Performance Evaluation
using DEA
Example: Automobile Firm
An Automobile firm is trying to determine relative efficiency of its six
factories. The firm is evaluating performances on Total Business and Total
Income as a function of Operating Expenses and Number of Employees.
Objective is to conduct a proper study to set a benchmark and goal for each of
the factories, based on the performances of the current year.
2
Example: Automobile Firm
Total Business Total Income Operating Expenses Number of
Factory
(INR in Lacs) (INR in Lacs) (INR in Lacs) Employees
1 4013 300 41 70
2 3353 235 35 60
3 4141 237 37 60
4 4403 194 39 60
5 5263 340 50 100
6 3119 199 32 80
7 2631 153 26 70
8 2995 164 42 80
9 1719 98 21 65
Which factory is performing better ?
3
Criteria for Performance Measurements
1. Total Business Turnover
2. Total Income
3. Operating Expenses
4. Total number of Employees
4
Criteria for Performance Measurements
1. Total Business Turnover (Output criteria – higher the better)
2. Total Income (Output criteria – higher the better)
3. Operating Expenses (Input criteria – lower the better)
4. Total number of Employees (Input criteria – lower the better)
5
How to determine the efficiency
score?
6
How to determine the efficiency
score?
Weighted sum of outputs
Efficiency =
Weighted sum of inputs
7
Model Development
Maximize efficiency score of factory1:
v1O11 + v2O21
max
u1 I11 + u2 I 21
v1O1i + v2O2i
s.t. 1 for all i = 1, 2,...,8
u1 I1i + u2 I 2i
v1 , v2 , u1 , u2 0
8
Model Development
Maximize efficiency score of factory2:
v1O12 + v2O22
max
u1 I12 + u2 I 22
v1O1i + v2O2i
s.t. 1 for all i = 1, 2,...,9
u1 I1i + u2 I 2i
v1 , v2 , u1 , u2 0
9
Model Development
Maximize efficiency score of factory p:
v1O1 p + v2O2 p
max
u1 I1 p + u2 I 2 p
v1O1i + v2O2i
s.t. 1 for all i = 1, 2,...,9
u1 I1i + u2 I 2i
v1 , v2 , u1 , u2 0
10
Generalized Model
Assume there are n decision making units (DMUs), each with m inputs and s
outputs. The relative efficiency score of a test DMU p is obtained by solving
the following model:
s
v O
k =1
k kp
Oki: Amount of output k produced by DMU i
max Iji: Amount of input j utilized by DMU i
m
u j I jp
j =1
vk: Weight given to output k
uj: Weight given to input j
s
v O k ki
s.t. k =1
m
1 for all i = 1, 2,..., n
u I
j =1
j ji
vk , u j 0 for all k = 1, 2,..., s; j = 1, 2,..., m
11
Generalized Model
➢ The above model is called DEA (Data Envelopment Analysis) model.
➢ DEA is receiving increasing importance as a tool for evaluating and
improving the performance of manufacturing and service operations.
➢ It has been extensively applied in performance evaluation and
benchmarking of schools, hospitals, bank branches, production plants, etc.
➢ DEA measures the relative efficiencies of organizations with multiple
inputs and multiple outputs. The organizations are called the decision-
making units, or DMUs.
➢ DEA assigns weights to the inputs and outputs of a DMU that give it the
best possible efficiency. It thus arrives at a weighting of the relative
importance of the input and output variables that reflects the emphasis
that appears to have been placed on them for that particular DMU.
12
How to solve a DEA model?
