I am trying to do computations on several vector layers to create a single new vector layer. the summation function for raster cell statistics would be ideal. The ideal output would be a grid vector layer, similar to a fishnet, where the output in each cell is the summation of all of the vector inputs cells. I have several layers that each represent a different valuation scale, Temporal collection, order value, customer remaining orders etc.
I have looked at merging the different layers together but there are too many disparate attribute fields as well as varying "cell" sizes. is there a toolset that has similar functionality for vector layers?
I thought Union will bring the individual shapes together and then creates individual features for each shape but doesn't run any statistics or calculations on the shapes. I have several hundred thousands individual features spread across several shapefiles and feature classes that I need the computations run on which is why I was thinking about cell statistics. I can set it up and let it run over night while I'm gone. I am hesitant to convert to a raster format, run the tool and then convert back to polygons because of the possibility of the spatial accuracy loss after so many file shifts.
Yes, I would avoid going the raster route, for the reason you described. I do this kind of overlay and summation all the time to prepare input data for a forest management model. I generally use a combination of the overlay analysis tools, and the Summary Statistics tool where I summarize values by Harvest Units, for instance. It just depends what you want your end result to be exactly. I wish I could be more specific, but it's difficult without more details. If it's a process you want to repeat, and it has many steps, you can put it together in modelbuilder for efficiency.