Targomo Coverage Score

The Targomo Coverage Score represents an estimate of the total unique area covered by the sources in the request. The Source Coverage Restriction allows users to use a large number of sources in areas of interest while still protecting Targomo services from being overloaded. To give an example:

Let us assume you have 5 sources all with a travel type of “bike”, with travel calculated using time, a max travel time of 1 hour, and no other default settings changed. For the sake of this example we will also assume the coverage maximum is around 3000km^2. In our default settings, bikes have a travel speed of 15 kmph, so this gives us a 15km Radius which approximates to a 707 km^2 area circle. If The circles are far enough apart that there is no overlap then your coverage would be 707*5 or 3535. So we have exceeded our source coverage limit, and now we want to reduce that. To do so we have 3 options.

Remove sources

The simplest option is to remove sources, this will reduce large sections of coverage thus reducing the total score. However, depending on your setup this may not be as useful as it seems. I will explain further on.

Reduce the travel time or distance or reduce the travel speed

Reducing the maximum travel time or distance will reduce the area covered for all sources. The same effect can be achieved by shifting down to a slower travel type like “walk” or by changing the default speed, again reducing coverage. Again simple and easy but possibly not applicable if your use case requires you to know information about what is at specific distances from your sources.

Move your sources closer together and/or focus on closely clustered sources

This is where you can achieve the most with this restriction. Since our focus is on the total unique area covered, if source coverage areas overlap with each other, the shared area is only counted once and thus the contribution to total coverage is much lower. In our above example if we assume that 3 of the sources are nearby each other with only 30% of their circles each not overlapping with each other, the previous coverage of 3535 is reduce to about 2544 which takes our total down by almost 1000. This means that even very large numbers of sources, when clustered closely together, have a much lower effect on coverage.

To maximise speed and efficiency, we use h3 hexagons and count the number of unique hexagons to approximate the area covered. The Coverage Score is calculated from this area and checked against the Coverage Restriction. In the event of a failure a coverage score will be returned in the response e.g.

Source Coverage limit exceeded. Coverage Score at 662, Max 500. Coverage can be reduced by reducing 
`maxEdgeWeight`,reducing travel speed or by placing sources closer together.