result)
100 Ruggiero2007

Ruggiero2007 Data: Ruggiero (2007).

Description
Simulated data of 35 DMUs with two inputs and one output.

Usage
data("Ruggiero2007")

Format
Data frame with 35 rows and 4 columns. Definition of inputs (X) and outputs (Y):

x1 Input 1
x2 Input 2
y1 Output 1

Source
Ruggiero J. (2007). Non-Discretionary Inputs. In: Zhu J., Cook W.D. (eds) Modeling Data Ir-
regularities and Structural Complexities in Data Envelopment Analysis. Springer, Boston, MA.
doi:10.1007/9780387716077_5

Extract the slacks of the DMUs from a dea or dea_fuzzy solution.

Usage

slacks(deasol)

Arguments

deasol Object of class dea or dea_fuzzy obtained with some of the DEA model func-
tions.

Author(s)

Vicente Coll-Serrano (<[Link]@[Link]>). Quantitative Methods for Measuring Culture

(MC2). Applied Economics.
Vicente Bolós (<[Link]@[Link]>). Department of Business Mathematics
Rafael Benítez (<[Link]@[Link]>). Department of Business Mathematics
University of Valencia (Spain)

Examples
data("Coll_Blasco_2006")
data_example <- make_deadata(Coll_Blasco_2006,
ni = 2,
no = 2)
result <- model_multiplier(data_example,
orientation = "io",
rts = "crs")
slacks(result)
102 [Link]

[Link] Summary conventional DEA models.

Description
Summary of the results obtained by a conventional DEA model.

Usage
## S3 method for class 'dea'
summary(object, exportExcel = FALSE, filename = NULL, returnList = FALSE, ...)

Arguments
object An object of class "dea" obtained by a DEA model function.
exportExcel Logical value. If TRUE (FALSE by default) the results are also exported to an
Excel file.
filename Character string. Absolute file name (including path) of the exported Excel file.
If NULL, then the file name will be "ResultsDEA" + [Link].
returnList Logical value. If TRUE then the results are given as a list of data frames. If
FALSE (default) all the data frames are merged into a single data frame.
... Ignored. Used for compatibility issues.

Value
Depending on the model it returns a single [Link] containing: efficiencies, slacks, lambdas, tar-
gets, references or a list of [Link] with the cross-efficiencies computed with different methods
(Arbitrary, Method II or Method III (see CITA)) or, in case the model is a Malmquist index, a single
[Link] with the coefficients for the different periods.

References
Charnes, A.; Cooper, W.W.; Rhodes, E. (1981). "Evaluating Program and Managerial Efficiency:
An Application of Data Envelopment Analysis to Program Follow Through", Management Science,
27(6), 668-697. doi:10.1287/mnsc.27.6.668
summary.dea_fuzzy 103

Examples
data("PFT1981")
# Selecting DMUs in Program Follow Through (PFT)
PFT <- PFT1981[1:49, ]
PFT <- make_deadata(PFT,
inputs = 2:6,
outputs = 7:9 )
eval_pft <- model_basic(PFT,
orientation = "io",
rts = "crs")
summary(eval_pft)

summary.dea_fuzzy Summary Fuzzy DEA models.

Description
Summary of the results obtained by a fuzzy DEA model.

Usage
## S3 method for class 'dea_fuzzy'
summary(object, ..., exportExcel = FALSE, filename = NULL, returnList = FALSE)

Arguments
object An object of class "dea_fuzzy" obtained with a fuzzy DEA model function
(modelfuzzy_guotanaka, modelfuzzy_kaoliu, modelfuzzy_possibilistic).
... Extra options.
exportExcel Logical value. If TRUE (FALSE by default) the results are also exported to an
Excel file.
filename Character string. Absolute file name (including path) of the exported Excel file.
If NULL, then the file name will be "ResultsDEA" + [Link].
returnList Logical value. If TRUE then the results are given as a list of data frames. If
FALSE (default) all the data frames are merged into a single data frame.

Value
If the model is that from Guo and Tanaka (modelfuzzy_guotanaka), it returns a [Link] with
columns: DMU, alpha cuts and efficiencies. For the possibilistic model (modelfuzzy_possibilistic)
it returns a [Link] with columns: DMU, alpha-cuts, efficiencies and the corresponding lambda
values For the Kao-Liu model (modelfuzzy_kaoliu), the result may depend on the crisp sub-model
used. It will contain a [Link] with the efficiencies (if any), the slacks and superslacks (if any),
the lambda values and the targets.
If exportExcel is TRUE, then an Excel file will be created containing as many sheets as necessary
depending on the variables returned.
104 Supply_Chain

References
Leon, T.; Liern, V. Ruiz, J.; Sirvent, I. (2003). "A Possibilistic Programming Approach to the
Assessment of Efficiency with DEA Models", Fuzzy Sets and Systems, 139, 407–419. doi:10.1016/
S01650114(02)006085

Examples
data("Leon2003")
data_example <- make_deadata_fuzzy(Leon2003,
[Link] = 2,
[Link] = 3,
[Link] = 4,
[Link] = 5)
result <- modelfuzzy_possibilistic(data_example,
h = seq(0, 1, by = 0.1),
orientation = "io",
rts = "vrs")
summary(result)

Supply_Chain Data: Sanei and Mamizadeh Chatghayeb (2013).