13
DEA Model
Assume there are n decision making units (DMUs), each with m inputs and s
outputs. The relative efficiency score of a test DMU p is obtained by solving
the following model:
s
v O
k =1
k kp
Oki: Amount of output k produced by DMU i
max Iji: Amount of input j utilized by DMU i
m
u I
j =1
j jp
vk: Weight given to output k
uj: Weight given to input j
s
v O k ki
s.t. k =1
m
1 for all i = 1, 2,..., n
u I
j =1
j ji
vk , u j 0 for all k = 1, 2,..., s; j = 1, 2,..., m
14
Solve a DEA Model
The fractional program shown in previous slide can be converted to a linear
program as shown below:
s
max vk Okp Oki: Amount of output k produced by DMU i
k =1
Iji: Amount of input j utilized by DMU i
m
s.t. u I
j =1
j jp =1 vk: Weight given to output k
uj: Weight given to input j
s m
v O − u I
k =1
k ki
j =1
j ji 0 for all i = 1, 2,..., n
vk , u j 0 for all k = 1, 2,..., s; j = 1, 2,..., m
Solution of DEA Model
16
Solution of DEA Model
Factory v1 v2 u1 u2 Efficiency
1 0.000069 0.002408 0.000010 0.014280 1.000
2 0.000136 0.002090 0.028554 0.000010 0.946
3 0.000081 0.002809 0.000010 0.016661 1.000
4 0.000220 0.000151 0.025626 0.000010 1.000
5 0.000095 0.001463 0.019980 0.000010 0.997
6 0.000148 0.002286 0.031225 0.000010 0.917
7 0.000183 0.002813 0.038435 0.000010 0.911
8 0.000205 0.000141 0.023790 0.000010 0.636
9 0.000409 0.000280 0.047588 0.000010 0.731
v1: Weight given to output 1– Total Business; v2: Weight given to output 2– Total Income
u1: Weight given to input 1 – Operating Expenses; u2: Weight given to input 2 – No. of Employees
✓ DEA assigns weights to the inputs and outputs of a DMU that give it the best possible efficiency.
✓ It thus arrives at a weighting of the relative importance of the input and output variables that
reflects the emphasis that appears to have been placed on them for that particular DMU.
DEA &
Efficiency Frontier
18
Data Envelopment Analysis
◼ Data Envelopment Analysis (DEA) measures the relative efficiencies of
organizations with multiple inputs and multiple outputs. The organizations are
called the decision-making units, or DMUs.
◼ DEA assigns weights to the inputs and outputs of a DMU that give it the best
possible efficiency. It thus arrives at a weighting of the relative importance of
the input and output variables that reflects the emphasis that appears to have
been placed on them for that particular DMU.
◼ At the same time, though, DEA then gives all the other DMUs the same
weights and compares the resulting efficiencies with that for the DMU of focus.
19
Data Envelopment Analysis
◼ If the focus DMU looks at least as good as any other DMU, it receives a
maximum efficiency score. But if some other DMU looks better than the focus
DMU, the weights having been calculated to be most favorable to the focus
DMU, then it will receive an efficiency score less than maximum.
20
Graphical Illustration
◼ To illustrate, consider seven DMUs which each have one input and one output: L1 =
(2,2), L2 = (3,5), L3 = (6,7), L4 = (9,8), L5 = (5,3), L6 = (4,1), L7 = (10,7).
9
L4
8
To read more about L3 L7
7
efficiency frontier,
6
please refer the L2
Output
5
attached papers. 4
L5
3
L1
2
L6
EJOR Paper 1
0
0 1 2 3 4 5 6 7 8 9 10 11
Input
Management
Science Paper
21
Graphical Illustration
◼ DEA identifies the units in the comparison set which lie at the top and to the
left, as represented by L1, L2, L3, and L4. These units are called the efficient
units, and the line connecting them is called the "envelopment surface" because
it envelops all the cases.
◼ DMUs L5 through L7 are not on the envelopment surface and thus are
evaluated as inefficient by the DEA analysis. There are two ways to explain
their weakness. One is to say that, for example, L5 could perhaps produce as
much output as it does, but with less input (comparing with L1 and L2); the
other is to say it could produce more output with the same input (comparing
with L2 and L3).
Graphical Illustration
◼ Thus, there are two possible definitions of efficiency depending on the purpose
of the evaluation. One might be interested in possible reduction of inputs (in
DEA this is called the input orientation) or augmentation of outputs (the output
orientation) in achieving technical efficiency. Depending on the purpose of the
evaluation, the analysis provides different sets of peer groups to which to
compare.
◼ However, there are times when reduction of inputs or augmentation of outputs
is not sufficient. In our example, even when L6 reduces its input from 4 units to
2, there is still a gap between it and its peer L1 in the amount of one unit of
output. In DEA, this is called the "slack" which means excess input or missing
output that exists even after the proportional change in the input or the outputs.
Thank You !!!
