True data on modal share in Central London

What happens to Olympic parks after they host the largest athletic competitions in the world? How can cities take advantage of these vast spaces and increase walkability standards to ensure these parks are accessible to all modes of transportation? How can these multi-purpose spaces be best leveraged and redesigned to serve the needs of the community 5, 10, 20 years post-Olympics? We explored these questions in a recent partnership at the site of the 2012 Olympics – London’s Queen Elizabeth Olympic Park.

A sprawling 560-acre site in east London, the Queen Elizabeth Olympic Park is home to the park grounds as well as a buzzing innovation district and recreation area. The park’s developers, the London Legacy Development Corporation (LLDC), have an ambitious goal of making it the most active area in the city and accessible to all with new amenities including shops, green space, and housing. LLDC came to Fyma to benchmark the park’s current activity as well as the surrounding streets that lead to the park’s entry points so that they have a baseline to work from to set measurable metrics to track progress toward their goal.

Before working with Fyma, the planning team conducted labor-intensive surveys to understand modal share and purchased multimodal studies based on models, not actual counts from the very streets in question. Given the significant change in human behavior during the pandemic, LLDC was eager to use actuals rather than forecasted data and turned to Fyma to transform the park’s video footage into actionable insights.

Over an eight-month period, LLDC and Fyma partnered to collect real-time and longitudinal data to explore and understand movement patterns. Gathering data was straightforward as the park has access to a camera network and within minutes of accessing the camera feed, we collected a series of data points including counts of cars, pedestrians, busses, vans, trucks, bicycles, motorbikes, and e-scooters. We helped LLDC understand footfall and traffic patterns and how they varied daily and seasonally. We interpreted the data to develop accessible and actionable metrics in five geographic areas and shared insights on how to reduce the share of vehicles, increase walkability, and promote a modal shift from cars to active mobility.

During the project, we discovered that despite the gut feeling of there being too many e-scooters, we discovered that in the summer of 2021 there were not enough of them to even build a dataset for our AI training purposes. This, however, changed once lockdowns ended and schools and businesses went back in September last year. E-scooters are located mostly in the Westfield Stratford City area, in fact 70% of the e-scooter traffic is there. We also found out that around the park, car modal share can reach peaks of 90% during week-ends. However, at Westfield Stratford City, walking is the most dominant mobility mode. There was a major increase (around 50%) in all traffic modes after the lockdown ended, with the Here East area still at peak traffic capacity, with cars dominating the urban landscape.

Fyma continues to work with the LLDC on understanding their modal share and the effect of various initiatives across the park. An interview about the collaboration with LLDC’s Director of Innovation, Emma Frost in Diginomica.

Fyma’s tech is GDPR-compliant and it’s AI has never seen a human face. We are committed to being a privacy-first company and will never store your video data – we simply run the video footage through our AI that measures objects and their movement and automatically erase all video footage.

Get started
Your request has been submitted!
Something went wrong! Please try again later.
Ready to chat?

See how Fyma can help you gather accurate occupancy insights right from the start.

Object detection
and recognition


Specific location