Adaptive Frequency Public Minibus Service for Towns and Small Cities

Abstract representation of a network of electric minibuses

This story follows up on Thoughts on the Energy (and Carbon) Saving Potential for Personal Mobility by looking at one aspect only: public transport in the form of a smart minibus service for towns and small cities. This kind of service would employ electric minibuses and operate at an adaptive frequency schedule generated in real-time, possible by using machine learning, data science and mobile phone technologies. The resulting service would be most suitable for towns and small cities.

Rationale of Public Minibus Service for Towns

An important objective in personal mobility has to be the minimisation of both energy consumption and carbon footprint while maintaining a quality of service through a portfolio of means of transport. Public transport is one piece of the puzzle, in particular for urban mobility. The reason I would like to focus on towns and small cities as well as a minibus service is twofold.

Why focus on towns and small cities?

First, due to the COVID-19 pandemic, it appears that the convergence towards larger and larger cities may be questionable henceforth; urban agglomerations are perhaps more conducive to pandemics and, despite economic synergy effects, larger cities may become less desirable. As a consequence, a large portion of the population has now realised that they can indeed be working from home, and they can be trusted to be working from home without degradation of their performance. It turns out that my initial estimate of the percentage of working from home employees and its energy saving potential of around 10% was too conservative. There is more to be had.

That being said, people could live anywhere. Homes do not have to be isolated places in the country side, however, with the entire population uniformly distributed over all inhabitable space. I reckon, an optimum for a society where a large portion of the population works from home would be a collection of distributed towns and small cities with high speed links between them. This concept may be reminiscent of the Garden City movement, “initiated in 1898 by Ebenezer Howard in the United Kingdom and aims to capture the primary benefits of a countryside environment and a city environment while avoiding the disadvantages presented by both.” Perhaps a renaissance of this movement could be expected in the near future.

Why focus on a public minibus service?

As depicted in the Thoughts on the Energy (and Carbon) Saving Potential for Personal Mobility, every person could follow a decision sequence to determine how to realise a mobility need, if at all: 1) could the mobility requirement be avoided all together, e.g. working from home, partial home schooling, use delivery service, etc; 2) could walking be a solution; this would require what in German is called Nahversorgung, local services, e.g. corner shops, bakeries, butchers, green grocers, etc.; 3) could cycling fulfil the mobility need; this would require a coherent, contiguous and safe cycle network. Failing those, and trying to avoid using an individually owned car, the next best solution would be a public transport service, in particular buses.

What is wrong with the current bus service? Cambridge, for instance, as many European cities, has a medieval city centre; it does not make sense to use large double decker buses, all trying to run in convoluted ways through the city centre. These buses are too large and cause road damage, too (big pot holes at the stops), a nuisance for cyclists; they run infrequently and in an unpredictable manner causing a vicious circle in their usage. Last but not least, since many bus services are expected to be profitable, i.e. maximising shareholder value, their proposition is a contradiction in terms with public service. They are too expensive and hence not interesting for the inhabitants, contributing further to feedback cycle towards zero.

For all these reasons I propose a public bus service employing electric mini-buses (and drivers), and operating at an adaptive schedule; the system is owned by the city and is run as a service for its inhabitants. As such it goes beyond shared taxi services such Sheruts, in Israel. In addition, it is conceivable to link this service to a larger transport network.

Towards a Public Shuttle Service for Towns

Instead of using a fixed schedule of large buses I could imagine a network of smaller, electric minibuses akin to the ones proposed by Arrival, perhaps a bit smaller. They would be running in a dense, dynamic network, at an adaptive frequency which is fed by a real-time customer requests, information feeds from CCTV, traffic model predictions, etc. The result is something between taxi and fixed route bus service. As shared mobility solution it is inherently more energy efficient than relying on individually owned and used cars.

Traffic Flow and Mobility Needs Analysis

As a first step it would be necessary to analyse the mobility requirements for the town or city in question, establish the traffic flows as function of space and time. I could image using different sources, e.g. to name but a few:

  • survey of inhabitants, from where to where and when people want to go,
  • with people’s consent, mobile phone traces of interested participants,
  • publicly available traffic flow or CCTV analysis.

Based on various data sources, mobility models can be determined and a business potential estimation can be carried out. In essence it will be important to know how many minibuses will be needed where and when to accommodate the anticipated mobility needs.

