Mobile phones have become ubiquitous in the past 15 years and not just in developed countries. Currently about 50% of people in Sub-Saharan Africa own a mobile phone, and this number is expected to continue to rise significantly.
Meanwhile, urban mobility has become a source of increasing challenges in developing and developed cities. The number of cars is growing faster than the infrastructure can catch up with and policymakers simply have no practical data on how populations are moving.
What’s the link between these two phenomena? Actually, the first can help solve the second. How? Think about this: every time you make or receive a call or a SMS, your phone operator keeps data such as time, location, length of the call, etc. From there we can deduce lots of
mobility insights: where are people commuting to and from? At what time? How long does it take them? Where and when are there traffic jams?
The Kampala case
Kampala is famous for its traffic jams. A one hour trip to the airport can easily become five hours during peak time. The city is growing fast and is expected to triple in size over the next 35 years. In order to achieve well-mastered urbanisation, Kampala needs roads that are large enough to absorb the traffic in the city, as well as a development plan to anticipate the future expansion of the city.
Like many other big cities of the region, Kampala has decided to massively invest in a Bus Rapid Transit (BRT) system. The critical question is, where should they start? We’ll show how telecom data can be used before, during and after the setup of a BRT, optimizing the resources needed as well as the system itself.
Public transport planning demands answers to questions as where should the bus lines be installed, as well as the bus stops, the interconnection hubs, etc. In order to make the best decisions, Kampala’s policymakers need to know how many people live and work in the different districts, the traffic demand, and how those numbers will grow in the future. Manual traffic counting and surveys are extremely time and money consuming, creating issues with the city council’s data. Those traditional techniques of collecting data are mostly inefficient and fail to give a holistic view of a city’s mobility.
The good news is, telecom data can be leveraged to obtain high-quality mobility insights, at low cost. Every time we make a phone call or send a text message, our telecom operator automatically stores metadata of our activity: where we are located, the date, the time, the duration of the call (but not the content of the conversations). Aggregating and analyzing this information, known as the Call Detail Records or CDRs, is like finding a goldmine in terms of mobility insights.
It gives a very precise, dynamic and complete view of mobility patterns within the city at fairly low cost. Since the aggregated data are anonymized, there are no privacy issues. As a result, Kampala’s policymakers can get insights about commuting patterns, commuting flows, and commuting times, which is key to shape the transport system and adapt the capacity of the lines at peak times. Knowing when and where those peak times occur can be helpful in avoiding bottlenecks because the city office can offer alternative routes or encourage/discourage people from traveling between certain times.
Infrastructure and public transport planning require huge investments from city authorities, governments and international organizations, such as the World Bank. Unfortunately, the impact of these investments is rarely measured. Again, telecom data turns out to be a powerful tool to assess both the short and long term impact of mobility investments.
For example, the city might be faced with a strike day or a road cut off due to work. Analyzing telecom data on those days creates a better understanding of the price of these temporary issues and helps transport planners take the most appropriate predictive measures for the future. In the longer term, monitoring traffic through CDRs can also help the transportation system evolve. New lines, new stops and new roads can be planned to keep things running as smoothly as possible.
Further on, CDRs can be used to test and improve a city’s resilience to some unpredictable events like floods, earthquakes or terrorist attacks. In such emergency situations, the focus is often put on short-term aid actions. Understanding population flows and locating the most vulnerable communities are crucial to prioritize actions and improve crisis management.
Mobile phones are among the richest data sources ever. They are distributed across entire populations all over the world and give vastly more information at a lower cost than traditional data collection techniques. When we aggregate the data from these millions of sensors, we can build mobility models across entire cities or countries just as we did in the above case study. Given the economic and social cost of traffic congestion in developing as well as developed countries, telecom data’s potential for solving the mobility issue is both huge and broadly underexploited.
By Nicolas Snel, Product Owner Data for Partners at Real Impact Analytics