Description
Data of 17 supply chain (buyer-supplier relationship in manufacturing).

Usage
data("Supply_Chain")

Format
Data frame with 17 rows and 8 columns. Definition of inputs (X) and outputs (Y):

X1 to X3 Inputs of buyers
I1 to I2 Outputs of buyers, Inputs of suppliers
Y1 to Y2 Outputs of suppliers
targets 105

Source
Sanei, M.; Mamizadeh Chatghayeb, S. (2013). “Free Disposal Hull Models in Supply Chain Man-
agement”, International Journal of Mathematical Modelling and Computations, 3(3), 125-129.

Tone2001 Data: Tone (2001).

Description
Data of 5 DMUs producing 2 outputs by using 2 inputs

Usage
data("Tone2001")

Format
Data frame with 5 rows and 5 columns. Definition of inputs (X) and outputs (Y):
x1 Input1
x2 Input2
y1 Output1
y2 Output2

Source
Tone, K. (2001). "A Slacks-Based Measure of Efficiency in Data Envelopment Analysis", European
Journal of Operational Research, 130, 498-509. doi:10.1016/S03772217(99)004075

Tone2003 Data: Tone (2003).

Description
Data of 9 DMUs producing 2 outputs, being second output undesirable, by using 1 input.

Usage
data("Tone2003")

Format
Data frame with 9 rows and 4 columns. Definition of inputs (X) and outputs (Y):
x Input
yg Output1 ("good" output)
yb Output2 (undesirable "bad" output)

Source
Tone, K. (2003). "Dealing with undesirable outputs in DEA: A Slacks-Based Measure (SBM)
approach", GRIPS Research Report Series I-2003-0005.

undesirable_basic Undesirable inputs and outputs for basic DEA model.

Description
This function transforms a deadata or deadata_fuzzy class with undesirable inputs/outputs accord-
ing to Seiford and Zhu (2002). Onwards, it is recommended to use a DEA model with variable
returns to scale (vrs).

Usage
undesirable_basic(datadea,
vtrans_i = NULL,
vtrans_o = NULL)

Arguments
datadea A deadata object, including DMUs, inputs and outputs.
vtrans_i Numeric vector of translation for undesirable inputs. If vtrans_i[i] is NA, then
it applies the "max + 1" translation to the i-th undesirable input. If vtrans_i
is a constant, then it applies the same translation to all undesirable inputs. If
vtrans_i is NULL, then it applies the "max + 1" translation to all undesirable
inputs.
vtrans_o Numeric vector of translation for undesirable outputs, analogous to vtrans_i,
but applied to outputs.
undesirable_basic 109

Value
An list with the transformed object of class deadata or deadata_fuzzy and the corresponding
translation vectors vtrans_i and vtrans_o.

References
Seiford, L.M.; Zhu, J. (2002). “Modeling undesirable factors in efficiency evaluation”, European
Journal of Operational Research 142, 16-20.
Hua Z.; Bian Y. (2007). DEA with Undesirable Factors. In: Zhu J., Cook W.D. (eds) Modeling
Data Irregularities and Structural Complexities in Data Envelopment Analysis. Springer, Boston,
MA.

Examples
data("Hua_Bian_2007")
# The third output is an undesirable output.
data_example <- make_deadata(Hua_Bian_2007,
ni = 2,
no = 3,
ud_outputs = 3)
# rts must be "vrs" for undesirable inputs/outputs:
# Translation parameter is set to (max + 1)
result <- model_basic(data_example,
orientation = "oo",
rts = "vrs")
Index

