Why aren’t transport planners using mobile phone data? Missed opportunities in transport planning
- StratPlanTeam
- Jun 18
- 5 min read
Mobile phone data can be annonymised for reliable, real time transport planning
As cities grow, mobility needs become increasingly complex. Transport planners are tasked with designing systems that reflect real-world behaviours—where, when, and how people move.
Traditionally, this insight has been derived from costly and infrequent travel surveys or data collected via road sensors and ticketing systems. But a far richer, more continuous source of information already exists: the data generated by our mobile phones. Known as Mobile Phone
Data (MPD), this form of passive data collection has been heralded as a revolution in transport planning, offering lower costs, greater coverage, and timely information. Yet, despite its promise, MPD remains underused in many planning contexts.
This article explores what MPD offers, where it’s being used effectively, and—more importantly—why planners aren’t making full use of it.
The appeal of mobile phone data in transport planning
MPD refers to the anonymised location information gathered from mobile devices as they interact with cell towers. This interaction provides a continuous trail of where people are, when they move, and which routes they take. Unlike conventional surveys or ticketing records, MPD does not rely on voluntary participation, predefined routes, or specific transport modes. It offers a broader, more dynamic picture of population movement.
Several compelling advantages make MPD especially attractive. It costs far less to obtain than large-scale surveys, which are labour-intensive and often infrequent. The data can be gathered at a national scale and refreshed continuously, allowing planners to detect trends, respond to events, and model future scenarios with far greater confidence. Moreover, MPD reveals actual behaviour rather than self-reported patterns, reducing the biases that plague survey responses.
MPD also excels in its spatial resolution. It enables analysts to see movement within and between neighbourhoods, helping to understand commuting flows, population density shifts, and regional travel patterns in much finer detail. As ticketless public transport systems become more common,
MPD offers a primary data source for passenger tracking without requiring physical validation points. Traffic modelling, which traditionally depends on sensors and historical records, can also be vastly improved by MPD, enabling a more detailed understanding of journey lengths, peak congestion points, and road use.
Practical uses: transport statistics and system design
One of the most established applications of MPD lies in transport statistics. Agencies have used it to calculate population concentrations, track flows between districts, and estimate aggregate travel distances such as total personal miles travelled (PMT). In contrast to the static snapshots offered by census data, MPD provides a moving picture, allowing authorities to monitor how travel evolves across days, seasons, or in response to policy changes.
For public transport, MPD helps map journey patterns even where ticketing systems don’t exist or where riders switch modes mid-journey. In the case of urban buses or light rail systems with open access, MPD may be the only consistent method to estimate patronage. As more systems adopt virtual ticketing or shift towards integrated mobility platforms, reliance on mobile data will only increase.
In road transport, MPD offers a way to supplement or even replace traditional methods such as vehicle counters or fuel consumption modelling. While it may not yet replace sensor-based accuracy for real-time speeds or traffic flow at specific intersections, MPD provides a broader context, capturing route choice, journey purpose, and even demographic nuances based on user profiles.
Furthermore, active transport—such as walking and cycling—has historically been difficult to track. Survey participation is typically low among users of these modes, and infrastructure like bike counters is rare. MPD helps overcome this by showing all forms of movement, not just those tied to motorised or ticketed modes.
Real-world experience: Indonesia’s shift from surveys to mobile data
One of the clearest demonstrations of MPD’s potential comes from Indonesia. Faced with financial and logistical challenges in maintaining traditional household travel surveys, the national statistics agency, BPS-Statistics Indonesia, began exploring MPD as an alternative. Conventional surveys were expensive, limited in geographical reach, and slow to deliver results. As a result, they were conducted only every two years and covered just five major urban centres.
In collaboration with Telkomsel, a major mobile network operator, and the Indonesian Ministry of National Development Planning, BPS launched a pilot project in Bandung. Using the mobile data of more than 50,000 subscribers, they developed algorithms to map commuting flows and delineate the boundaries of the metropolitan area. This allowed them to identify actual travel patterns, comparing them to legal metropolitan definitions. The results revealed mismatches, offering a more evidence-based basis for planning decisions and regional policy.
Validation was carried out using a travel diary survey app completed by volunteer participants, allowing for the cross-checking of algorithm outputs against self-reported journeys. This iterative process improved the reliability of the MPD-based models. Moreover, the entire process took significantly less time and money than a traditional survey, and it delivered more granular insights.
So why isn’t mobile phone data widely used?
Despite these advantages, many transport agencies remain hesitant to adopt MPD. A major reason is the complexity involved in interpreting the data. Unlike structured surveys, mobile data requires careful filtering to distinguish between trips and non-trips, determine dwell times, and define appropriate zoning systems. For example, distinguishing a stop at a traffic light from an actual destination requires algorithms calibrated with local knowledge and transport context.
There are also concerns about data privacy. Even anonymised datasets raise questions about surveillance and public consent. In many jurisdictions, planners are still navigating legal and ethical frameworks for how mobile data can be accessed, stored, and used without infringing on individuals’ rights.
Another challenge is methodological maturity. While the data is available, there’s still work to be done in refining how mode of transport is inferred, especially for shorter trips. Errors in classification can mislead investment priorities or infrastructure design. Researchers caution that mobile data is best used over long sampling periods to allow for statistically significant conclusions, particularly when attempting to assign trips to specific modes.
Institutionally, many planning bodies are structured around legacy data systems. Integrating
MPD into existing processes may require new skills, partnerships with telecom providers, and updated analytical frameworks. There may also be resistance within organisations unfamiliar with the intricacies of digital data processing or concerned about relying on third-party providers.
Opportunities on the horizon
Nonetheless, the direction of travel is clear. As cities become more digitally enabled and as planning becomes increasingly data-driven, MPD will move from optional supplement to essential foundation. The richness of mobile data makes it well-suited to address new challenges, including real-time crisis response, climate-resilient transport design, and equitable access assessments.
Importantly, MPD enables longitudinal monitoring. Changes in commuting behaviour following a new policy—say, congestion pricing or extended public transport hours—can be tracked almost in real time. This allows for a more adaptive approach to transport planning, where interventions can be refined in response to actual outcomes rather than projections.
There’s also untapped potential in layering MPD with other digital sources such as GPS tracking, social media activity, or IoT-enabled sensors. These integrations could offer a multidimensional view of urban life, allowing planners to understand not just movement, but purpose, sentiment, and experience.
Conclusion - its time to accelerate use of Mobile Phone data for transport planning
Mobile phone data represents a transformative shift in the way we understand and plan transport systems. It offers unprecedented detail, scale, and frequency in monitoring how people move. Yet, uptake remains uneven, largely due to institutional inertia, concerns over data protection, and the technical complexity of extracting reliable insights.
Still, the examples from countries like Indonesia show that these challenges can be overcome, and that the rewards are significant. With the right safeguards and analytical rigour, MPD can become the backbone of modern transport planning. As travel behaviours grow more dynamic and cities more interconnected, relying on outdated data collection methods is no longer sufficient. Embracing mobile phone data will not only make planning more responsive and inclusive—it will help cities move smarter and better into the future.
See more here: https://www.georgejamesconsulting.com/
References
Comments