The Future of Multimodal Transport in a Self-Driving World
Published on October 4, 2019
Rohit Talwar
Rohit Talwar – Futurist Speaker
By Rohit Talwar, Steve Wells, Alexandra Whittington, and Helena Calle
What are the scenarios for multimodal transportation in a self-driving future?
Planning Transport for a World of Unknowns
City transport planners around the world are faced with an incredibly complex task of determining the transport infrastructure for the next 20-50 years. The complexity derives from the need to determine the transport and supporting infrastructure required for modes of transport that don’t all exist, carrying people who may not yet have been born, working in jobs and industries that may not yet have been created, with huge uncertainty on the resulting mobility implications. Now, the likelihood is that mass transportation—especially into city centers—at peak times will be required for at least the next two decades. However, planners are challenged by the fact that mobility patterns could change quite quickly. So, for example, economic growth could drive job creation and a demand for more transport into key work locations. However, a rapid rise in the pace of automation could see many jobs eliminated with a dramatic impact on transport demand.
One big excitement factor is autonomous vehicles which could potentially offer a more personalized service. For example, could autonomous vehicles pick travelers up from their homes and be loaded onto trains at the station? When the train reaches the city center terminal, the “train” would break into component vehicle parts and take each traveler to their end destination, thereby providing a “first mile/last mile” solution. While such developments may be a decade or more from fruition, planners need to be thinking of these possibilities now to avoid the need for costly infrastructure rework projects in the future.
Another interesting development we’ve been looking at is Flying Taxis. Following successful trials of single person passenger drones in 2017, commercial services are due to be launched in China and the UAE in the next few years. The technology is slated to continue to improve and, around the world, by 2023 more than 20 countries might have licensed the use of both single and multiple occupant passenger drones. From a planner’s perspective, again it is hard to determine the possible requirements for such services which may not even be part of current government strategy or have regulatory approval.
Trends Driving Multimodal Transport Thinking
One of the forces that is clearly important to transport is the autonomous driving trend that permits robotic devices and artificial intelligence to complete the supply chain. The future thrust toward self-driving cars is a key one to watch. A number of powerful separate trends are combining here to challenge our thinking about the different possible future multimodal scenarios we must prepare for. These include the automation of transportation, changing commuting behaviors, the evolution of delivery/supply chains, the application of AI to transport planning, the growth of home working, automation of jobs, and changing patterns of disposable income. There are also questions here about what might be the new opportunities and risks resulting from multimodal transport for global companies, and what might be the resulting innovative future growth strategies in multimodal transport? The big risk here is that we under- or over-provide and invest either too little or too much in transport infrastructure relative to likely demand. Predicting the exact volumes and transport mix is a difficult challenge under current economic and technological uncertainty.
The Use of Mobility Scenarios
Our approach to de-risking the planning process is to envision a range of possible future scenarios to help identify alternative strategy approaches. For example, one scenario we envisioned is Data Drives the Future. In this 2030 scenario, Transport for London (TfL) runs Greater London’s multimodal transport network using a fully integrated AI-based Travel and Transport Management System (TTMS). Vast amounts of data are processed using human expertise, AI-based transport infrastructure planning and traffic management algorithms, and predictive analysis—drawing on sensors in roads, pavements, and public transport access points.
What does this future look like? In this scenario, traffic and pedestrian flows have grown exponentially smoother, transport’s environmental impacts have declined dramatically, and London is ranked first globally on mobility. A single control center automatically manages and matches services to demand—combining autonomous buses and surface and subway trains, and road and rail signaling. Live predictive analytics allows greater use of road and track space. Autonomous boats ply their trade on the Thames from Putney in the west to Woolwich Royal Arsenal in the east. An automated fail-safe mode restricts public access to capacity-sensitive areas like underground stations and riverboat piers.
Manual drive cars of all fuel types would still exist, but only autonomous electrically powered vehicles are permitted in the city center. A constant flow of data between autonomous vehicles (their current location, destination, and purpose of the journey) and the central system would be used to re-route traffic around congested areas. The system also gives priority to public transport and emergency services. The system’s associated app provides pedestrians’ personal digital assistants with navigational information.
