Seville is one of Europe’s largest cities to be without a mass transit network.
The Andalucían capital houses more than 1.7 million residents in its metropolitan area while serving as a major tourist hotspot for Spaniards, Europeans and global visitors. Meanwhile, its metro system offers just one line to carry citygoers, with private cars accounting for up to 50% of all trips.
Local and regional governments have been pushing to change the landscape in recent years. The city’s Sustainable Urban Mobility Plan 2030 (PMUS) and the regional Transport and Mobility Infrastructure Plan of Andalucía 2030 (PITMA) align on three priorities: expanding Seville’s high-capacity transit with potentially three more metro lines as well as tram and bus corridors; completing planned expansions of cycling networks; and improving integration across modes of transport such as with mobility hubs, park and ride facilities and integrated fares.
The vision for Seville is there but the next step is to turn its mobility plans into targeted intervention.
That’s what Bolt and ETH Zürich’s shared mobility research lab has sought out to do over the last year. The Sustainable Urban Transitions (SUT) Lab combines the front-line data of a global mobility giant with advanced modelling from a world-leading technical university to feed urban mobility results back to city officials and inform their transport decisions.
Seville was one of the first cities to join the Lab last year, aiming to investigate four key mobility areas: where to improve cycling infrastructure; where the safety hotspots are; where traffic bottlenecks occur; and how metro expansion should be integrated with walking, cycling and shared mobility.
The first-year results are now in.
Seville’s cycling infrastructure
To determine where Seville should prioritise its cycling infrastructure, the SUT Lab analysed more than 90,000 Bolt e-bike trips to firstly examine travel behaviour across the city.
The study, which took place between December 2024 and December 2025, found that the average e-bike trip was 2.6 kilometres and 16 minutes long. Men, visitors and younger users were identified as overrepresented riders.
The study found that tourists and locals ride similar distances, though tourists ride slower and for longer durations with trips more focused in the city centre and scenic areas. The crossover in tourist and local e-bike behaviour is one finding that stood out for ETH Zürich’s Postdoctoral Researcher Dr. Xuan He.
“The cycling network plays both the role of local transport infrastructure and supporting visitors mobility,” Dr. He tells Zag Daily.
For both locals and visitors, the built environment matters. Bike lanes, sidewalks, shelters, greenery and public transport stations were amongst the street features most strongly linked to higher e-bike volumes. Bike lanes in particular were found to have a strong association with route choice.
The link between e-bike volume and bike lanes, the SUT Lab concludes, demonstrates the importance of maintaining and upgrading Seville’s cycling network particularly on high-use corridors, bridges, riverside routes and routes connecting to public transport.
“Seville has strongly developed its cycling network over the last two decades. Therefore, the recommendation that appears more frequently is not about network expansion but about targeted improvements in the network,” explains Professor for Infrastructure Management at ETH Zürich Bryan Adey.
Identifying critical safety hotspots
Leveraging telemetry data from Bolt’s e-bikes, the SUT Lab was able to add another layer of safety data to complement official crash statistics in Seville.
Data from 10,400 Bolt rides was analysed to flag potential safety-critical events that could be seen as warning signs before a crash may occur. These events include instances of harsh braking, sharp lateral movements and strong accelerations associated with poor road conditions.
Distinct hotspots were identified in parts of the city centre and increasingly towards the urban periphery, including in some parks, while riverfront paths appeared comparatively safe.
The statistically strongest risk factors were higher riding speeds, higher road speed limits and nighttime riding. Female and older riders showed lower odds of experiencing potentially safety-critical events.
“Seville can use this insight to focus interventions on high-risk paths such as specific peripheral corridors, poorly maintained roads, nighttime routes and areas where micromobility mixes with high-speed traffic,” Dr. David Zani, Chair of Infrastructure Management at ETH Zürich, says.
Dr. Zani adds that speed management can be treated as both a “street-design” and an “operational” issue.
