Making a city smarter not only relies on understand technological opportunities but also human decision-making, as Miller Crockart explains.
Cities are about people – a fact that can easily be forgotten when experts talk about roads, healthcare and education as though they are abstract and unconnected monoliths rather than things people use.
Understanding how and why people use services is vital for making decisions on how they can be optimised for maximum efficiency across inter-connected networks that form the basis of modern cities.
For example, why do so many schools and institutional buildings have significant traffic problems around drop-off and pick-up times? Because very few have been located or designed thinking about the needs of the users. And when British educational policy changed to allow parents to, within reason, send their children to the school of their choice rather than the one nearest their homes, was any thought given to how that might affect traffic and any mitigation measures?
Why do we need traffic modelling?
Smart cities require an understanding of what people want and need and how one decision might impact on a wide range of other, superficially unconnected, services.
Technological advances, the wide availability of data and improved processing power, enables computerised models to be built that predict the effects of population changes and determine what decisions must be taken to ensure a successful outcome, whether that’s related to mobility, infrastructure, policy or regulation.
Such models take into account factors including population demographics, public transport availability and network pinch points which can constrain growth. They can even reflect how weather impacts peoples’ habits and how remuneration affects the distance commuters will travel to work.
This allows analysis of the knock-on impact of making a certain decision. For instance, building a new railway station may reduce the number of long-distance car journeys but increase the number of shorter journeys between a traveller’s home and the station. Alternatively, the traveller may choose to take a bus, thereby increasing demand for bus services and reaching capacity limitations on a particular route – maybe justifying a new or amended route.
The ability to foresee and analyse the ripple-effect of individual choices means that planning for smart cities can be made using informed decisions; from identifying and quantifying a problem to finding a suitable solution. Using empirical analytical models which take data from the wide pool of information sources means cities really can make sure the solutions they have identified provide a good outcome for all stakeholders.
By understanding why and how people move about reveals how they make their choice of transport mode or what time they need or prefer to travel.
Increase efficiency
Modelling helps cities deliver services more efficiently to cope with ever-growing populations without everything grinding to a halt.
The latest software can model a range of measures from analysing a change in public transport timetabling and routing to network changes such as the number of access points for a new housing estate or business district. In terms of traffic, does the one-way system reduce conflicts and smooth traffic flow or simply increase the miles travelled? Improvement in processing power and new computational techniques, means models which used to take days to run now only take minutes, so they can be run any number of times for varying scenarios to identify the optimum solution.
Mobility as a Service (MaaS) is an interesting proposition, but for people to reduce their reliance on personal transport mode – their car – they need to know there is a reliable, convenient and cost-effective alternative. And, MaaS service providers also need to know how to plan and operate a financial viable business.
To provide that smart city transport, authorities will need to understand how modes fit together in their city and how an incident on one part of the travel network will affect the rest. When traffic managers have real-time predictive traffic state data, they can immediately optimise the network if there is, for example, an accident on one particular route. Optima software can support decision makers decide whether, for example, it is better to reroute people on a secondary or tertiary route.
If a problem on the rail network causes more people to travel by road, how many buses are needed at the station and should those buses be prioritised through the network to allow people to get their journeys back on schedule? And if so, what will be the impact on other modes and would the overall performance of the movements across the networks as a whole, be better or worse?
By fusing real-time data with traditional predictive models, the latest modelling and simulating (Optima Public Transport ETA) is giving operators the time and tools to answer these questions.
Consideration of smart cities’ efficiency must include the ‘Uberisation’ of parts of the transport network.
Autonomous vehicles
It is important to grasp the topic of automation now, long before these vehicles come to our cities. This is at two levels: What will the introduction of AVs have in terms of demand on the city road network – plus or minus – and how will AVs interact with other road users and infrastructure?Consider this, autonomous vehicles will have their own schedule and route choice and traffic managers could end up with no authority to guide them. So authorities might consider allowing AVs to operate but requiring full disclosure of the vehicle’s movements in order to be able to give guidance and grant access to the road network (possibly for a charge). The city cannot disavow its responsibility to manage and control the road network.
PTV is already simulating the interaction of autonomous vehicles with other road users such as private cars, public transport, heavy goods vehicles, pedestrians and bicycles along with the interaction with new trends such as e-hailing, car, bike and ride sharing that will facilitate their arrival on our roads. This is being done at a microscopic level and in a safe virtual environment where millions of vehicle miles can be simulated and evaluated to provide an insight into what will happen when these vehicles are on our streets. It is about modelling that driving behaviour and decision making, whether it is human or automated, and demonstrating how that interacts with technology at a ‘smart’ or integrated level.
The latest traffic software can cope with these demands; scheduling movements of vehicles (public and private) and influencing multi-modal integration, enabling government and OEM’s and fleet owners and operators to be involved in every step of the decision-making, ensuring positive disruption and co-operation with our updates.
Roads, people and goods
Monitoring of data in real-time in order to optimise the use of finite resources at any given moment is already a reality. By fusing real-time data with existing models it is possible to predict the propagation of an event or incident on a network to identify and test mitigation strategies before they need to be implemented.PTV is working with academic partners around the world on these positive disruptive topics. In Australia, for instance, we are working with the University of Melbourne and state roads organisation,
ABOUT THE AUTHOR: Miller Crockart is vice president of global sales and marketing traffic software at computer modelling specialist PTV Group.
