Next generation analytics for next generation transport systems

Since the start of the decade, global stability has been severely compromised. The destabilising factors are well known: COVID-19, concerns about climate change and the geopolitical fallout from the tragic invasion of Ukraine. These global tremors are shaking the very foundations on which transport planning and policy have been built. In particular, remote working, social distancing, emissions targets and spiralling fuel prices are posing new conundrums for transport operators.

This ‘perfect storm’ is calling into question many of the traditional data-led approaches to planning which have been predicated on – first, extrapolating passenger trends into the future; and second, correlating these trends with relatively stable economic growth. And yet, in parallel, comes the promise of a more intelligent, data-driven service, as transport related-data grows exponentially, new analytics technologies emerge and operators evolve towards Mobility-as-a-Service (MaaS). The challenge facing operators and authorities is how to reconfigure transport analytics in this new ‘non-linear’ paradigm in order to realise the goal of intelligent transport systems. The experience of Ireland’s National Transport Authority (NTA) suggests that, while new approaches will undoubtedly be needed, data analytics will continue to be key.

Role of data analytics in overcoming COVID-19 constraints

During the lockdown era, public health was the primary consideration across virtually all jurisdictions. Data and, in particular, data analytics played a crucial role in optimising fleet usage subject to the constraints of COVID-19.

In Ireland, NTA was tasked with reducing capacity on the public transport system to levels as low as 25 per cent, depending on the prevalence of COVID-19. To achieve this objective, while at the same time providing a service that would meet the needs of commuters, accurately predicting passenger levels was crucial. In order to do this, we needed to overcome a common bus data problem – a lack of data on alighting patterns. This was achieved by building a social distancing model for our bus fleet based on our Boarding & Alighting Tool (BAT). The BAT is a statistical model which uses boarding data on outward and return/ onward trips in order to predict where passengers are alighting. Tasked with meeting the singular objective of social distancing, NTA’s BAT tool has proven to be extremely effective.

Implications of changing passenger trends in post-pandemic era

However, as we emerge from the lockdown era, more complex multifaceted challenges are emerging. As of late January 2022, public transport usage in Ireland was back up to 66 per cent of pre-lockdown levels, and this is before a more general easing of restrictions on 28 February 2022. While passenger numbers have risen again, the passenger trends are different to the pre-lockdown era, given the new hybrid work practices and the preference for core working hours over the standard 09:00 to 17:00.

In addition, lockdown has seen the public ‘get on their bike’ due to increased e-mobility usage via e-scooters and e-bikes. These trends – hybrid working, remote learning, e-mobility etc. – directly affect transport policy and planning. For example, the trend towards micro-mobility has the potential to solve the ‘last mile’ problem, which has been a significant disincentive to public transport usage. And while it is uncertain how permanent these changes will be, it is generally agreed that there is no returning to the old orthodoxy. All of this will have significant implications for network capacity and service design.

In this new era, transport planners, who have long sought demand levers over and above the relatively blunt instrument of fare changes, can now adjust demand to supply. In a sense, planners, armed with the right data, can now help to ‘flatten the curve’ of peak hours travelling patterns. Even a small change to peak hours can make a significant difference. Not only can this make public transport more attractive, but it can also lead to a more efficient use of the fleet. Efficiencies can be achieved by both maximising the load of the fleet across a broader time span, and by delivering faster journeys outside peak hours, thereby reducing the mpg ratio.

Aside from the environmental benefits, the economic case for fuel efficiency has become self-evident in recent weeks. In order to make this vision a reality, data analytics will be key. However, given that past trends will no longer be indicative of the future, new richer data sources harvested from next generation transport technologies will be needed in order to optimise transport systems for the new normal. In addition, the next generation data will need to be complemented by advanced analytic approaches, such as digital twins.

Utilising emerging data technologies

Digital twinning is a relatively new form of simulation designed around a two-way flow of information between sensors and models. By having better and constantly updated data related to a wide range of areas, combined with the additional computing power of the cloud, digital twins are able to model a greater combination of scenarios than traditional model simulations. This approach is extremely relevant for transport modellers who are seeking to optimise systems in the context of increasingly complex travel patterns, combined with new and multifaceted policy objectives.

However, in order to realise this optimisation, near real-time data of high quality is needed. To this end, the Irish government has just given NTA the green light for an ambitious re-design of our bus network, which will involve, among other things, investment in next generation ticketing and AVL systems. The introduction of these systems will open up a new data vista which will allow NTA to leverage new and emerging technologies, such as digital twins.

Realising the goal of an intelligent transport system

In short, public transport, having just emerged from the trials of the COVID-19 era, is experiencing even greater uncertainty as recent global events exacerbate the longer-term structural challenges facing the sector. This, in turn, is calling into question many of the traditional analytic approaches used by planners and policy makers. Nevertheless, emerging commuting and working patterns, accelerated during lockdown, combined with the timely advent of next generation systems and advanced analytics technologies, means that transport planners now have a golden opportunity to realise the goal of a truly intelligent transport system.