∗ datasets EconomyLong, 16
Airlines, 3 eff_dmus, 20
Coelli_1998, 6 efficiencies, 17
Coll_Blasco_2006, 7 [Link], 18
Departments, 13 efficiencies.dea_fuzzy, 19
Doyle_Green_1994, 14 Electric_plants, 21
Economy, 15 extreme_efficient, 22, 63
EconomyLong, 16
Electric_plants, 21 Fortune500, 23
Fortune500, 23 Fried1993, 24
Fried1993, 24 FuzzyExample, 25
FuzzyExample, 25
Golany_Roll_1989, 27 Golany_Roll_1989, 27
Grifell_Lovell_1999, 28 Grifell_Lovell_1999, 28
Guo_Tanaka_2001, 29 Guo_Tanaka_2001, 29
Hotels, 30
Hotels, 30
Hua_Bian_2007, 32
Hua_Bian_2007, 32
Kao_Liu_2003, 37
Leon2003, 39 [Link], 33
Libraries, 40 is.dea_fuzzy, 35
Lim_Zhu_2015, 42 [Link], 34
PFT1981, 89 is.deadata_fuzzy, 34
Power_plants, 92 [Link], 36, 53
Ruggiero2007, 100
Supply_Chain, 104 Kao_Liu_2003, 37
Tone2001, 106
Tone2003, 107 lambdas, 38
Leon2003, 39
Airlines, 3 Libraries, 40
Lim_Zhu_2015, 42
bootstrap_basic, 4
make_deadata, 4, 7, 8, 13, 15, 21, 24, 25, 27,
Coelli_1998, 6
31, 32, 41, 43, 43, 90, 93, 100, 105,
Coll_Blasco_2006, 7
107, 108
cross_efficiency, 8, 15, 27, 43, 75
make_deadata_fuzzy, 26, 30, 38, 40, 45
cross_efficiency_fuzzy, 10, 11, 30, 40, 54
make_malmquist, 16, 17, 28, 47
Departments, 13 malmquist_index, 16, 17, 28, 49
Doyle_Green_1994, 14 maximal_friends, 36, 52, 63
model_additive, 58, 63, 66
Economy, 15 model_addmin, 61

110
INDEX 111

model_addsupereff, 60, 64, 86, 87

model_basic, 13, 21, 25, 32, 38, 41, 54, 56,
58, 66, 75, 87, 90, 100
model_deaps, 70, 78, 79, 83
model_fdh, 72, 105
model_multiplier, 10, 15, 24, 27, 43, 54, 56,
69, 74
model_nonradial, 31, 72, 76, 79, 83
model_profit, 72, 78, 78, 83
model_rdm, 80
model_sbmeff, 4, 36, 53, 72, 78, 79, 82, 86,
107, 108
model_sbmsupereff, 66, 83, 84, 87, 93
model_supereff, 66, 69, 86, 86, 93
modelfuzzy_guotanaka, 30, 40, 53, 56, 58
modelfuzzy_kaoliu, 54, 55, 58
modelfuzzy_possibilistic, 40, 54, 56, 57
multipliers, 88

PFT1981, 89
[Link], 90
plot.dea_fuzzy, 91
Power_plants, 92
[Link], 94
print.deadata_fuzzy, 94

read_data, 95
read_data_fuzzy, 96
read_malmquist, 97
references, 98
rts, 99
Ruggiero2007, 100

