In Analytic Hierarchy Process (AHP), after an initial stage of data collection, the next step is related to GIS analysis, and focuses on converting collected vector maps to raster and spatial analysis functions (e.g., map overlay using Boolean functions, distance analysis and reclassification).
The following phase includes calculation of criteria weights using a developed tool adopting AHP principle. Here, the Consistency Ratio is computed in order to verify the consistency of the experts’ judgment.
Finally, the last step covers the integration between criteria weights and maps, producing a map representing land suitability, risk/hazard or best location.
My question focuses in the reclassification of the factors (criteria) maps that are used to produce the final map. In different papers that I read the number of classes varies from 2 to 5. In same cases authors use 9 classes (the same number of the Pairwise Rating Scale proposed by Saaty - ?).
Thanks Hamid,
I agree with you, but I also think that the definition of too many classes turns the classification subjective. In my opinion, more than five classes is not so reasonable.
But the reason I address this question relates to the fact that in some published papers the authors defined the same 9 classes of Pairwise Rating Scale proposed by Saaty, with the same odd scores [?] (e.g., Maris et al., 2008). Possibly a misinterpretation of the method? One thing is the Pairwise Rating Scale, other thing completely different is the map classification.
Regards
Maps represent spatial phenomena tighthat some science are especialized in analysing. A good weight for spatial economic analysis is land value. Eventually for life phenomena water density is a good solution. Tomaz
Dear Luis;
I think I guess your question and its answer but I have to know the papers that you talk about to give you my clear answer. Please let me know which studies you are talking to. I have to look at them and read them to answers your question according to the study.
Have a nice day
All my best
Burak
Dear Burak,
I already sent you the paper I mentioned as an example.
Regards
Luís
Dear Luis;
I hope I can explain in a short and clear way.
If you would like to ask me questions during your research study, I will be happy to give my clear replies and to help you.
If you would like to write and submit a paper, I can also contribute as a coauthor of your paper (second, third, fourth etc.).
I hope your research will be very good and informative for the next generation researchers.
Have a nice day All my best
Burak
Dear Burak,
Thanks. The same applies to you. If you need to share any question, I will be pleased to answer if I could.
The possibility of writing and submit a paper together is something that we could consider.
Best Regards
Luís
Dear Luis;
Okey. When you ask me to be your coauthor please send me your e-mail, I will be your coauthor in the subjects that I can give support and contribute. You already know me e-mail adress, I think. If you do not, please have a look at my papers and presentations. My e-mail adress is there. Please feel free to contact to me anytime you wwant to.
Have a nice day
All my best
Burak
Dear Luis I do not have enough experience on spatial analysis but I do in AHP. The reasons that supports Saaty's scale are borrowed from psychological literature. One of them is The Magic Number Seven (attached) that is the famous nine levels but when you research on original Saaty's sources you grasp that really there are only five levels (1=equal, 3=somewhat more important, 5=much more important, 7=very much more important, and 9=absolutely more important)
The even numbers are original used to solve controversies in group judgement when consenus fail to be reached. In any case AHP is a methodological tool and in this vein it is never overcomes the expert knowledge of the application context.
Please apologize me I forgot to attach the paper, but you have it know
- Badges
- Science topic
- Similar topics
- Geoscience
- Asia
- India