Title: DEA (Data Envelopment Analysis)
1DEA (Data Envelopment Analysis)
- Toshiyuki Sueyoshi
- New Mexico Tech
- Dept. of Management
2Data Envelopment Analysis
- (1) Relative Comparison
- (2) Multiple Inputs and Outputs
- (3) Efficiency Measurement (0-100)
- (4) Avoid the Specification Error between
- Inputs and Outputs
- (5) Production/Cost Analysis
3Case 1 input 1 output
Table 1.1 1 input 1 output Case
4Efficiency Frontier
E
G
Output
F
C
A
H
B
D
0
Employees
Figure 1.1Comparison of efficiencies in 1
input1 output case
5Efficiency Frontier
E
G
Output
F
C
A
Regression Line
H
B
D
0
Employees
Figure 1.2 Regression Line and Efficiency
Frontier
6Table 1.2 Efficiency
1 C gt G gt Agt B gt E gt D F gt H 0.4
7Efficiency Frontier
Output
C
D2
D1
D
0
Employees
Figure 1.3 Improvement of Company D
8Case 2 inputs 1 output
Table 1.3 2 inputs 1 output Case
9Production Possibility Set
G
F
C
A
I
Offices/Sales
E
D
B
Efficiency Frontier
H
0
Employees/Sales
Figure 1.4 2 inputs 1 output Case
10C
A
Offices/Sales
A1
A2
B
0
Employees/Sales
Figure 1.5 Improvement of Company A
11Case 1 input 2 outputs
Table 1.4 1 input 2 outputs Case
12A1
B
C
A
Efficiency Frontier
D
F
Offices/Sales
Production Possibility Set
E1
G
E
0
Customers/Offices
Figure 1.6 1 input 2 outputs Case
13Case Multiple inputs Multiple outputs
Table 1.5 Example of Multiple inputsMultiple
outputs Case
14(No Transcript)
15CCR model
16R A Reference Set
17Primal Problem
18Dual Problem
19Slack
20Reference Set
21Example Problem
Table 1.6 2 inputs 1 output Case
22Primal Problem
23Dual Problem
24Efficiency Frontier
E
A
D
A1
F
C
0
Figure 1.7 Efficiency of DMU A
25(No Transcript)
26Table 1.7 Results of DEA
27BCC model
- Variable Returns to Scale
28CCR model
29BCC model
30Dual Problem
BCC model
31Efficiency Frontier of CCR model
Efficiency Frontier of BCC model
(B)
d
c
Output
b
(C)
a
(A)
0
Input
Figure 2.1 Efficiency Frontier and Production
Possibility Set
32Cost Analysis
33Efficiency Frontier of CCR model
Efficiency Frontier of BCC model
f
g
g
E
b
c
b
h
h
i
i
d
e
j
j
k
P
0
34Dual Problem