Crowd Mapping – making sense of the chaos

Public transport in most developing countries is an ever-changing mix of independent operators all following their own rules, routes, schedules and rate hikes. The result is an erratic and often unreliable system. A recent study conducted in the city of Kampala, Uganda, shows that smartphone technology could offer a brilliant solution and bring some structure (and huge benefits) to this chaotic environment

For myriad reasons, the infrastructure in developing cities, especially in Africa, stagnated while globalisation and the demand for public transport exploded. Most governments were too slow (or too poor) to react and private transport is too costly – only nine percent of commuters in Kampala make use of private transport.

This created a prime opportunity for clever entrepreneurs to pounce on the gap in the market and, as a result, privately owned minibus taxis (also known as matatus or trotos) transport 83 percent of all Kampala commuters on a daily basis.

The problem is that there seems to be no rhyme or reason to the schedules, stopping points, routes, or even pricing. All of this is seemingly done at the whim of the driver or owner of the taxi.

Look closer, however, and you’ll see some patterns emerge – all guided by commuter needs and existing or temporary infrastructure, such as unofficial taxi ranks. These underlying patterns are what prompted the clever people behind the Kamapala study to see if they could help the citizens of this dynamic city to officially map its organic public transport system.

The truth is, cities such as Kamapala desperately need some type of structure when it comes to public transport. Without a clear picture of how the city moves its citizens, commuters cannot effectively plan their trips to and from work, school, or any other destinations. This is inefficient and costs both commuters and operators a lot of time and money, which, in turn, has a negative effect on the economy as a whole.

As the population of Kampala grows and the city expands, the problem will only get worse. In 2012 the greater Kampala region had a population of 3,15 million people.

According to the World Bank, this will double to 6,3 million by 2020 and by 2040 Kampala will be a city of 12,6 million people. Efforts to organise the transport system therefore need to be made now, before the task simply becomes too overwhelming.

How smartphones make a difference:

In 2015, the Department of Electrical and Electronic Engineering at Stellenbosch University started a project to map Kampala’s public transport system using smartphone technology. The project was funded by GoMetro – a mobile App that supplies public transport information to commuters in cities around the world.

Students and volunteers, armed with the energy of youth, smartphones and the GoMetro Pro App (this version allows the GPS tracking and storage of route data) hit the streets of Kampala to start the first phase of the mapping process. This involved documenting the GPS coordinates of all official taxi ranks, as well as the routes that originate from each one.

They then moved on to record all the minor taxi ranks (that have developed organically over the years) as well as their connecting routes.

All of this information was collected on GoMetro’s central database. The team also noted the number of taxis departing from each taxi rank, as well as the frequency of departures. During this phase they also collected information on destinations by determining where each taxi was headed.

Phase two was all about tracing the routes leading from all the taxi ranks. In this case, students and volunteers boarded taxis and set out toward various destinations. The GoMetro App automatically stored their GPS locations every 30 seconds and the team members manually tagged all stops along the way.

This information was used to work out the exact route used by the taxis and to calculate the number of stops (formal and informal) on each journey.

In the third and final phase, the team made use of QGIS, a free Linux Geographical Information System, to compile all the data. All the GPS information collected was overlaid on a road map of Kampala. Using specialised software, the team was able to plot all the trends and figure out how to classify all the routes and stops.

In the end, the hard work paid off and the results were fantastic. By mapping the overall trends, the team was able to put together a schedule showing the times when the majority of taxis tend to depart all the mapped taxi ranks.

This allowed them to confidently set up an informal schedule. For example: If a commuter goes to taxi Rank A at 07:00, they are likely to find ten taxis ready to depart along Route B heading to destination C.

The results may not be as accurate as train schedules in Sweden, perhaps, but it is a huge step in the right direction for Kampala. Using the GoMetro App, Kampala residents and visitors can now effectively plan their time and journeys with reasonable accuracy.

Taxi operators, on the other hand, can run a more efficient business by ensuring they’re at certain stops at specific times instead of driving around looking for fares.

In the long run, this study and the use of the GoMetro App could lead to a more structured and effective public transport system for Kampala. Through crowdsourcing it is possible to keep refining the existing data and to add any new routes that spring up, or reclassify routes that grow from minor to major status.

Tireless volunteers and the magic of smartphone technology have given the city of Kamapala a major boost toward developing a more efficient and organised public transport system. All parties benefit from the process and it proves that developing cities don’t need to spend millions (that they don’t have) to create a new public transport system from scratch.

In the end, the project was a huge success. By optimising and organising the system that’s in place, Kamapala is showing the way forward for developing cities around the world. It also goes to show that students with smartphones aren’t always such a bad thing…