When it needed to replace a six-year-old transport modelling system, Oxfordshire County Council saw an opportunity to innovate by identifying the outcomes needed, then giving firms the challenge of delivering on them.
To bring the required skills and technologies together, it has initiated a unique collaboration of partner firms invested in the project. It’s the first council in the UK to adopt this ‘solutions-led’ procurement approach and has positioned itself as a partner in the project, securing a stake in its success and potential to generate new income streams.
The partners in the new Transport Modelling Project are agent-based modelling firm Immense Simulations, location-based big data specialist GeoSpock, and two real-time journey analysis providers – Alchera Technologies and Oxford SME Zipabout. Oxford Computer Consultants is managing the consortium and will be handling user research, UX design, data visualisation and system integration.
The consortium will work together to build a ‘Smart City in a Box,’ a cloud-hosted agent-based model for transport scenarios which will allow Local Authorities and Infrastructure- and Housing-Developers to access transport modelling as a service. This ‘virtual world’ for Oxfordshire’s complex transport infrastructure will enable the council to assess demand for different types of travel and inform strategy and funding for improving transport provision in the region.
Lewellyn Morgan, head of Oxfordshire County Council Innovations team, explains they are replacing a system installed in 2013 which was based on mobile phone data. While pioneering for its time, the team felt they could do more to benefit from innovations resulting from projects like this:
“We found we paid private companies who benefited by winning more contracts. We had nothing against this, but next time round we thought if we’re going to be innovative, we should have a way of reaping the benefit as well as carrying the risk.”
During the next six years the county council worked with the UK Transport Catapult, since evolved into the Connected Places Catapult, to extend its connections and knowledge in the industry. The team wanted to look beyond the normal channels and seek out emerging companies in sectors such as the creative industries, where modelling is used extensively.
“Clearly the world was changing quickly and we began looking to industries such as gaming and learning from them,” Morgan says.
“We wanted to build a virtual real world where we can run multiple scenarios and have lots of users with a usable front end. Games do this all the time and build complex worlds with thousands of users.
“We weren’t experts but had enough confidence to aim for a step-change and there were people who could deliver. We could specify what we wanted and put it over to them.”
Finding the right partners meant initiating a process different to anything the team had done before, as this time they were asking firms to take a financial stake in the scheme. While the council needed firms who ticked boxes on innovation history, technological capabilities and financial strength, they also needed partners who would attract investment.
“We knew we were driving innovation and that is a massive opportunity for the industry,” explains Morgan. “As such, we expected partners to pull in investment on the back of the contract.”
While SME’s and startups were invited, Morgan admits the selection criteria presented barriers. A few at second-stage funding did come forward and provided they had proof of their achievements and passed all the checks, were in with a chance.
The team also found asking firms to become partners didn’t work for everyone, as firms are unsurprisingly focused on understanding where the boundaries for risk and responsibility are drawn: “Many boards struggled with the risk factor of the contract, because we were looking at a joint venture.”
Modelling is a key component in any transport planning process as it enables planners to run complex, multi-modal scenarios for transport systems that will help to inform decisions on infrastructure and funding. In Oxfordshire, the transport options involved in getting from A to B can range from walking and cycling to catching a bus, train or taking the car, plus plenty more in between.
But cost is a big issue, as modelling complex environments generates big data sets that require massive processing power and, importantly, the skills and resources to interpret them.
Councils in the UK have deskilled in recent years and often rely on consultants for basic modelling work. Even at normal levels, Morgan admits “modelling is really expensive” and that 70 per cent of Oxfordshire County Council’s transport spend currently goes on exactly that, a figure he describes as “horrendously high.”
Bringing skills back in-house isn’t easy either. When Morgan’s department tried to recruit a programmer, it couldn’t fill the vacancy. He points to the relative mundane nature of the work, compared to working in creative agencies, as a reason for this.
All this adds up to the council not getting the information it needs to interpret the various patterns of transport use in the region: “We don’t use models as much as we’d like and keep the number of scenarios down to a minimum. We might like to run 30 but we’ll run 10.”
This new project aims to deliver a transport modelling product that is relatively simple to use across the industry, from transport operators to developers. It hopes that cutting complexity for the user, by providing a direct portal that handles multimodal transport and cuts admin costs, is something people will be interested in using.
Morgan recognises there’s a built-in bias towards highways in the council’s transport policy and the current system is no good at factoring in cycling and walking alongside the use of highways. The transport model needs to be “a truly multi-modal system and take a more holistic approach, rather than assess each mode of transport side-by-side.”
He is keen to give people more choices about how they travel and make those choices easier, with solutions such as journey planning tools based on predictive analytics.
He also feels the need to make better use of public transport and this is where a data-centric model could inform transport providers on where to meet demand with supply. If, for example, the model shows that 50 people travel from A to B by car each day, the service providers could offer cheaper trips, such as chartered buses from the park & ride sites, as they know the demand exists to justify them.
Now the consortium has been formed, the project will go into proof-of-concept phase, assessment phase, then build, which is estimated to take about two years all in. Strategically, the team is looking at building a basis for transport planning for the next 10 years, with a modelling system that will inform what works and what doesn’t, what skills are needed and to adapt to new technologies and transport modes.
The consortium model itself, is intended to be flexible: “It’s an open consortium and we would expect to bring in more partners over time,” Morgan points out.
The project’s success will generate new revenue streams for the council, feeding into its targets for financial savings and potentially generate income. The procurement model itself has value elsewhere and Oxfordshire County Council has a longer-term plan to licence it out.
Reynold Greenlaw, director of innovation delivery for Oxford Computer Consultants compares the project with conventional contracts: “This is very different to the usual way of working with local government and the consortium including Oxfordshire County Council are innovating not only on technology but how to share the benefits of the results together.”
Morgan hopes all those who took part have benefited from a new way of thinking they can take forward: “We believe we have really challenged the industry to think differently about how they handle the technical elements of a project and about how to work with local government.
“Personally, I think this is the benchmark of how we should do lots of procurement and I hope we encourage more people to take this approach.”