I am working with ArcGIS 10. I am working on a study that is about providing heat generation plants with forest harvest residues. I used the closest facility module in order to find the closest facility to the forest stands from which I will collect forest residues. I would like to take into account the capacity of my heat generation plants as well. That means that I cannot bring to them more forest residues that they can handle. I would like that the amount of forest residues that are left over (due to capacity limitations of the heat generation plants) are transported to a terminal. Is it possible to handle such a situation in Network Analyst and could you give me some hint of how? Best rgrds Dimitris
There is no out of the box way to sum up capacity but it should be very easy to do. Using Closest Facility, when you load locations to add in the Incidents (forest stands), you can map over the amount of residue at those locations. To do this, create the CF layer, then open the Attribute table for Incidents and add a field called Residue of type double. Now in the NA Window, right-click the Incidents and Load Locations. This brings up the form to choose your forest stands to load from. In the Location Analysis Properties, you will see "Residue" listed. Click on the Field and drop down and pick the appropriate field containing the residues at that location. Load the locations. Now you will have the residue field loaded on the Incidents. Solve the CF layer and now you can Sum up the residue being brought into that facility.
Location-Allocation does not yet support capacity. But you can solve for 4 sites. I am not sure why you are getting only two? Perhaps you choose minimize facilities and not minimize impedance. Spread out the candidates where the school can be built then choose a minimize impedance and put an impedance cutoff to a reasonable distance for how far the school will take students from. Then see how the students are allocated to each school. Again, this will not give you the perfect solution as it is not guranteed to take only 400 students but should generate your alternatives to where to best put them based on other constraints. You can choose a set of 4 reasonable locations and then run them through the Optimum Allocation tools: http://resources.arcgis.com/gallery/file/Geoprocessing-Model-and-Script-Tool-Gallery/details?entryID... Jay Sandhu
Hi Jay, thanks for your fast response. I simplified my problem a little bit - maybe too much.
So, here is the initial situation: I have a rural area with 600000 people in about 1000 villages and about 4 small cities. The road network has about 3000 km. My subject is to expand the existing charging station infrastructure for electric vehicles (17 sites). So I calculated from the population and tourist stays of each demand point the expected demand for charging per day (this is still a long-shot guess). As layer for potential facilities I take either the same 1000 villages or 15000 neighborhoods. For these I want to calculate different expansion stages (50, 100, 200 charging stations). Therefore I need a kind of capacity limitation, because otherwise one station supplies the demand of 100000 persons and another one only for 50.
Th description of the python script you suggested sounds good, but even when I reduce the amount of demand points and facilities it cancelling the calculation because of memory shortage. I am not sure if the tool uses a heuristical approach like the location-allocation tools in the NA.
Last week was Esri User Conf so missed seeing your question. Since you are siting electric charging facility, rather than putting a capacity constraint, put a distance or travel time constraint. So choose the "Minimize Facilities" problem type and set a travel time constraint of 60 minutes and solve. Adjust as needed. Hint: Take into account how far the vehicles can travel on one charge. Once you know what is the minimum set of facilities needed you can switch to maximize coverage and pick your expansion strategy of 50, 100, etc until you serve everyone. Jay Sandhu