This is all made possible by embedded road sensors monitoring surface and sub-surface conditions. Traffic types and flows are constantly monitored against the TTMS’ comprehensive historic road status database—proactively undertaking maintenance. This reduces requirements for lengthier and more extensive subsequent repairs, minimizing traffic disruption by accurately re-routing transport resources during repairs, maintenance, and emergency situations, and predicting the implications.
These scenarios combine a range of driving forces on traveler behavior and demand and allow for the creation of alternative storylines about how the future could play out. There is also a strong element of visioning—articulating what we’d like to see happen in each scenario. This type of approach lets planners and policy-makers prepare for a range of possibilities and bring flexibility into the definitions of tangible goals and visions under different prevailing circumstances. In each case, the goal is still to serve humanity as best we can but the resources available and strategies we adopt might vary dramatically from scenario to scenario. These scenarios then become the basis of communication with leaders—enabling them to make better informed and more robust decisions.
The Impacts of Multimodal Transport on Individuals and Society
The personal example of multimodal transport that individuals might notice most easily is that of changes in how consumer goods are obtained and the resulting impact on personal time and expenditure. In future, the use of multimodal delivery services could see drones as one of the modes of “last mile” transport, as Amazon and Domino’s and others have experimented with.
A drone could deliver a pizza directly from the kitchen, or packages from a centralized drone port, where the goods have already been delivered by truck, train, or airplane. Another strategy would be the increased use of shared autonomous vehicles (such as Uber) for both consumer and goods transport. Passengers and goods might travel together to the same destinations, allowing for passengers to have the option to ride free and act as de facto delivery persons.
In terms of public transport, might train stations that are less crowded outside peak travel times host pop-up digital shopping malls? Autonomous and shared transport implies greater convenience and ease, so perhaps people will have more time on their hands to browse and shop during commutes. Conversely, train stations may become places where people can pick up consumer goods en route to their homes or jobs. This could transform their commute into a profitable or time-saving activity, or at least eat away at some of their personal transportation costs by becoming part of the supply chain.
Challenges Posed by Multimodal Transport for Governments
Clearly there is a major challenge around planning for the desired capacity, funding the infrastructure, building it, and coordinating schedules across the participants in the ecosystem. In terms of self-driving transport (trains, buses, cars, and taxis), one question government and local planners should be asking is “Will parking soon be obsolete?” In 10-15 years, the idea of a stationary vehicle may be an anachronism. In future smart cities, self-driving vehicles should be enabled by data to orchestrate smooth mobility with almost no stopping required, other than to let passengers on and off. Self-driving cars are but one current in a larger wave of change sweeping us toward this vision; another is the erosion of car ownership.
The rise of endeavors like Uber, Lyft, and Ola Car have popularized mass ride sharing, and it is imaginable that in a smart city environment, such programs could become the norm. The environmental and economic benefits, like cleaner air, and less traffic, might encourage creation of more pedestrian areas and green spaces—improving public health. Most of all, the ability to creatively reclaim spaces that are now devoted to parked cars could enhance the quality of life in cities, a key consideration in terms of the legacy for future generations. Policies that engender these changes in transportation can be win-wins for elected officials. Another challenge that multimodal transport poses for governments is how to implement a more commercial approach to offer a more attractive service for customers. A fluent dialogue between the governments and multimodal transport companies is essential to face the challenge of providing an appealing transport service.
Conclusion—Planning as Scenario Visioning
The rise of autonomous vehicles seems inevitable; what we can’t yet determine is what proportion of the total vehicle fleet will be self-driving, or what demand might look like. Equally we have little certainty about the future shape of the workforce and hence what the patterns of traveler demand might be. Given the scale of the uncertainties, planners must look to the use of a hybrid scenario thinking/visioning approach to articulate different possible driver combinations and outline the resulting multimodal transport strategies that they might pursue under each scenario. The key here is to put the citizen at the heart of the process and to be clear what the benefits and downsides are for the population in each case.
- Can transport business models expand to include the provision of a “one stop shop” door to door service, maybe by running fleets of autonomous vehicles that can utilize rail infrastructure as well as roads?
- What futuristic uses can rail companies put their infrastructure to? For example, could track routes become autonomous drone highways?
- How can train companies work with bus companies, hire car providers, and hotels to integrate traveler information and travel needs? What AI-led data sources can enhance the planning and provision of integrated travel services?