Accra Trotro Network map

Details:

Map design for Trip Lab at concordia University.
2016 as a Designer


Trip Lab team
Zachary Patterson, Natalie Wiseman


Case Study written 18/07/27

Introduction

In the winter of 2016, I approached Zachary Patterson for a very special. Patterson is a research chair in Transportation and Land Use Linkages for Regional Sustainability. The project concerned the elaboration of a user friendly, schematized map of Accra’s trotro network. In this case study, I will discuss the context of the trotro network, as well as how I developed a concept for a user friendly public transit map.

Final map of the Accra trotro network Illustrator 2016
Bus route table Illustrator 2016

Existing infrastructure

In Accra, Ghana, there is a network of small buses called trotros. Together, they contribute to the transportation of the city’s population. However, this network has the particularity of being operated by drivers that have licences to operate on certain routes. Therefore, any driver can claim an unclaimed bus route and exploit it with a trotro. With a system like this, whether the route is in operation or not remains unknown because there are no global systems or timetables imposed upon the drivers. Therefore, the approach of the trotro network is much more organic, and makes for an interesting challenge when it comes to mapping the network.

Mapping the network

Patterson’s team had previously mapped these routes with a GPS and a stop log to know exactly how many, and which bus routes were active, as well as where they were stopping. With the data, they were able to produce a map of the network. The result of this exercise was a complex network of spaghetti lines that were geographically accurate, but hard to follow. The only tags present were at the beginning and end of the lines, with no clear indication of where bus routes connected, where they shared a street segment or terminus or simply where they stopped
The first step was to understand the data that was collected during this mapping operation. The line numbers go up to the 900s, but in total, there are less lines. The first observation was that there existed a densification of the lines towards the city center, as well as on the main highways and major intersections of the city. I wanted to know how the bus routes interacted with each other, and how commuters in Accra could benefit from a better understanding of these interactions.
With the help of a colleague, we developed a map of nodes that represented roughly the stops and terminuses for the bus routes. This map showed how the bus lines interacted together and formed a rough hexagonal shape in the city center, with lines that extended beyond in the sprawls of the city.
I developed a network of 21 nodes that joined in a hexagonal shape in the city center. I had this map printed in large format, and I started to write down the lines that were circulating and stopping in common areas. This method was not optimal. I had to manually combine data from the terminus node map, the geographically accurate bus routes map, as well as with geographical data about roads, neighborhoods and landmarks. In a better world, the process could have been done with programming. Along the way, 44 extra nodes were created to better represent unique terminuses that had different names. So there it was: a network of 65 nodes. I just had to make it look nice.

Major terminus nodes mapped QGIS 2016
Cleaned nodes projected on major terminus map Illustrator, QGIS 2016
Cleaned nodes Illustrator 2016
Aligning the cleaned nodes on a 60° grid Illustrator 2016
60° grid nodes Illustrator 2016
60° grid nodes projected on major terminus map Illustrator, QGIS 2016
60° grid nodes projected on major terminus map with line numbers. Illustrator, QGIS 2016

Schematizing the network

The goal of this project was to create a map that could inspire opportunities to develop and organize the trotro network. In a sense, it was never made to be useful or precise. It was made to be different. With this information in mind, I knew the best approach to this would be inspired by modernist schematized maps. Maps such as Harry Beck’s London Tube map of 1920, as well as its contemporary descendant. Massimo Vignelli’s MTA map for New York of 1972. But surely the most influential maps were Vignelli’s Washington Metro trial maps. This is where I got the idea to use a 60° grid. This grid could fit the hexagon of Accra’s Trotro network, as well as provide control over the extension of the network beyond the city center. 

Schematized hexagon map projected on major terminus nodes map Illustrator, QGIS 2016
Schematized hexagon map Illustrator 2016
Aligning hexagon map Illustrator 2016
Hexagon map Illustrator 2016

To help navigation, I split the network in six regions, roughly taken from the overall shape. North-East, East, South, South-West, West, North-West and center. In the city, these regions would also correspond to different areas along highways, landmarks or waterfronts. This segmentation into six pieces also helped understand how all these areas joined in the city center, or the hexagon. Doing this, I also found out that multiple routes were taking similar paths across the city, between Accra’s city center and Kotoka International Airport, along highways and main arteries. In trying to help navigation, I ended up discovering patterns in the network that were previously invisible to me. 

North East area Illustrator 2016
West area Illustrator 2016
South East area Illustrator 2016
East area Illustrator 2016
North West area Illustrator 2016
City center area Illustrator 2016

After schematizing the map, I ended up cleaning up my file, aligning and offsetting lines so they would not overlap and create false connections. I added route numbers onto the actual lines, so that we would understand which lines took what paths. I tried to add a tag at the beginning and at the end of each route. Some lines had different paths for round-trips, which added further complexity to the map. 

Final map before refinements Illustrator, QGIS 2016
Placing route numbers on map Illustrator 2016
Offsetting lines Illustrator 2016
Placing lines on offset guides Illustrator 2016
Offsetted final map before aligment of route numbers Illustrator 2016

Conclusion

The final result is very pleasing and interesting to watch. Today, two years after the completion of this project, I still find it interesting to watch this map, and I still would like to know how Accra’s people use this trotro network. I am pretty sure they would not use my map, but this study made me understand how complex networks can be mapped and schematized to produce meaning that is absent from geographically accurate maps.