Professional Documents
Culture Documents
SIG Pertambangan
(Theory of Spatial Analysis : Metode AHP)
Oleh :
Irvani
Referensi :
Materi/Pokok Bahasan
I Pendahuluan (P.1)
Teori AHP 1
AHP
Information is decomposed into a hierarchy of
alternatives and criteria
Information is then synthesized to determine
relative ranking of alternatives
Both qualitative and quantitative information can
be compared using informed judgements to
derive weights and priorities
Hierarchical tree
Selecting
a New Car
Style
Reliability
- Civic
- Saturn
- Escort
- Miata
Fuel Economy
- Civic
- Saturn
- Escort
- Miata
- Civic
- Saturn
- Escort
- Miata
Ranking of criteria
Weights?
AHP
pair-wise relative importance
[1:Equal, 3:Moderate, 5:Strong, 7:Very
strong, 9:Extreme]
Style
Reliability
Fuel Economy
Style
1/1
1/2
3/1
Reliability
2/1
1/1
4/1
Fuel Economy
1/3
1/4
1/1
Ranking of priorities
Eigenvector
[Ax = x]
Iterate
1. Take successive squared powers of matrix
2. Normalize the row sums
Until difference between successive row sums
is less than a pre-specified value
1
0.5
2
1
0.333 0.25
Row sums
12.75
22.3332
4.8333
39.9165
3
4
1.0
3.0
1.75 8.0
5.3332 3.0 14.0
1.1666 0.6667 3.0
squared
Normalized
Row sums
0.3194
0.5595
0.1211
1.0
Difference is:
0.3194
0.5595 0.1211
0.3196
0.5584
0.1220
0.3196
0.5584
0.1220
- 0.0002
0.0011
- 0.0009
Preference
Style
.3196
Reliability
.5584
Fuel Economy .1220
Selecting
a New Car
1.0
Style
.3196
Reliability
.5584
Fuel Economy
.1220
Ranking alternatives
Style
Civic
Civic
1/1
Saturn
1/4
Saturn
Escort
Miata
4/1
1/4
6/1
1/1
1/4
4/1
Reliability Civic
Civic
1/1
Saturn
Escort
Miata
1/2
1/5
1/1
Saturn
2/1
1/1
1/3
1/2
Escort
4/1
4/1
1/1
5/1
Miata
1/6
1/4
1/5
1/1
Escort Miata
5/1
1/1
3/1
1/1
4/1
2/1
1/4
1/1
Eigenvector
.1160
.2470
.0600
.5770
.3790
.2900
.0740
.2570
Miles/gallon
Fuel Economy
(quantitative
information)
Normalized
Civic
34
.3010
Saturn
Escort
Miata
27
24
28
113
.2390
.2120
.2480
1.0
Selecting
a New Car
1.0
Style
.3196
- Civic .1160
- Saturn .2470
- Escort .0600
- Miata .5770
Reliability
.5584
- Civic
- Saturn
- Escort
- Miata
.3790
.2900
.0740
.2570
Fuel Economy
.1220
- Civic
- Saturn
- Escort
- Miata
.3010
.2390
.2120
.2480
Ranking of alternatives
Style Reliability
Fuel
Economy
Civic
.1160
.3790 .3010
Saturn
Escort
Miata
.2470
.0600
.2900 .2390
.0740 .2120
.5770
.2570 .2480
.3196
*
.5584
.1220
.3060
.2720
.0940
.3280
Handling Costs
Dangers of including Cost as another criterion
political, emotional responses?
Separate Benefits and Costs hierarchical trees
Costs vs. Benefits evaluation
Alternative with best benefits/costs ratio
MIATA
.9840
CIVIC
1.3771
SATURN
.9791
ESCORT
Normalized
Cost
Cost/Benefits
Ratio
$18K
.333
$12K
.222
$15K
.2778
$9K
.1667
.5639
Complex decisions
Many levels of criteria and sub-criteria
10
Application areas
strategic planning
resource allocation
source selection, program selection
business policy
etc., etc., etc..
Group AHP
Teori AHP 2
11
12
Goal
General Criteria
Secondary
Criteria
Alternatives
Handling
Braking Dist
Turning Radius
Ford Taurus
Economy
Purchase Cost
Lexus
Maint Cost
Power
Gas Mileage
Time 0-60
Saab 9000
13
1 -9 Scale
Intensity of Importance
Definition
Equal Importance
Moderate Importance
Strong Importance
Extreme Importance
2, 4, 6, 8
Reciprocals of above
Rationals
14
Maintenance Cost
Gas Mileage
Purchase Cost
to
Maintenance Cost
Purchase Cost
to
Gas Mileage
Maintenance Cost
to
Gas Mileage
15
1/3
1/5
1/3
Consistency
Ratings should be consistent in two ways:
(1) Ratings should be transitive
That means that
If A is better than B
and B is better than C
then A must be better than C
(2) Ratings should be numerically consistent
In car example we made 1 more
comparison than
we needed
3M = 5G
M = (5/3)G
We know that P = 3M and P = 5G
16
1/3
5/3
1/5
3/5
has Rank = 1
That means that
all rows are multiples
of each other
wP
15
23
0.65 , w M
5
23
0.22 , w G
3
23
0.13
17
Properties:
- The number of nonzero Eigenvalues for a matrix is
equal to its rank (a consistent matrix has rank 1)
- The sum of the Eigenvalues equals the sum of the
diagonal elements of the matrix (all 1s for
consistent matrix)
Therefore: An nx n consistent matrix has one
Eigenvalue with value n
Knowing this will provide a basis of determining
consistency
Inconsistent matrices typically have more than 1 eigen value
- We will use the largest, max , for the computation
1/3
1/5
1/3
=A
w = vector of weights
Must solve: Aw = w by solving det(I A) = 0
We get:
max = 3.039
find the Eigen vector for 3.039 and normalize
18
Measuring Consistency
Recall that for consistent 3x3 comparison
matrix, = 3 max
Compare with
from inconsistent matrix
max
Use test statistic:
C.I.
n
Consistency Index
n 1
Measuring Consistency
For Car Example:
C.I. = 0.0195
n=3
R.I. = 0.52 (from table)
So, C.R. = C.I./R.I. = 0.0195/0.52 = 0.037
Rule of Thumb: C.R. 0.1 indicates sufficient
consistency
Care must be taken in analyzing consistency
Show decision maker the weights and ask for
feedback
19
Procedure:
(1) Normalize each column
(2) Compute geometric mean of each row
Limitation: lacks measure of consistency
1/3
1/5
1/3
3
1
Normalized
1/3
.65
.69
.56
.22
.23
.13
.08
.33
.11
wp = [(.65)(.69)(.56)]
wM = [(.22)(.23)(.33)] 1/3
= 0.63
1/3
= 0.05
wG
= [(.13)(.08)(.11)]
= 0.26
0.67
Normalized
0.28
0.05
20
Terima Kasih
21