Deutsche Version

Flexible Rural Carsharing with Container-based Fleet Logistics

Initial situation: Vehicle sharing systems everywhere except where you need it

Carsharing is often mentioned as a key element of a transition of the transportation sector towards sustainability: It should facilitate multimodal mobility with the targeted use of shared motorized vehicles only on that trip sections, that can't be covered well by foot, by bicycle or by public transport. This would not only lead to less kilometres travelled by car, but also less energy consumption per average car-kilometre as a single person would not necessarily use a car designed for 5 passengers and 130 km/h for a short trip. Furthermore, there are less vehicles required and for the most trips less range too. Thus, less energy and less raw materials for the manufacturing of cars and their batteries is required.

In practice, vehicle sharing systems met these expectations only to a very limited extent: Very flexible sharing systems for cars, bicycles or electric scooters developed dynamically in big cities, where it was quite easy to get around without a car already before. Outside urban agglomerations, carsharing is offered rarely and only under the condition that the vehicle must be returned to that point where it has been taken. Therefore, for many trips it is impossible to cover only the "last mile" by carsharing unless the user pays an inadequately long rental period compared to a very short trip length.

An important reason for the lack of flexible carsharing in rural and suburban areas is the unsolved challenge of the redistribution of vehicles which is necessary to make them reliably available in the whole territory covered by the system. In central urban areas this problem is less significant as various origin-destination pairs with different trip purposes are rather compensating each other concerning the distribution of vehicles. If inequalities in distribution and availability of vehicles arise, public transport and taxis are always available. In rural and suburban areas the requirements are completely different: Carsharing must promise much higher reliability and massive use by commuters would lead to shortage of vehicles during the day in the outskirts and during the night in the centre of the respective region.

Two solution approaches can be combined in order to overcome the problem of vehicle redistribution:

  1. Efficient logistic solutions, making redistribution easier

  2. Reduction of the demand for vehicle relocation by optimized division of labour between carsharing, scheduled public transport and demand-responsive public transport

Efficient fleet logistics: transportation of light electric vehicles on roll-off containers

This concept is based on the Renault Twizy as an electric vehicle that is available on the market and can be parked very efficiently on small space: on a tailor-made special type of a roll-off container five vehicles can be parked perpendicularly. Other light electric vehicles potentially suitable for this concept could be the Microlino (only four vehicles per container, but better comfort and in case of perpendicular parking more practical entrance through the front of the vehicle) or weather-protected tilting three wheelers as the Toyota i-ROAD, Roo or Nimbus Halo (more than five vehicles per container possible, but either not yet or nor more on the market and more complicated to steer). The Roll-off container is designed symmetrically to its longitudinal axis. It is accessible from both sides so it is not necessary to distinguish between "left-hand" and "right-hand" containers in fleet logistics.

Similar to many bikesharing systems, vehicles can be taken and returned only at rental stations but the system offers the flexibility to return the vehicle at another station than that where it has been taken.

The rental station consists out of two or more pitches for one roll-off container each with sufficient manoeuvring space between them. The only fixed installation of the rental station are several charging poles for recharging the vehicles between the rides. Most light electric vehicles can be charged through domestic power sockets. Users are told by the rental system which of the vehicles they shall take and to which exact position of the destination station they shall return it. So it is guaranteed that a shortage of vehicles can always be relieved by replacing a completely empty by a completely full container and the other way round a shortage of free slots can always be fixed by replacing a completely full by a completel empty container. After exchange of empty and full containers a rental station shall never be completely empty or completely full, so it is not necessary to place all vehicles on containers. Therefore, next to the space for the containers, there are also some parking slots for rental vehicles directly on the road surface.

Before moving a roll-off container full of rental vehicles, the vehicles must be secured, e.g. with straps around the whole vehicle or with clamps, fixing the wheels of the vehicles to the container. Furthermore charging cables must be disconnected. Probably it will be more practicable to connect the cables permanently to the charging poles than to the vehicles. The containers are moved by hook lift trucks which are widely used in waste management. Synergies between waste management and relocation of sharing light electric vehicles could be achieved if both together is tendered by the same public entities.