“An effective response could be a combination of speed calming measures, better protected cycling infrastructure and dynamic speed regulation such as geofencing in sensitive areas where technically and legally feasible,” he explains.
The report sets out a series of practical actions for Seville to prioritise such as investigating locations with repeated warning signs, improving pavement quality and maintenance, strengthening lighting, managing speeds more effectively and paying closer attention to peripheral areas where cycling infrastructure may be less protected or consistent.
How ride-hailing data can identify traffic bottlenecks
Steering away from micromobility, ETH Zürich then looked to Bolt’s ride-hailing data to better understand congestion and traffic patterns across Seville.
GPS data from more than 62 million data points were translated into link-level speed maps for Seville, with the aim of providing an approximation of traffic speeds across the city.
The maps distinguished between faster traffic conditions on motorways and slower conditions on urban streets. As expected, most road segments showed an increase in speed during off-peak periods compared to peak periods.
Some arterial links, however, showed lower speeds during off-peak periods. The Lab plans to investigate these segments further which could perhaps reflect behavioural differences like taking more relaxed trips during off-peak periods, or traffic signal timing that is optimised for peak periods.
Whilst the analysis requires further validation, it provides insights into traffic patterns by season, day of the week and hour of the day and the model can be extended to other cities with available ride-hailing data across different time periods and geographical scales.
In the next phase of the study, the SUT Lab research team aim to develop scalable workflows that use the ride-hailing data for more detailed congestion analysis, infrastructure impact assessment, traffic-model calibration and real-time or predictive traffic monitoring.
What Seville’s metro expansion should look like
With these findings in mind, the SUT Lab developed an agent-based model to simulate travel patterns across a second metro line currently in construction and a planned third line.
Simulated passenger demand estimates were found to be in line with measured daily passengers for Metro Line 1, suggesting strong usage and a natural mix between the two lines.
The research found a clear steer towards public transport after the implementation of Lines 2 and 3, identifying car-to-public-transport shifts of up to 18% at the district level.
The strongest gains were found to be in districts along the existing corridor. Sevilla Este, Triana, Los Remedios and Palmera-Bellavista were cited as areas that would particularly benefit.
“Metro expansions should be planned as a wider access strategy,” says Dr. Grace Kagho, Pioneer Fellow at ETH Zürich. “The simulations show that Lines 2 and 3 can shift trips from car to public transport but the benefits depend on how well the new lines connect to walking, cycling, buses, shared mobility and feeder services. Without that integration, there is a risk of concentrating benefits only along the Metro corridor.”
Pairing the Metro expansion with multimodal connectivity such as mobility hubs could lead to higher ridership, Dr. Kagho argues, making the investment more cost-beneficial while the benefits can spread to typically underserved areas.
Where Seville is headed next
Seville will remain onboard the SUT Lab for another year, with each of the four sub-studies to be further refined.
ETH plans to expand the e-bike behaviour analysis with further environmental and user-experience dimensions. Heat is one topic that will be focused on, with Seville and much of Southern Spain often exceeding temperatures of 40℃ in the summer.
“Heat is highly relevant because it shapes whether people ride, when they ride, which routes they choose and who feels able to use active modes,” ETH Zürich Professor of Transportation and Mobility Planning Eva Heinen explains. “This is a growing research topic worldwide and Seville provides the conditions to study it.”
On safety, the researchers will seek to refine their critical event analysis to better understand the specific infrastructure triggers of unsafe manoeuvres.
The ride-hailing pipeline for congestion analysis and model validation will also be refined, while the simulated metro expansion will expand to cover a larger area which models broader metrics such as accessibility, emissions, health benefits and cost-benefit ratios.
One year of the study is complete and the conclusion for Seville is clear: the city has the right vision for urban mobility but its transition will not depend just on infrastructure spend like new metro lines, but on the quality of first and last mile connections, the maintenance of cycling infrastructure and the integration of new mobility modes.
