W2_Afra_Best Alternative using AHP
1.
Problem Recognition
Manager or decision maker need to use an
appropriate method to make a complex decision. One of those methods is AHP (Analytic Hierarchy Process). Rather than prescribing a "correct" decision, the
AHP helps decision makers find one that best suits their goal.
AHP tool will be used
on this blog to choose the best and correct Gas turbine depends on
some
specifications and criteria.
2. Feasible alternatives
They are four Gas turbines with different specification has been
selected to choose between by using five criteria (most important and common)
which are: manufacturer, output, heat rate, price, Efficiency and Shaft speed. The
price of each gas turbine shows in the table below:
(A, B, C, D are dummy names for gas turbine)
A
|
332 Million dollars
|
B
|
234 Million dollars
|
C
|
222 Million dollars
|
D
|
123 Million dollars
|
Table 1: Gas Turbines Price
3. Development of the outcome of Alternative
Decision hierarchical tree below places all criteria and alternative gas
turbines. This decision hierarchical has three levels: the top level shows the
overall decision which is to select the best gas turbine for a plant, the second
level shows all criteria need to be considered while selecting the best
turbine, the third level shows all four alternatives.
Figure 1: Decision hierarchical tree
4. Selection
Criteria
Pair-wise comparison will be used to select the best alternative, this
technique the relative
priority for each criterion against each other criterion
as well as the priority for each alternative
(gas turbine) against each other
alternative. The Pair-wise comparison uses a scale that ranges
from equally preferred
to extremely preferred. Here is the scale
Figure 2: Pair-wise
comparison scale
5. Analysis and
comparison of the Alternative
The table below develops a weight for every single criterion by developing single Pair-wise comparison matrix for the criteria. all elements on the diagonal assign to 1 since When we compare any alternative against itself the
judgment must be that they are equally preferred.
Manufacturer
|
Output
|
Heat Rate
|
Shaft speed
|
Efficiency
|
|
Manufacturer
|
1
|
5
|
2
|
7
|
4
|
Output
|
1/5
|
1
|
2
|
4
|
2
|
Heat Rate
|
1/2
|
1/2
|
1
|
2
|
3
|
Shaft speed
|
1/7
|
1/4
|
1/2
|
1
|
3
|
Efficiency
|
1/4
|
1/2
|
1/3
|
1/3
|
1
|
Total
|
2.09
|
7.25
|
5.83
|
14.33
|
13.00
|
Table 2: Pair-wise comparison matrix
The table below shows the normalized
pairwise comparison matrix. Along with
the average for each row.
Manufacturer
|
Output
|
Heat Rate
|
Shaft speed
|
Efficiency
|
Total
|
Average
|
|
Manufacturer
|
0.48
|
0.69
|
0.34
|
0.49
|
0.31
|
2.31
|
0.46
|
Output
|
0.10
|
0.14
|
0.34
|
0.28
|
0.15
|
1.01
|
0.20
|
Heat Rate
|
0.24
|
0.07
|
0.17
|
0.14
|
0.23
|
0.85
|
0.17
|
Shaft speed
|
0.07
|
0.03
|
0.09
|
0.07
|
0.23
|
0.49
|
0.10
|
Efficiency
|
0.12
|
0.07
|
0.06
|
0.02
|
0.08
|
0.35
|
0.07
|
Total
|
1
|
1
|
1
|
1
|
1
|
Table 3: Normalized pairwise comparison matrix
Furthermore, to calculate consistency index, following formula will be used.
CI = 0.0180. To identify the random index, the table below will be
used.
RI value is equal to 1.12
According to above formula, CR is equal to 0.0161. Based on Saaty’s empirical suggestion that if C.R. <
0.10 is acceptable.
Table below is the alternative ranking matrix
Manufacturer
|
Output
|
Heat Rate
|
Shaft speed
|
Efficiency
|
|
A
|
0.0882
|
0.2769
|
0.3728
|
0.1824
|
0.2459
|
B
|
0.3126
|
0.2403
|
0.3728
|
0.4759
|
0.2534
|
C
|
0.5154
|
0.1922
|
0.2052
|
0.2910
|
0.2624
|
D
|
0.0837
|
0.2906
|
0.0491
|
0.0507
|
0.2383
|
Table 4: Alternative ranking matrix
The alternative ranking is the product between the
alternative ranking matrix and criteria ranking.
