Virtually, no queues into France

It is a seemingly unending scenario. But could a solution be on the horizon?

Electronic Specifier caught up with Simon Barnes, Funding and Partnership Development Manager at the University of Kent, to discuss the logistics and impact of a virtual queuing system.

Simon’s role is to work with academics and industry professionals to share research and innovation, particularly on large and complex challenges and projects.

What is a virtual queue?

If, for example, a truck breaks down whilst crossing the channel, a virtual queue means that a driver three-hours away will be informed of the delay and asked to park in their closest lay-by, until the backlog has cleared. Once the delay has been resolved, the driver will then be informed they are ok to continue their journey.

Simon and his team came up with the concept at the university, where they often look at regional challenges.

But how can they achieve this?

The idea brings together a plethora of technical and human resources, with Barnes insisting that data is the key to unlocking the whole system.

“The way that both the Euro Tunnel and Dover operate is it’s a ‘turn up and go’ system … you can just decide [to go] at the last minute … No one organisation really has the database to manage the virtual queuing system in its own right.”

The next level for traffic flow

According to Barnes, around 11,000 trucks go through Dover and the Euro Tunnel a day, so if there’s an issue at the tunnel or port of Dover, the implications are significant.

To really understand the theory of holding, potentially, hundreds of trucks in various locations throughout England, Barnes, along with the University’s School of Computing, accessed a network of cameras along the national road network to determine the number of trucks passing through each day.

The statistics were gathered over a three-month period, with the team noting peak time for traffic.

Smart technology is the key

The government are trialling a digital boarder system which the team are hoping to use as part of their virtual queuing system.

“A truck … will need to be sealed. It will need to have a smart tag attached to it which includes a GPS locator, and the driver will need to effectively verify his identification when he starts the journey – through some kind of system of face recognition or voice recognition. And then we need to track that truck,” comments Barnes.

In Lithuania, a virtual queuing system is already working successfully, so the team knows the system works in theory, although on a much less complex scale.

Modelling and simulation forecasting … data security, AI, and machine learning … and 5G too!

Once data starts to be collected, Simon and his team then plan to use modelling and simulation to ensure they are being proactive in their approach to combatting the traffic issues. It is a plan that will also integrate technology such as smart tags, biometric identification and 5G.

It is hoped that the University will act as an independent representative for any data gathered, as they do not have a commercial interest in it. They also have an institute for cybersecurity – so they know about the risk of sharing data.

And with technology like federated data and federated learning using AI and machine learning to move away from big data risk, there is less chance of a breach.

Barnes explains: “You ask the question, the AI then dives into everybody’s database, which is kept siloed on its own … It draws out the information to give you the results but doesn’t actually draw core or raw data out … it [is] a way of being able to combine datasets or information without compromising data integrity. So, everyone can keep their data protected.”

A system only works if the people do

Another hurdle to crossing the channel is getting drivers and companies onboard with the system.

Barnes states that unless you can get around 90% or more compliance then, no matter how great a system you have in place, it won’t work, and the benefit won’t be felt by motorists.

An early warning system

It is hoped that by determining patterns of truck behaviour, the system will also be able to serve as an early warning for potential human trafficking, or truck hijacking.

For example, if a truck is in-bound from another country, the system should be able to estimate when that driver will arrive at their destination. But, if that truck for any reason took a detour, then that could be an indication that something is wrong.

So, if a truck behaves in a way that is unexpected, then authorities will be notified. Thus, the hope is the potential for human trafficking and large vehicle hijacking will be drastically limited.

So, when will it be available?

Barnes says: “To some extent, [that] depends on how the trials go for the EcoSystem [the government trial] … it could take three years for everyone to agree to do it, [and] to get it up and running.

“It’s a really good example of collaboration potentially with SMEs, large data businesses, and universities to solve a problem.”

The teams have been working on the project for the last three years, and they still have some way to go. But ultimately, if they succeed, it will mean a smoother and safer channel crossing – for everyone.