If there is always space for one container left empty, five rental vehicles on one container can be relocated quite quickly and easily by first putting down one container and then picking up the other one. Using 7,5 m long containers it is also possible to use trucks with trailers, so up to ten vehicles can be relocated at once in compliance with the maximum total length of 18,5 m for a truck with a trailer. Despite the much more complicated procedure of loading and unloading roll-off containers to and from hooklift trucks with trailers it is quite common to use such trailers in waste management.

Optimized division of labour between carsharing, scheduled public transport and demand-responsive public transport

A comprehensive carsharing offer for rural and suburban areas is a valuable addition to public transport, but it can't replace public transport for the following reasons:

  1. Long distances should not be covered by carsharing vehicles in order to reduce demand of vehicles, battery capacity and recharging time.
  2. Despite the optimized redistribution of vehicles it would be inefficient to use carsharing for commuter flows that are bundled and clearly unbalanced concerning time and direction: Either there is still a high effort for vehicle redistribution or many vehicles are again used just twice per day.
  3. Despite the fact that some light electric vehicles can be used with an moped licence (category "AM" in some EU countries) there are still parts of the population that have mobility needs but are not allowed or not able to drive such a vehicle.

In order to achieve optimal rural and suburban mobility at reasonable costs for the public we need an optimized division of labour between the publicly accessible components means of transport:

  • Conventional, scheduled public transport is divided into two fields:
    • A network of backbone services, consisting out of railway lines suitable for passenger transportation and bus routes with high demand. These services operate from the early morning to the late evening with trains or buses every hour or half hour in both directions. These lines follow the principle of a network-integrated regular interval timetable (hub-node-structure with optimized interchanging times).
    • In peak hours, not only the frequency of trains or buses on these backbone routes is increased but there are also additional routes operated only during peak hours, connecting smaller villages to the scheduled public transport network.

  • In this way scheduled public transport relieves the carsharing system from a big part of the journeys in peak direction: in the morning from the periphery to the center, in the afternoon from the center to the periphery. At the same time the flexible carsharing option solves the unpopularity of typical communter buses caused by the fact that commuters have no possibility to get home if they want to stay longer in the city than usually. With the flexible carsharing (without the requirement to bring the vehicle back to the place where it has been taken) commuters can regularly use scheduled trains and buses, but if they return later from work they can use a shared light electric vehicle. The carsharing system offers comprehensive mobility over the whole territory and should be used mainly for those diverse journeys that do not lead in the morning into the cities and in the afternoon back, but tangentially between rural and suburban locations or in opposite direction to the commuter peaks. Beside unusual commuter routes (e.g. people working in a more peripheral location than they live in or working during the night) this includes the mostly underestimated mobility needs for leisure activities, business appointments, shopping, administration, medical care or care for children and elderly relatives.

  • Locations away from the backbone routes of public transport are made accessible for people without driving licence by demand-responsive public transport (paratransit). This service also covers the whole territory, but compared to the carsharing system it is less flexible as users have to book a ride via phone, web or app and have to accept some waiting time.

  • Complementary to the flexible carsharing system with light electric vehicles there should still be the possibility of carsharing with 5-seat automobiles or vans for weekly shopping or transportation of bulky items. Sharing of these vehicles can be organized more conventionally with the obligation to return the vehicle at the same point where it has been taken.

Considerations and rough estimations concerning usage and density of the rental station network

The expected usage of the system is given by various trips, differing from spatial, temporal and purpose perspectives: In the smallest villages covered by the system, there is few transport demand in total, but for lack of scheduled public transport it would also serve for common commuting from home to work and back. Therefore, at these most peripheral rental stations, vehicles often stay overnight and move in the morning to the next local or regional center and in the evening back. In contrary to individually owned cars, staying the whole workday at the workplace or a park & ride facility, the carsharing vehicles can be used in between for various other purposes: for atypical commuter trips leading not in the morning to the center and in the afternoon back to periphery, but also for leisure, shopping or business trips. For such trips, the carsharing system does not only connect smalles villages to local centres and public transport, it also serves for trips between local or regional centres mutually, in particular on tangential origin-destination pairs poorly covered by scheduled public transport.

