Artificial Intelligence gives Queen Elizabeth Olympic Park a performance boost

Mark Samuels Profile picture for user Mark Samuels May 4, 2022 Audio mode
Summary:
The London Legacy Development Corporation is using real-time data to help ensure the effective flow of people and vehicles around the park.

 

An image of the Olympic rings at the Queen Elizabeth Olympic Park in London
(Image by Waid1995 from Pixabay )

Queen Elizabeth Olympic Park is using artificial intelligence (AI) to create real-time insight that helps improve the flow of visitors around its venues, transport systems and retail facilities.

The London Legacy Development Corporation (LLDC), which manages the park, has plugged 32 of its CCTV cameras into an AI platform from computer vision specialist Fyma for the past two years. Emma Frost, director of innovation at LLDC, explains to diginomica how this technology is providing new insights into how the park and its facilities are used.

Frost describes the park, which covers 560 acres, as “a really interesting space”. She says it has continued to evolve rapidly since the London Olympics in 2012. While many of the sport venues remain, the park also provides access to a range of business, culture, education, and retail facilities, such as the nearby Stratford City Westfield shopping centre.

These venues and facilities are linked by a network of roads, pathways and public transport hubs. Frost says AI has allowed her innovation team to transform the park’s CCTV cameras into smart devices that provide new data-led insights. The technology is helping to track the movement of people and objects around the area, which helps LLDC make important decisions about how space is used, improved and developed. Frost explains: 

Having a real-time data feed on the movement of people and different modes of transport in the area is very important to us because we need to know what the user numbers are and what the user approaches are in and around the park all the time, because it's changing so rapidly.

LLDC was also keen to use the Fyma AI platform, which has been adopted by businesses and local governments across other European cities and towns, because of its strong data ethics principles, says Frost. The platform is trained to never recognize or process human faces, for instance. 

Fyma blurs out human faces on images used to train its AI system, so algorithms – and the data science teams that process the information – never see any human faces. Camera-feed data is automatically deleted once it passes through AI analysis. Frost explains:

There's no grey zone there; the data is never actually captured. And as a public organization, it was very important for us – as we start entering into the world of AI and understanding different technology applications for better urban futures – that we were really mindful of data ethics, data standards and data protection at the very beginning of any of our trials and initiatives.

Analyzing alternative transport use

More than 43 million objects have been detected in the surrounding area of the park to date. Delving further into the data gives more detail on specific areas, such as transport. Take the example of Waterden Road, which is the route leading to the Westfield Strafford City shopping centre. Since the project began, Fyma has detected more than nine million people, around 1.6m buses, over 532,000 bicycles, and over 100,000 e-scooters. Frost says:

We've seen a mass an influx of new modes of transport. So, for example, we were the first place to test e-scooters. They're not legal on high roads. But because we're a private estate, back in 2018 we were the first place in the UK to actually test and run a trial for e-scooters.

The team is also using the AI system to analyze how people are using e-scooters. They consider key questions, such as the routes they’re taking and where potential conflict points with pedestrians and other forms of transport might occur. Frost adds:

Using the AI platform with our existing camera network means we get a real-time read on all of that intelligence. We actively had to train the AI platform to be able to detect the e-scooter shape and form. And that was a big part of this trial to make sure that we could get high-accuracy readings on detecting these scooters.

Insights from the AI system, which also examines trends such as bicycle use, to the popularity of bus stops, and onto the flow of people across the park, will help LLDC as it works to make the area more accessible, user-friendly and sustainable in the coming years. Fyma’s system was originally implemented as part of a six-month trial last year. Now the technology has proved its worth, Frost is keen to do more. She says:

We're now going to extend that trial. We're going to be expanding from the 32 cameras that primarily looked around the edges of the park and we’re going to cover most of the inside of the park, with some key priority areas. So, it's really an extension to look at more detailed areas between venues and road networks, so we can analyze the next level of detail down.

Future opportunities

A good example of where they want to go next is using AI to inform LLDC’s retail strategy. Frost explains:

That’s about understanding in minute detail about the movement of people walking and where the dwell times are largest, and how those levels change over weekends, whether there’s seasonal patterns or whether there's a West Ham United game taking place in the London Stadium or not.

LLDC and its partners could use this data to create real-time operational advantages, such as changing opening hours to reflect higher visitor numbers. More generally, she says the application of Fyma at the park shows how any organization looking to use AI must be aware of important cultural considerations. Frost says: 

With that comes all of the questions of the unknown and dealing with the ethics, the risk appetite and the uncertainty – and sometimes, just a lack of understanding. So there's a big cultural and education piece that needs to come alongside any introduction of technology projects – and data projects, especially.

LLDC is using Fyma’s data science capability to help understand the insight that’s being captured. In the longer term, Frost says LLDC will have to think carefully about how it partners with other organizations to access data talent. The organization might even need to develop expertise internally. What’s crucial to recognize, says Frost, is that the data-enabled work she’s undertaking now forms part of a much wider digital innovation strategy: 

The park has been built with amazing digital connectivity. Now, we’re really trying to understand what we need to put in place in terms of the learning and the culture and the management of that infrastructure.

A grey colored placeholder image