Elly Yates-Roberts |
In the first article of this series, we introduced the concept of the Intelligent Transportation System, or ‘ITS’ – coordinated systems for moving people and freight which run on data and match supply to demand in the most efficient way possible. This article will start to drill into how this can be accomplished.
Let’s focus on a specific aspect of the ITS – the coordination of passenger transport within an urban environment. In today’s world there are usually multiple modes of transport available for any given journey, which might include interchange with longer-distance transportation at some form of terminal. Chaining different forms of transport together yields ‘multi-modal’ journeys, and there are even applications available, such as Citymapper or Moovit, which provide some level of support for identifying options and coordinating such journeys.
So where’s the problem? The issue is that the underlying forms of transport that constitute these journeys are typically not coordinated with each other in any meaningful way. This causes friction at the transition between different modes. The more steps in such a journey, the less reliable it becomes, and the efficiency of the journey usually trends downwards.
To make this problem worse, these applications can do relatively little to help passengers dynamically manage their journeys as the plan starts to fail. Passengers might even be locked into ticketing structures which are inflexible and incentivise them to carry on with a flawed plan rather than switching to a better one.
The most significant underlying success factor to address these issues is the availability of real-time data that describes the behaviour of the different components of the transportation system and integrates this with real-time demand for services. Note that for this we do not need personally identifiable information, information of true commercial value, or proprietary information that might represent one company’s competitive edge over another.
If this data is readily available, individual operators of transportation services can immediately start to use this information to improve their own operations, even though there is no optimisation being done at a ‘central’ level.
I’ll provide a few examples of this to highlight the opportunity. An urban rail station will represent a start point for many intra-urban journeys, and the end point for others. Historically, a taxi booked to collect someone from a train would have to wait until the train gets in and the passenger shows up. The vehicle and driver are sitting idle, taking up space and possibly generating emissions.
Now let’s contrast this to a future mobility service that takes advantage of real-time data. They can associate requests for service with the arrival of a particular train, and then allocate vehicles dynamically to provide that service. Assuming sufficient scale, the operator can work at optimising the positioning of vehicles in the area while having them do other jobs nearby until they are needed. If the train is late, the workload is pushed back in the schedule to compensate. In this way, the overall system efficiency is optimised and enhanced.
Even operators of less flexible services can benefit from such real-time data. At a minimum, their schedules can be optimised to take account of probable delays based on historic performance, and slack in their schedules can then be tactically deployed to increase intermodal opportunity – for example near rental bike stands.
Such real-time performance and state information can also be used to help car-sharing companies and micromobility providers figure out where to place their inventory of vehicles at different times, and even influence demand to help achieve the best positioning. That might sound like a stretch, but dynamic pricing of journeys with better rates available for journeys that suit the operators overall plan better can help with overall system optimisation and is likely to be particularly effective in an under capacity scenario – rather than applying ‘first come first served’ to resource contention, you take into account the demand that provides the greatest benefit to the system by getting vehicles in the right place for the next journeys.
There’s nothing here that is outside the scope of today’s analytical services and machine learning – provided you have the data. So to get started on the ITS journey, you need to invest in a city mobility data hub and persuade your operators to populate it with data.
The next article in this series will explore the separation of duty between different software solutions in the following generation of ITS.
John Stenlake is director of vehicle innovation and mobility for automotive, mobility and transportation at Microsoft
This article was originally published in the Autumn 2021 issue of Technology Record. To get future issues delivered directly to your inbox, sign up for a free subscription.