slacks, 101
[Link], 102
summary.dea_fuzzy, 103
Supply_Chain, 104

targets, 105
Tone2001, 106
Tone2003, 107

undesirable_basic, 108

DeaR Conventional and Fuzzy Data Envelopment Analysis
No ratings yet
DeaR Conventional and Fuzzy Data Envelopment Analysis
117 pages
Benchmarking
No ratings yet
Benchmarking
59 pages
Frontier Efficiency Analysis in R
No ratings yet
Frontier Efficiency Analysis in R
53 pages
Fundamental Data Structures
No ratings yet
Fundamental Data Structures
277 pages
MIRT: Multidimensional IRT Package
No ratings yet
MIRT: Multidimensional IRT Package
170 pages
Performance Metrics for Regression Models
No ratings yet
Performance Metrics for Regression Models
99 pages
MIRT Package Overview on GitHub
No ratings yet
MIRT Package Overview on GitHub
232 pages
ANOVA and Design Functions in R
No ratings yet
ANOVA and Design Functions in R
129 pages
Effective Pandas: Patterns For Data Manipulation 1st Edition Matt Harrison Ebook
100% (3)
Effective Pandas: Patterns For Data Manipulation 1st Edition Matt Harrison Ebook
211 pages
Effective Pandas: Patterns for Data Manipulation 1st Edition Matt Harrison resource guide
100% (1)
Effective Pandas: Patterns for Data Manipulation 1st Edition Matt Harrison resource guide
212 pages
COPASI User Guide and Tutorial
No ratings yet
COPASI User Guide and Tutorial
139 pages
Mystic: Non-Convex Optimization Tools
No ratings yet
Mystic: Non-Convex Optimization Tools
218 pages
Mesa Agent-Based Modeling Guide
No ratings yet
Mesa Agent-Based Modeling Guide
5 pages
Mirt
No ratings yet
Mirt
200 pages
Social Network Analysis Tools in R
No ratings yet
Social Network Analysis Tools in R
187 pages
Olson2017 Book DescriptiveDataMining
No ratings yet
Olson2017 Book DescriptiveDataMining
120 pages
Discretization Algorithms for Classification
No ratings yet
Discretization Algorithms for Classification
24 pages
Effective Pandas: Data Manipulation Guide
No ratings yet
Effective Pandas: Data Manipulation Guide
391 pages
MIRT: Multidimensional IRT Package
No ratings yet
MIRT: Multidimensional IRT Package
183 pages
Zero To Deep Learning
100% (6)
Zero To Deep Learning
753 pages
Simulation and Analysis of Random Fields
No ratings yet
Simulation and Analysis of Random Fields
374 pages
EFA.dimensions: Factor Analysis Tools
No ratings yet
EFA.dimensions: Factor Analysis Tools
56 pages
Mike 21 Flow Model FM
No ratings yet
Mike 21 Flow Model FM
144 pages
sfaR: Stochastic Frontier Analysis Tools
No ratings yet
sfaR: Stochastic Frontier Analysis Tools
64 pages
Pavilion CalculationReferenceManual
No ratings yet
Pavilion CalculationReferenceManual
231 pages
K Rip A Shankar Thesis
No ratings yet
K Rip A Shankar Thesis
63 pages
What-If Analysis Dashboard Guide
No ratings yet
What-If Analysis Dashboard Guide
32 pages
Effective Pandas Patterns For Data Manipulation Treading On Python Matt Harrison Independently Published 2021
100% (2)
Effective Pandas Patterns For Data Manipulation Treading On Python Matt Harrison Independently Published 2021
40 pages
Dynare
No ratings yet
Dynare
74 pages
Dynare
No ratings yet
Dynare
74 pages
Continuous-Time Approaches For Short-Term Scheduling of Network Batch Processes: Small-Scale and Medium-Scale Problems
No ratings yet
Continuous-Time Approaches For Short-Term Scheduling of Network Batch Processes: Small-Scale and Medium-Scale Problems
22 pages
Pavilion ControlReferenceManual
No ratings yet
Pavilion ControlReferenceManual
118 pages
Spatial Capture-Recapture Tools in R
No ratings yet
Spatial Capture-Recapture Tools in R
336 pages
RTC-Tools: Real-Time Control Manual
No ratings yet
RTC-Tools: Real-Time Control Manual
76 pages
ASCEND Modelling Tutorial Guide
No ratings yet
ASCEND Modelling Tutorial Guide
139 pages
Manuel Référence Dynare v4.3.3
No ratings yet
Manuel Référence Dynare v4.3.3
99 pages
Statistical Methods for Quality Science
No ratings yet
Statistical Methods for Quality Science
147 pages
AlphaPy Machine Learning Framework
No ratings yet
AlphaPy Machine Learning Framework
180 pages
Manual Dynare
No ratings yet
Manual Dynare
99 pages
An Embarrassment of Pandas
No ratings yet
An Embarrassment of Pandas
16 pages
ANDES Manual Overview and Installation
No ratings yet
ANDES Manual Overview and Installation
899 pages
R Package for Applied Regression
No ratings yet
R Package for Applied Regression
147 pages
Zero To Deep Learning With Keras and Tensorflow (Francesco Mosconi)
No ratings yet
Zero To Deep Learning With Keras and Tensorflow (Francesco Mosconi)
769 pages
Zero To Deep Learning With Keras and Tensorflow Compress
No ratings yet
Zero To Deep Learning With Keras and Tensorflow Compress
769 pages
TOSA 0.23.0 Specification Overview
No ratings yet
TOSA 0.23.0 Specification Overview
86 pages
Lazy Trade
No ratings yet
Lazy Trade
71 pages
GSC Geochemistry EDA Package
No ratings yet
GSC Geochemistry EDA Package
209 pages
Moocore
No ratings yet
Moocore
47 pages
Econometric Tools for Risk Analysis
No ratings yet
Econometric Tools for Risk Analysis
240 pages
Statistical Analysis Package HH
No ratings yet
Statistical Analysis Package HH
124 pages
Lattice: Trellis Graphics for R
No ratings yet
Lattice: Trellis Graphics for R
159 pages
Actuarial Functions in R Package
No ratings yet
Actuarial Functions in R Package
91 pages
exdata
No ratings yet
exdata
218 pages
Meaning of 'R' in tracem-p
100% (1)
Meaning of 'R' in tracem-p
94 pages
Eco Lab User Guide
100% (1)
Eco Lab User Guide
62 pages
Species Distribution Modelling in R
No ratings yet
Species Distribution Modelling in R
60 pages
Pyomo Documentation Version 4.1
No ratings yet
Pyomo Documentation Version 4.1
113 pages
Pyomo Documentation Version 5.1.1
No ratings yet
Pyomo Documentation Version 5.1.1
113 pages