Considering the scenario of GIS usage for multi-criteria decision analysis to evaluate the land suitability, do you think is it appropriate to assign weightage to the spatial layers based on the published literature?
I think it would depend on the problem you are analyzing, which you don't mention in your question. In a general basis, I would suggest to use an Analytical Hierarchy Process in order to assign weights to the different layers, comparing them by pairs, according to the works by Saaty. You can find a good description of this process in GIS in this paper by Eastman.
www.geos.ed.ac.uk/~gisteac/gis_book_abridged/files/ch35.pdf
Hi Khan,
Yes, he said the right thing. If you are not aware of the exact weighing factors, can use AHP.
It depends on the issue you work. if you have contacts with the experts in the field, who is familiar with the study area, you can get the weighting values for each factors you used in the analysis.
Good luck.
vijith
@Garcia and Vijith: I am trying to develop a land suitability map for the utility scale solar power plants for non-urban area of region of interest. The parameters to be considered are: slope, insolation, distance to roads etc. I will surely look into the material you suggested.
I was wondering that the topography of any location does vary (and consequently the incoming solar radiation and climatic conditions) so assigning weightage to spatial layers based on published literature will be appropriate.
You can use Fuzzy overlay which is not require any weightage assignments. However, you just convert your layers to membership functions and add them together.
I agree with Hamid Reza, you can use the AHP method, developed by Saaty
Hi,
Pairwise comparison based on AHP given by Saaty would be the best method to find weightage values.
I would agree with the other answers. However, I would add that you would first need to standardize your input layers somewhat. Could you define which slope, insolation, distance to roads etc. is of high and which of low value for solar power plants? You then weight those layers. Whether you do that expert or literature based doesn't really matter as long as you have good reasons to do so and the results are valid.
I think Rank sum Method is the easiest way for giving weight to factors.
Following what Beni suggested, I would like to know, what kind of expert opinions are considered valid and sufficient for a PhD level research? I would like to use the same method for my PhD. Thank you.
Dear Harry Radzuan. The question of what you consider to be an expert is not an easy one. I tended to use the Definition of Meuser/Nagel (1991) which – if I remember correctly – state that an expert shapes the field of research and has decision power. But in the end an expert is what you, as a researcher, consider to be an expert. So as a PhD-student, I think it largely depends on your supervisor, whether he considers the people 'expert enough' :)
BTW: You might find this publication helpful. It is accepted and should now enter typesetting soon.
http://josis.org/index.php/josis/article/viewFile/368/199
Hello Beni, thank you so much for the explanation. I will consult with my supervisors and see what they think about this. :)
And thanks for the paper too!
- Badges
- Science topic