Calculating loss of value from a wind turbine on surrounding properties depending on distance?

Question

I have to evaluate where installing a turbine would be most efficient, depending on wind speed and price of properties surrounding each cell in a raster. The cost on value of properties is computed using some estimation based on the distance of the turbine to nearby properties.

I know the price of each property in the area. I have to find a way to calculate how much each turbine would cost, in terms of loss value, on the surrounding properties.

I thought of doing this by making buffers for each given distance, joining the resulting layer with the layer containing the price of properties, and then making a buffer reducing the buffer containing the value of properties to the cell (therefore, in the same, negative amount of the first buffer). However, the value of properties does not appear in the last buffer layer:

(example) Cell => 600 m buffer => join buffer with layer containing value of properties => buffer - 600 m => cell

Is there a way to keep information from the original cells on the buffers created?

Is there an alternative way to calculate the loss of value caused by a cell on the surrounding cells?

Answer 1

I think you're trying to convert over to vector too quickly. Distance can be modeled quite easily with rasters, allowing you apply your cost model across the continuous surface. Now, the premise of a fixed loss/distance ratio is probably deeply flawed, but if you use the centroid of each polygon to determine the spot value on the surface, you'll have a specific cost per feature.

Answer 2

I suggest that you do everything in raster.

1) convert your polygon value to a raster

2) use focal statistics with an home made filter (representing your distance rule) http://help.arcgis.com/en/arcgisdesktop/10.0/help../index.html#/Focal_Statistics/009z000000qs000000/

your home made filter must include the weight of each pixel in the surrounding of each cell.