The missing parts of GTFS (I)

A not so perfect world

GTFS is an easy to use standard. There are many good things, like  the time the bus arrives and departs, the connections with other lines, frequencies, GPS positions, trip shapes on a map, everything. Really everything ? Alas, that is not the case. The origins of GTFS are, like any project, in the mind of a single individual, but a standard should not be the work of one person. And this is the weak link in the chain. Don’t understand me wrong. GTFS has a part called GTFS Extension that leaves the door open for modifications. But what modifications should be made ?

The missing standard committee.

A standard needs a committee to elaborate it. GTFS has not been a standard from the beginning. It has evolved from a pain point. It had helped people by improving the experience of using the public transportation, in the bus or on the subway. Great. But not perfect.  And by discussing with professionals from transportation companies, I realized that the situation is far more complex. This article will present such an issue.

The subway connection

No, it is not a movie title. On average, people use two transportation lines on their journey from home to work and vice-versa. In fact, this implies automatically three more lines.

five-linesThis misconception is due to the fact that we overlook the small details. When someone calculates the time required to get from home to work by using public transportation, it takes into consideration the time from A to F. This is the total time we need. But, when it comes to GTFS, it contains only the individual times from B to C, respective from D to E. It is true that time from A to B, or from E to F are hard to estimate. What is even more important, time from C to D is almost impossible to measure. At rush hour, it might go from 5 minutes to a full one hour, while late at night in may be a 3 minutes walk.  But GTFS has no mean to measure it, because walking is not public transportation. Walking is just walking.

From C to D, GTFS has the file transfers.txt. It is just one entry, one estimation. No mathematical model to take into account the density of population, the speed with which the crowd moves. And it cant the way it has been designed. After all, GTFS has been designed to improve a bit the experience of people, not to solve complex problems.

From A to B, there are average speeds that can be considered. But can I compare my speed withe the generally accepted average ? And from E toF, can I have an average speed, when, in the morning, many people rush in the same direction. It may have the same problem as the central segment C to D. What to do ? How can we solve this ?

The solution

Before starting, let’s say that there is no miracle solution. Not today. But, one thing is sure, there should be a way to measure the speed of CE, respective EF segments at various hours. Based on the measures, some  guy passionate of transportation,should be able to suggest or propose a mathematical model. It might not be perfect, but it is better than nothing.

Then, based on that model, some other guy, even more passionate about it, should be able to introduce a new type of vehicle called human walk. At a top speed of 6-7 km/h, this new vehicle might be the missing link. Or not. But it is worth a try.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s