A
|
0.19
|
B
|
0.32
|
C
|
0.36
|
D
|
0.13
|
Table 5: Alternative ranking
The alternative that having highest weights will be
selected. Gas Turbine C is the highest ranked Turbine.
6. Selection of the
preferred Alternative
The mentioned criteria above are not enough to judge. After alternative
ranking has been defined, we should compare with the price using benefit-cost
ratio.
Cost
|
Normalized Cost
|
Alternative ranking
|
Benefit-Cost Ratio
|
|
A
|
332 .000
|
0.36
|
0.19
|
0.53
|
B
|
234 .000
|
0.26
|
0.32
|
1.23
|
C
|
222 .000
|
0.24
|
0.36
|
1.5
|
D
|
123 .000
|
0.14
|
0.13
|
0.93
|
Total
|
911.000
|
Table 6: Benefit-Cost Ratio
Based on benefit-cost analysis, Gas Turbine C has the highest ratio.
6. Performance Monitoring
and the Post Evaluation of result
To find more
accurate result, this tool needs to be done by someone who understands the
problem and criteria very well. To make sure about the result I'll do another
blog posting in coming weeks with the same problem but by using non-compensatory
models.
References:
1.
Suresh. (2017). Ahp calculations. Slideshare.net.
Retrieved 13 November 2017, from https://www.slideshare.net/lakshanasuresh/ahp-calculationshttps://en.wikipedia.org/wiki/Analytic_hierarchy_process_%E2%80%93_car_example
2. W5_WW_ Analytical
Hierarchy Process. (2017). GARUDA AACE 2015. Retrieved 13 November
2017, from https://garudaaace2015.wordpress.com/2015/03/28/w5_ww_-analytical-hierarchy-process-2/
3.
How to make a decisions:
The Analytic Hirerrchey Process. (1990) (pp. 15-19). North Holland. Retrieved
from https://www.google.com.om/url?sa=t&rct=j&q=&esrc=s&source=web&cd=5&cad=rja&uact=8&ved=0ahUKEwiR9PH-
_bvXAhXQIOwKHfveC8YQFgg_MAQ&url=https:%2F%2Fwww.researchgate.net%2Ffile
.PostFileLoader.html%3Fid%3D5879e59e615e27bbf27ef4e3%26assetKey%3DAS%253
A450351808684035%25401484383646513&usg=AOvVaw0VtE6wBMM3Umm_zqlcjAw5
_bvXAhXQIOwKHfveC8YQFgg_MAQ&url=https:%2F%2Fwww.researchgate.net%2Ffile
.PostFileLoader.html%3Fid%3D5879e59e615e27bbf27ef4e3%26assetKey%3DAS%253
A450351808684035%25401484383646513&usg=AOvVaw0VtE6wBMM3Umm_zqlcjAw5
4. Ishizaka, A. (2017). Clusters and
pivots for evaluating a large number of alternatives in AHP. Retrieved 13
November 2017, from http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0101-74382012000100006
WOW!!!! AWESOME posting Afra......
ReplyDeleteImpressed to see you start out with such an advanced tool/technique. What you might be interested in reading is this paper by Ms Lita Liana from Shell Indonesia. http://pmworldjournal.net/wp-content/uploads/2014/01/pmwj19-feb2014-liana-analytical-hierarchy-for-roi-oilandgas-projects-indonesia-FeaturedPaper.pdf
She used AHP to calculate the Minimum Atractive Rate of Return (MARR) on oil and gas projects based on their RISK PROFILE. I have a suspicion that what she did is something that might be of value to OPWP?
Take a look and see what you think...
BR,
Dr. PDG, Jakarta
Thank you Dr. PDG for your impressive comment.
ReplyDeleteHonestly I already have experience with AHP, because of that I started with it : )
Regarding the topic, sure, I will go and read about the it ! I will be happy if I could find out something that might be of value to OPWP.
Afra.