Even if the individual vehicles leave the smallest stations (small grey dots) only about twice a day for a short trip, the trips between bigger villages or small towns lead to good utilization of the vehicles.

The smallest workable rental station has space for ten vehicles: five of them directly on the road and five of them on a roll-off container. Furthermore there must be space for one further roll-off container in order to avoid time-wasting container manoeuvring when replacing an empty by a full container or vice versa.

In order to minimize the risk of completely empty or full rental stations, the average occupancy of the parking slots should be 50%, this means five vehicles at the smallest possible rental station. If there shall be one arrival and one departure per average vehicle parked at the station, the required usage is 10 trips per day (departing and arriving trips counted separately). Inhabitants of peripheral counties in Austria made in 2013/14 an average of 2,7 trips per day out of their homes, roughly two of them by motorized means of transport. If there are 100 inhabitants in the catchment area of a rental station, they do typically 200 motorized trips per day. In order to achieve the desired rental station usage of 10 trips per day, about 5% of the total traffic of the inhabitants must take place by the use of the shared light electric vehicles. This usage could consist out of the following paradigmatic user groups:

  • 4 of the 100 inhabitants make the carsharing system to their main means of transport with an average use of one and a half trips per day. This is a solution for those parts of the population living without a car in the household, but able and allowed to drive a light electric vehicle.

  • further 10 of the 100 inhabitants use the carsharing system as a part of a multimodal mobility lifestyle and take a shared vehicle on average for two trips per week (e.g. once a week there and back). This is considerable in particular for households with less than one car per adult.

  • 10 of the 100 inhabitants are visited once per two weeks by somebody who uses shared vehicles for travelling there and back.
For estimating the territorial coverage feasible by such a flexible carsharing system, population grid data published in the StatAtlas of Statistik Austria can be used. As an approximation, a grid cell of 1x1 km can be regarded to as the maximum catchment area of a rental station, in this case the maximum distance from the most remote part of the cell to the rental station would be about 700m (half diagonal line of a square with 1 km side length). In reality, the exact distribution of population and an accordingly optimized positioning of the rental stations would of course lead to significantly shorter average access distance.
If there is a rental station placed in every 1x1 km grid cell inhabited by at least 100 people, the result as a network represented by the purple and orange squares in the following map:

In a much more pessimistic perspective with a minimum population of 500 people in the 1 km² catchment area of a rental station, the network would be reduced to the orange squares. In this case the contribution of the comprehensive carsharing system to the goal of a nationwide mobility guarantee without an own car would be significantly smaller: One third of the inhabitants of Austria would live outside the catchment area of a rental station, in case of minimum 100 inhabitants per rental station (orange+purple squares) it would be only one tenth. In the suburbs of some bigger cities, for tangential origin-destination pairs or in off-peak hours it would anyway be a significant improvement over existing scheduled public transport.

A frequent position in discussions about transportation is that for low income groups posession and maintenance of an own car would already today represent a critical financial burden and a significant increase of fuel price or the shift to more expensive battery electric cars would not be affordable for broad levels of the population, in particular if direct and indirect subsidies for battery electric vehicles would be cut. If this assumption is true, in the long term a flexible carsharing system could attract much more than 5% of the total transportation demand and comprehensive, nationwide operation should be definitely reconcilable with efficient vehicle utilization.
Regardless of the high long-term demand potential, the short-term potential is much lower: Most people in the region have solved their daily mobility needs in some way, mostly by sufficient car ownership. Although a flexible carsharing solution can support the decision to live with one instead of two cars in the household or even without a car, such a solution will not be made shortly after the implementation of the carsharing system. A reduction of level of individual motorization is realistic rather after some months or years, when the carsharing system is perceived as proven and reliable and the car needs an expensive repair or must be replaced by a new one. As time goes by, demand can also grow because people are chosing new places to live or to work because of good accessibility without an own car. The question of the minimum required population density for efficient operation concerns mainly a pioneer phase of a flexible carsharing system. In this phase it seems reasonable to offer it at a very low price in order to make the use of the shared vehicles financially competitive for car-owners too. Even if this leads in the starting phase to similar losses as poor usage at a higher price, there is the huge advantage of making as many people as possible familiar with the use of the system.