StreetLight Data is on a mission to use big data to highlight real-world traffic patterns and make cities better places to live and work, by Sue Weekes
Sustainable transportation expert Laura Schewel co-founded StreetLight Data in 2011. At the time she was developing new techniques and approaches for analysing data as part of her PhD at UC Berkeley’s Energy and Resources Group. She believed her work could help unlock the barriers to data-driven transportation and joined with co-founder Paul Friedman who had experience of creating commercial data products.
The company’s platform takes data from smartphones, connected cars, fleet management systems and other sources and turns it into valuable mobility metrics. Its technology is used by provincial and regional governments across the US and Canada. The company recently published the e-book, Commutes across America: Where are the Longest Trips to Work? She talks to SmartCitiesWorld.
SCW: How would you sum up the goal of StreetLight Data?
To use scarce resources more efficiently. Whether you are talking about a major bridge or something small like parking meters, infrastructure decision-making requires planning and the impact of those decisions lasts a long time. If we get the decisions right, you can save a lot of money and positively impact people’s lives. StreetLight Data came about because I found that I wasn’t the only person in transportation looking for data, and there was nothing. Given how important the effects of transportation are to climate change and quality of life, I thought this was absurd.
SCW: Minnesota and Virginia Departments of Transport are among those using the tool to analyse transport patterns across the state. Are these typical use cases for the software?
Yes, fairly typical. We tend to have two types of clients: those who are undertaking a particular project, such as developing a new bike lane who would buy a project subscription, and those who buy unlimited analytics for state-wide use. The latter is where you can really unlock transformational benefits. Having the tool immediately available helps to change the behaviour of staff in agencies as they just look something up like they would on Google. It makes big data much more natural.
SCW: What are the main challenges you face communicating the benefits of the tool to government agencies?
It takes time to change mindsets and bring about behavioural change but we are starting to see it happen in forward-thinking DOTs such as Ohio and Minnesota. The other challenge is fragmented decision-making and budgets. We like to see central procurement and central data collection. Another challenge is that agencies, in general, aren’t used to buying software-as-a-service from a small business like us. The person who uses the data gets immediate benefits and they explain this to colleagues. It’s a case of one step at a time.
SCW: What did you hope to achieve with Commutes Across America
Around 30 per cent of travel is due to commuting. It’s something many people do every day and the length of a commute can be an indicator of happiness so we wanted to measure this. We started slicing and dicing the data and it became really interesting. Cities make really powerful decisions that create very different conditions and manifest in different ways. These decisions can have a major impact on a daily commute and in turn a person’s quality of life. If you are experiencing a very long commute, it is difficult to share that information with a city council and for the council to analyse the entire neighbourhood. Being part of a data ecosystem makes it much easier. Data can also dispel misconceptions. For instance, there is a perception that traffic jams often come from external traffic coming into a city but the data reveals that they are often due to internal traffic from the citizens.
SCW: Did some of the findings surprise you?
We could see that an average commute in one city could be much longer than the average commute in a similar city in a neighbouring state. Maine has the longest median one-way commute at 9.8 miles. That’s 72 per cent longer than the 5.7-mile median one-way commute in Wyoming, which is the state with the shortest commute in miles. We also found that in some cities there was no relationship between length of commute and socioeconomic factors but in others, it really mattered. We found that the likelihood of having a college degree is the most frequent, highest correlating factor of length of commute. For example, if you live in an Atlanta ZIP code where many residents do not have a college degree, you’re much more likely to have a longer commute than other Atlanta residents.
SCW: How will this data help to build smart cities?
It allows you to do two things: to design solutions that have a higher chance of working because decisions have been data-driven, and it allows you to talk to citizens about why you are making those decisions. The latter is really important because municipalities can say they have decided to give a bus route to this part of town because these are the people who have the biggest need. Identifying commute inequity is really important. One of our aims is to build data story-telling tools and the report gives us lots of stories to tell. Commuting is just the beginning of thinking about transportation in big cities but it is something everyone can relate to and talk about.
SCW: In general, what challenges does the smart cities movement face?
A few cities have done a good job of branding themselves as a smart city. But companies are throwing everything at them and it’s too much. There are big companies who see potential paydays for IoT infrastructure projects and there is some over-engineering going on. It is still very early and I feel some cities are spending a huge amount on assets that will be out of date in a few years or simply aren’t right and that is dangerous. Additionally, as a small business one of my concerns is that if big companies donate services for free in pilots it can stifle innovation. Big companies can give a lot for free but small companies can’t. Similarly, scaling up from pilots can be difficult. There needs to be more graduated programmes from pilots.
Overall, I think a lot of people and companies are working on the technology but we need to think about what this does to the city and the quality of life for people living in it and ensure we make smart cities equitable so they provide benefits for everyone.
The ebook Commutes Across America can be downloaded at smartcitiesworld.net/whitepapers/smarter-commutes---streetlight-data
If you like this, you might be interested in reading the following:
Tapping into StreetLight data
MnDOT will use StreetLight data to bring new traffic intelligence such as origin-destination and select link studies
Ohio DOT strengthens mobility intelligence
StreetLight Data will enable ODOT to transform Inrix data into actionable intelligence
Virginia taps into traffic intelligence
The US state is using mobility analytics company StreetLight Data’s technology to access on-demand traffic and transportation data