I have a question, I wonder if anybody can help me with it here.
basically I want to create random points, but I have difficulties with the conditions I want it to be created.
1. I have a field for the number of points in each feature (COUNT)
2. I have a vector/raster with a value for weighting points distribution. (Weight)
I know that I can use "Create Random Points" and do the distribution based on a field, but as far as I know I have no control over their distribution. I wonder if I can do this NUMBER of random points created based on the field that I have and their DISTRIBUTION based on the value of the vector/raster that i have as their weight for distribution.
I hope somebody can help me with this
What type of method are you proposing to do this? There are few/no tools available that don't produce point distributions based upon a specific pattern (ie random, circular, grid, linear etc)
the image actually shows the polygons with the values I want the points to be distributed based on (basically the more distance from service area the less probability of point density in the area)
In this image you can see the random points created based on the field (but completely random), i want their distribution be based on the value of the polygons. I used create random points for this part.
I hope this is clarifying my question better.
Thank you again
Take a look at the Sampling Design Tool addin for ArcGIS 10. You can create random point and select how many points will be allocated to each attribute in a specified field. This can be done proportionally (by area), equally, or manually.
Thank you Ken
Actually I saw this earlier today but didn't get to play with it to see if it actually can do what I want or not. I will give it a go in an hour.
Only one question, do you get to introduce number of samples manually or it can read it from attributes? I am trying with a small region but the actual model will have big number of different points to be created.
Thank you Ken
It seems to me that this can be the right tool for me to use, but it was throwing an error when I was trying to use it. At the time I have some problem with my network analysis, after that I will come to see if I can actually fix the issue.
It seems to me that the two stage random can be the solution in my case, but I am not sure. I should have an answer by tonight.
Please correct me if I am wrong.
I assume I can use a field/raster value as the weight for distribution of sample points and another field as the count of the actual points.