Stats about Kuwait’s Public Bus Network

Posted on Posted in Uncategorized

I have lived in Kuwait nearly 4 years. Before starting this project, I had caught the bus about 3 times, twice as a last resort. One was during a failed earlier attempt at this project when I was running the Kuwait Mapping Meetup. I mapped the stops of the 999 route with a friend. We stopped due to lack of committed team members – and that was mainly because none of us really caught the bus. So, why are we trying this again? Simple – if it were pretty much any other country, one of the first things I would’ve done as soon as I arrived was learn the bus routes to work and other useful places in town. I would never have expected to spend 4 years and catch the bus less than a handful of times. To learn, I would look at a map and a schedule and herein lies the reason why I haven’t caught the bus. These two things are basically not available in Kuwait. As discussed in the last post, the KPTC route map is not compelling. You cannot zoom in, the routes are all the same colour and the option to display one route at a time means you cannot easily see where to change routes. We have solved this problem in our route map using their data. Speaking of which, let’s look at some stats.

  1. Routes: 37
  2. Total length: 1433km
  3. Average length: 39km

What about the catchment? What is a catchment? This is like a valley with a river in the middle. Up to the top of the mountain ridges either side of the valley is the catchment for the water feeding the river. The same idea can be applied to bus routes, they are the rivers and people are the water. What then is the mountain ridge in this context? We chose walking a time of 10 minutes. Just like water won’t flow up and over a ridge then down into the next valley, we assumed a human – especially during Kuwait’s 3 month summer of 50° centigrade temps – will not walk more than 10 minutes to catch a bus. Assuming the average human in Kuwait walks 4km an hour, this is 4000/6=666m every 10 minutes. A very simple way to calculate the catchment area of the network, within the ‘mountain ridges’ defined by a 10 minute walking distance, was to measure a 666m buffer from all 37 lines comprising the routes. This is an area of 57,965 hectares (not double counting where it overlaps with nearby routes).

The total inhabited area of Kuwait is about 80,503ha and 47% (38,034ha) of this is covered by the catchment. The reason why there is not a direct relationship between catchment area and populated area covered by the catchment is there are parts of the catchment outside the populated area. For example, as shown on the map, most of the 500 route going north to Iraq is on the 6th Ring Road and Highway 80 and not many people live in those parts of Kuwait.

There are a couple of other considerations to do with the catchment estimate. First, it needs to end at the coast because it’s unlikely people would walk directly from the water to the bus! Second, the simple method of measuring 666m from everywhere along the route overestimates the catchment because, for example, the bus doesn’t stop on bridges and it’s only supposed to stop at … bus stops!

This is a nice segue into one of the three data streams (an idea covered in the next blog post) of this project. In order to build out a compelling database of Kuwait’s bus service, we had to collect, in ascending order of difficulty, the bus routes, bus stops and schedules. We have the routes, are partway through collecting the stop locations and have barely started on the schedules. With the stops we are able to do some important things, such as make a more accurate catchment estimate and use them in the routing service of a public transport app. Regarding catchment, the below image shows a 10 minute walk from each of the stops we’ve so far collected along the 999, 105, 106, 66, 602, 24 (partial) routes: 14,943ha. The catchment calculated using the routes for the same location 18,185ha. This illustrates that a route based catchment is overstimated at a rate of 18%, given that the stop catchment is 82% of the route catchment.

In the above image the stop based catchment is shown in beige, route based catchment in mauve. One can see significant portions of the route based catchment are not serviced by stops.

Leave a Reply

Your email address will not be published. Required fields are marked *

To create code blocks or other preformatted text, indent by four spaces:

    This will be displayed in a monospaced font. The first four 
    spaces will be stripped off, but all other whitespace
    will be preserved.
    
    Markdown is turned off in code blocks:
     [This is not a link](http://example.com)

To create not a block, but an inline code span, use backticks:

Here is some inline `code`.

For more help see http://daringfireball.net/projects/markdown/syntax