Using Maximize Capacitated Coverage Effectively

5188
11
08-13-2015 07:26 AM
JacquelynPutnam
New Contributor

Hey everyone! I am currently trying to use the maximize capacitated coverage problem solver of location-allocation to allot a study area population to emergency shelters. I have a layer of approximately 250 shelters, all of which have a capacity as large as 800 and as small as 10. My study area is quite large; at present I have 1268 census block group cetroids. Obviously each demand point contains a multitude of census data such as raw population. I would like to allocate population to the shelter that is closest by drive time, but am having trouble because the demand points cannot be apportioned to the facility, it's all or nothing. My end result is very unrealistic because most shelters reach all the way out into rural areas where the demand point's population is small, bypassing literally thousands of others who would reach the shelter first.

How can I rectify this? From the other posts in the forum it would appear I may have to break up the demand points into smaller populations. If so, how do I want to go about this? Create Random Points where the population contained in the census block group is the number of points created inside the block group polygon?

Thanks in advance for any insight you may have. I'm at all total loss!

0 Kudos
11 Replies
JaySandhu
Esri Regular Contributor

You are right that it is "all or nothing" allocation for each demand point. Besides the approach you mentioned about creating random points inside the census polygon, you could add the same demand point multiple times with less demand. Let's say you have a demand location with 100 units of demand. You could add that same location ten times with a a demand of 10. These points will be sitting on top of each other but they may be allocated to different facilities as needed.

Jay Sandhu

0 Kudos
FaroukUmar
New Contributor
Please, is there any way to create these multiple points of lower demand in Arc GIS? What is the best and quickest way to do it?
0 Kudos
JaySandhu
Esri Regular Contributor

You can simply load the points to the Location-Allocation layer multiple times! Make sure to load them with lower demand. For example if you load them five times, then make sure the demand is 5 times less.

Or you can start editing the point feature class, copy the points and paste them again as many times as you want.

Jay Sandhu

0 Kudos
FaroukUmar
New Contributor

Thank you very much. I have made the multiple demand points (29001). However, the maximise capacitated coverage analysis takes about 30 minutes to run and sometimes doesn't run. The 29001 demand points are loaded within 3minutes without any problem. The network dataset has a spatial index. Could this be due to the number of demand points I have? If I use fewer demand points it works perfectly. What is the solution, please?

0 Kudos
JaySandhu
Esri Regular Contributor

When you say that it does not run sometimes, do you get an error message or it hangs? What error do you get?

You can try to add a distance/time cutoff to cut down on the amount of searching. For example if your demand is spread out over 10 miles and you are allocating 10 facilities, then perhaps a cutoff of 3 or 4 miles should be enough. It basically means a facility will not search beyond that distance for available demand.

Jay Sandhu

0 Kudos
FaroukUmar
New Contributor

I am using  5000 and sometimes 10,000 metres as the distance cutoff. The problem is when I click the solve button, and it takes about 30minutes to locate the facilities. In some cases, it stays at  0% without progressing.

0 Kudos
JaySandhu
Esri Regular Contributor

The amount of time needed to solve location-allocation will depend on the number of demand points, number of candidates and how many facilities you are choosing to locate. So it can definitely take a long time if the problem size is big. How many demand and candidates do you have and how many are choosing?

Jay

0 Kudos
FaroukUmar
New Contributor

I have 838 required facilities and 350 candidate facilities. I am choosing 1000 facilities. I also have 29000 demands points after breaking the original ones to points of lower value.

0 Kudos
JaySandhu
Esri Regular Contributor

So 1188 facilities choose 1000 (even when 838 are fixed) are a lot of combinations! And solving for capacity does add another layer of complexity. So it can take some time to find a solution. I cannot estimate the time needed (based on what hardware/memory/etc that you are using), but it could be a while!

Jay Sandhu