Transport operators the world over are being challenged by demand exceeding capacity but are often unable to build additional infrastructure as a result of availability of space and/or funding. For example, in just five years, the annual number of passenger journeys on Transport for London (TfL) services has increased by half a billion – and demand continues to grow rapidly.
Congestion frustrates customers, increases the wear and tear on assets and slows down the network. At peak times, traffic flows unevenly, and buses and trains become increasingly delayed at each stop as the crush of passengers trying to get on board stops swift arrivals and departures. In other words, congestion creates more congestion.
Yet even as some elements of the network are overcrowded, others have spare capacity. We are increasingly seeing the use of real-time demand forecasting and pre-emptive signal sequencing and information publication – enabling optimised use of the infrastructure capacity, and relieving the pinch points that appear during the busiest periods.
Passenger information has significantly improved in recent years, with announcements, websites, emails and app notifications alerting people to hold-ups and suggesting alternative routes – but with everyone diverted, these messages can create new bottlenecks elsewhere. Similarly, while motorists’ satnavs try to divert them around traffic jams, the different providers – lacking coordination or predictive capabilities – simply divert everyone in the same direction thereby creating new queues on minor roads that are even less able to cope with demand.
Technology and data requirements
Technologies have made significant advances in recent years but our infrastructure capacity remains stretched, particularly in the capital. Focussed investment in tools has significant potential to improve journey times, travellers’ experiences, and investment returns across all of our major cities and our national transport networks. Given the growth of mega-cities and digital infrastructures around the world, they could also produce a useful export industry for British business.
We already possess much of the data and the underlying capacity and demand models we need to solve these problems, and the connectivity to get the right messages out there. Transport Authorities & Executives such as Highways England and Transport for Greater Manchester know this, and have been developing data analysis and communications systems to optimise use of road space, whilst Network Rail are developing the basis for similar solutions through the Digital Rail programme. In London, the established databases serving Oyster card, contactless payment, Congestion Charge and Santander Cycles users provide powerful assets: contact details for many travellers, plus an understanding of their typical travel patterns. Recognising this, TfL are investing heavily in tools and resources to analyse data and tailor customer communications.
The growth of data collection in respect of individual customers together with multiple engagement channels and the use of individual customer accounts allows transport operators to segment travellers experiencing congestion or delays and suggest to each group a different way to reach their goal – making best use of the available capacity and enhancing customer service. Using emails, SMS and apps, operators can offer passengers incentives if they take a particular route, travel at a particular time, or use a particular mode. And with real-time data coming in on recipients’ behaviour, operators can quickly adjust their messages to focus on the most effective incentives and the most responsive travellers.
Meanwhile, TfL is investigating the use of mobile phone network data to track increases in road traffic in real time; this, along with the growth in ‘connected cars’ – which transmit data on their movements and satnav destination – will soon provide transport managers with enhanced tools to predict and respond to the formation of traffic jams in real-time. Automated routines can then be used to amend traffic light phasing to ease the bottleneck; given coordination with the in-car navigation providers, it may also be able to segment incoming drivers and calculate the best way to get each to their destination without creating fresh hold-ups elsewhere. In the absence of automation these systems will also provide enhanced decision support for Traffic Managers. Computer-based learning will capture the outcome of the action taken and record this to educate future automation or suggestions to Traffic Managers.
Mobility as a Service necessitates knowledge of journey purpose
Knowing the purpose of the customer journey is a more challenging issue but one that is of importance to address in order to manage demand both optimally and equitably. A family of four travelling on holiday with a car full of luggage are unlikely to change mode of travel; however an individual travelling for a business meeting is potentially more likely if he/she has relative certainly of not missing that meeting. The problem lies in collecting this data and, currently, most passive collection solutions (e.g. Mobile Network Data and GPS analysis) are able to determine “normal” routes – identifying what is likely to be a home to work route and what is a-typical, but not an individual’s propensity to change or the purpose of an a-typical route. Short of asking all travellers to register normal journeys and modal shift preferences there is a clear requirement for more granular but non-invasive mechanisms for determining journey purpose and responding to this with tailored options to deliver Mobility-as-a-Service (‘MaaS’).
The optimal solution would result in strategic management of demand across public and private transport. Then, for example, a London-bound driver heading down the M40 into a major traffic incident could be told how much time they’d save by stopping at the Oxford park & ride and taking a train or coach; but only if this is appropriate based on the purpose of the journey.
Ultimately, this combination of data, analytics and personalised messages could dramatically strengthen our ability to use infrastructure at close to its optimum load, but not above it – taking full advantage of the system’s capacity, whilst avoiding the need for investments that only pay off at the busiest times.
Trust in accuracy and consistency of messages is key
As we develop these systems, we will encounter many hurdles around the technology, the data-gathering, the analytics techniques and the communications systems. As ever, though, the biggest challenges are likely to lie in persuading and organising people. Travellers will only listen to messages if they trust the source: if they’re confident both that organisations’ use of their personal data is both ethical and transparent, and that altering their route will produce the promised benefits. Passenger instructions and communications need to provide consistent messaging which, if acted on, delivers beneficial outcomes – both on an individual basis and for the transport system as a whole – without penalising users of the transport network with seemingly little overall benefit (e.g. camera-enforced average/temporary road speed restrictions). This applies to routing, mode used and speed restrictions (e.g. managed motorways for vehicle drivers). Only in this way can sufficient trust be built in the user base to provide confidence that instructions will be adhered to in order to deliver the aforementioned outcomes.
Transport authorities and operators must work together to deliver results
This in turn requires good co-ordination between TfL and other transport infrastructure managers and operators way to manage the flows of data around the system, with the tools and relationships to gather data from – and transmit messages via – all the key actors guiding and carrying travellers around our transport infrastructure. As we move from connected to autonomous cars – strengthening technology’s role in deciding vehicles’ routes – we’ll increase our ability to manage traffic flows across the whole network; and the role and powers of a central authority will become still more crucial to realising the potential of these technologies. Integration and interaction should be broader than operators in a single mode (e.g. road) and should bridge both public and private transport such that, for example, passengers delayed en route to an airport for a flight can be fast-tracked through security scanning and an informed decision made on delaying flight departures (subject to regulatory and other commercial considerations). Similarly, park and ride operators who can predict demand and arrival times for customers in real-time may amend bus schedules to optimise both the customer experience and vehicle scheduling.
If we develop these systems in the right way, we’ll not only route travellers around congestion, but also reduce the amount of congestion in the first place – helping citizens, transport managers and infrastructure investors to get the best out of our hard-pressed transport networks. At the same time we will help operators and authorities to maximise capacity of their transport networks and to realise cost efficiencies in enhanced management.
KPMG’s Transport Advisory Practice comprises 30 full time partners and 118 full time staff specialising in providing advisory services to the sector spanning strategy and operating model development, economic appraisals, financial modelling, technology, data insights and assurance. Integration of transport systems is a core mindset of our Transport Advisory Practice and fundamental to our approach to delivering services. Recent examples of our intelligent transport systems experience include: supporting metropolitan transport authorities/operators worldwide on the implementation of integrated Fare Payment systems, advising several national road and highways authority on the development of advanced traffic management system for their strategic road networks, building predictive traffic modelling and management solution using a wide array of data signals, developing predictive asset management and vehicle maintenance solutions using demand data and asset information; and developing multi-modal revenue forecasting models for bus, tram and rail operators.