Adaptive Schedule and Routes

A bus is not a taxi; it has more or less fixed routes with designated stops. Quite often this network is rather course in space and time which is one reason why people prefer individual forms of transport. Another reason might be the fact that there is always an uncertainty around any schedule (with all agents competing for the road); consequently, one has to take into account an allowance in time when taking a bus: better be at the stop a bit earlier in case the bus arrives and leaves before schedule. Once the frequency is high enough and the intervals become small, the uncertainty becomes irrelevant; there would always a bus within an acceptable amount of time, say 5 minutes.

Fortunately, the bus frequency does not need to be high at all times; both schedule and routes could be adapted to the needs in real-time. This should nowadays be possible. Given the mobility analysis and real-time data feeds about people’s movements, e.g. public events, as well as users making reservations, a fleet can be dynamically deployed to certain areas or times. Conversely, the bus does not need to be routed through areas where and when there are no clients. The location of all buses is always know and a schedule can be dynamically adapted to the current needs in real-time. Of course, a contingency of drivers and buses will need to remain on standby to make that work; the better the modelling the smaller this contingency needs to be.

User Interface and Experience

Most journeys are not spontaneous but known in advance. Everybody usually plans journeys ahead, knows about them in advance, or they are imposed by the schedule of others, employers, school, doctors, dentists, you name it. Hence in this system, a customer can and will lodge a request for a journey, as a single event, or even repetitive events, and expect that a small bus will be at the bus stop; the mean waiting time would be much smaller. The customer would receive real-time notification of the buses status, e.g. the expected arrival time and the estimated time to reach the bus stop on foot.

All that would require a mobile phone App associated with the service. NFC on phones could be used for check-in on shuttle, and getting off again. The same App be used for billing. It is even conceivable to sell flat fee tickets to people who do not have a smart phone and supply the phone as part of the service. That combined with attractive prices would result in a virtuous circle: uptake in usage, fewer cars, buses getting through leading to a more predictable service. For instance, the city of Vienna, Austria, charges 1 EURO per day to use the entire public transport network; most people take advantage of this kind of service. If Cambridge had such a service for £1/day, I am sure that most people would not hesitate to become subscribers.

The Correct Response to Mobility Needs

In an era where horses were mostly used for mobility, about 100 years ago, Henry Ford allegedly said: “if you ask people what they want, they would say, faster and more reliable horses”. Transposed into our era of individually owned vehicles, if you asked people what they wanted, they might say: “more reliable cars, more energy efficient ones, and perhaps fitted with some autonomous driving capability.” On both accounts people extrapolate from current technology and do not question the mobility requirements by making a step back: what problem are we trying to solve here?

The question or questions should be: why do we need such high levels of personal mobility?, What can we do to reduce them? For those we cannot, what means of transports are needed to deliver these needs at a certain quality of service and by minimising the ecological footprint? It would make more sense to first question the needs for mobility. How can society be organised to avoid personal mobility? Urban planning with a provision for local amenities at short distance that encourages walking and cycling would be a starting point. Public means of transport would be another piece of the mobility portfolio, including but not limited to trains, trams and minibus services.

One further note on the use of individually own cars. A common reason to take ones car to town is the transport of goods purchased in town. Using a bicycle or a bus is just not an option when it comes to lugging your purchases home. Here again it is important to question the solution (individual car ownership and journey) and think about alternatives, perhaps following the concept of browsing for goods in town reached by bicycle or bus, then order a pristine copy delivered to your home as explained in Reducing Resource Expenditure for the Procurement of Household Goods. No need to drive to town to get stuff that has been taken there by a van beforehand.

Towards autonomous services

Building on the adaptive minibus service, the next evolutionary step would logically be Level 4 autonomous services, starting from very, very constrained operating domains that have been adapted to mobility case, e.g. the Smart Urban Region Austria Alps Adriatic using self-driving electric vans by NavYa. Gradually these domains could be extended as technology matures on one hand. On the other, and in parallel to building human driven dense public transport services, urban planning can help to extend Level 4 autonomy by creating safe and dedicated paths for autonomous, shared means of transport. The urban landscape would change and evolve with the public transport network including autonomous means of transport such as minibuses.

Works on the technological foundations of autonomous vehicles at Five, UK. Interested in personal mobility, renewable energy and regenerative agriculture.