Menu Home

A data-driven solution to build cycle paths — Making urban mobility efficient and sustainable

In this article, Vittoria Pauri, Giulio Consiglio, and Michele Nasini from Bocconi University, Italy explain how the use of big data can help make our urban mobility planning efficient and paradigms environmentally sustainable.

The global demographic trends show that every year more and more people live in urban areas. Hence, urban mobility has been a pressing issue for governments for the longest time. Innovations in the mobility could certainly help. Autonomous vehicles and supersonic trains immediately come to mind when thinking of innovation in the mobility sector. However, it might take a while until autonomous vehicles become widely adopted, leading to a complete revolution of the mobility paradigm.

In the meanwhile, some data-driven solutions could be adopted by governments to solve the problem of traffic congestion. Traffic congestion costs drivers and economies time and money. In the most congested cities, drivers waste more than one hundred and fifty hours per year stuck in traffic, according to INRIX (a private analytics company based out of Kirkland, Washington). The INRIX Global Traffic Scorecard provides robust insights on road traffic by analyzing data from diverse data sources, including not only road sensors, but also phones, cars, and trucks. This kind of mobility analysis has tremendous potential to support mobility-related policy decisions taken by governments.

Bicycle riding as a solution to traffic congestion

One solution which is always proposed when coming to the long-standing problem of traffic in urban mobility is bicycle riding, especially in smaller cities. In Milan, the second largest Italian city, for example, the average commute is less than 4 kilometres, as reported by the Guardian. This means that riding a bicycle, in such a small and dense city, might be an actual solution to avoid spending hours stuck in traffic jams (precisely an average of 98 hours per year, according to the INRIX Scorecard). However, to cycle a bike, cycle paths are needed.

Milan’s reaction to COVID-19: building more cycle paths

The lockdown imposed by governments in the wake of COVID-19 has led to unprecedented reductions in motor traffic congestion and air pollution levels. New measures are being implemented by many cities to gradually go back to normality after the lockdown period and to try to maintain these reduced levels of traffic and pollution. These measures are designed to bear a significant impact on public travel routes and transportation services: crowded subways and buses will remain a thing of the past for a while.

People will need to respect the social distancing norms on the subway. The capacity of public transportation services will be drastically reduced and waiting time will increase. Bicycle riding provides an alternative that is convenient and environmentally sustainable. Hence, the Milan administration recently announced that 35 kilometres of streets will be transformed to expand cycling and walking spaces.

More interestingly, the outbreak of the COVID-19 pandemic offers us the rare opportunity to rethink and finally change our consolidated mobility paradigms for the better. This is the moment when bicycle mobility can be promoted. Governments that are interested in addressing the long-standing issue of traffic congestion in their urban centres — to make urban mobility environmentally sustainable — should start designing policies for building cycle paths and encouraging the use of these paths by their citizens now.

Data-driven solution to build cycle paths

There are many bicycles on the same roads as cars. Clearly, the current number of cycle paths are just not enough to accommodate the needs of even the current number of bicycle riders. This only implies inefficiency in planning cycle paths. The yearslong processes of feasibility studies and planning have to be substituted with a data-driven approach to identify where cycle paths need to be built.

To do so, huge amounts of data on urban mobility are needed. The INRIX approach to mobility analysis should be considered a good starting point. Data can be collected from phones and cars. Examples of data that can be used are the ones provided by the geolocations of apps and phones, vehicle-to-everything (V2X) or vehicle-to-infrastructure (V2I) communication systems, and data generated by bike-sharing and scooter-sharing companies. By merging all these datasets, governments could get precious insights on population movement, traffic peaks, and accidents.

Since all public administration units have a limited budget, the inefficiencies in prioritizing spending between different locations for building cycle paths could be drastically reduced by using insights from these datasets. It is indeed possible to implement city-specific algorithms to calculate a bicycle suitability index. The city-specific algorithms could be implemented by detecting patterns behind the data that is collected from thousands of movements per day.

By using different kinds of algorithms and methods, such as the Fuzzy Hierarchical Process Method or a minimum-path algorithm, each road in a city can be classified with a score, which would be higher if the road is suitable for bicycles (i.e., it is wide enough, the pavement is regular, is not to trafficked, etcetera.) and low if it is not. At the end of the process, all the roads are mapped and classified (e.g., from absolutely bicycle-suitable to not bicycle-suitable) and the local administration can prioritize building cycle paths on the absolutely bicycle-suitable roads.

This blog culminated from a student group project at Bocconi University, Italy (comprising Vittoria Pauri, Giulio Consiglio, and Michele Nasini) dedicated to analyzing how the city of Milan should start using big data to build cycle paths more efficiently.

Categories: AI and Urban Planning Blog

AI Policy Exchange

AI Policy Exchange is an international cooperative association of individuals and institutions working at the intersection of AI and public policy, with the mission to produce deliverables that can create an AI-literate society and inform better AI policies.

%d bloggers like this: