Free-floating Carsharing Systems Research Articles

A Space-Time Model for Demand in Free-Floating Carsharing Systems

A novel model approach is proposed to estimate the spatiotemporal distribution of demand for free-floating carsharing. The proposed model is based on a Poisson regression model for right-censored data and estimates possibly time-varying demand rates of small subareas of a service region based on booking data with spatiotemporal information on pickups and dropoffs of cars. The approach allows operators to gain insights into the spatiotemporal distribution of demand for their service and to estimate the loss of demand due to unavailability of cars. Moreover, it can also be used as an input to improve the design of the service, through relocation techniques or to analyze the service with macrosimulation models. In addition, the approach is applied to a case study with real data.

Area of Operation Planning for Free-Floating Car Sharing Systems

Area of Operation Planning for Free-Floating Car Sharing Systems

The benefit of complete trip information in free-floating carsharing systems

Free-floating and instant-access carsharing systems are two features of the most flexible carsharing systems. In the former, users are allowed to freely park the car in any legal parking spot within the boundaries defined by the service operator. In the latter, users are not requested to make any reservation in advance before picking up the car. This paper aims at evaluating the importance of complete trip information in free-floating instant-access carsharing systems. To this aim, we consider a system, referred to as the look-ahead system, where users can reserve a car in advance and are also allowed to directly pick up a car without any reservation. In both cases, the user is requested to specify complete trip details, including the estimated usage duration and the location where the car will be returned. Taking advantage of a complete knowledge of the trip details, the service operator can assign a reservation request to a car that is in use at the time of reservation, provided it will become available at the time and location the user will need it. In its nature, the operational setting we consider is dynamic, as trip information is revealed to the service operator at the time the user requests a car. We also investigate the possibility of suggesting to the user that makes a reservation a pickup location different from the desired one, provided it is not too distant from the latter. We compare the performance of the look-ahead system with the case no information is anticipated, which resembles the service provided by the main carsharing operators currently active in Milan (Italy). Additionally, we use, as a benchmark, the static case where all the information about the users requests (pickup and return locations, as well as delivery time) is known before the start of the planning horizon. We consider a system with no car relocations performed by ad-hoc operators. This enables us to measure the pure benefit of anticipated information, i.e., the benefit coming only from knowing in advance where and when the vehicles will be returned, purged from potential vehicles availability related to relocation operations. The matching between user requests and cars is obtained, for the look-ahead system, iteratively at fixed time intervals through the solution of a Binary Linear Program (BLP) and, for the benchmark case, through the solution of a single BLP. No optimization model is needed for the case where no information is anticipated. A simulation study, based on real-world data from the city of Milan, shows that the look-ahead system can satisfy a number of requests much greater than the case without anticipated information and close to the benchmark case. Moreover, we perform a sensitivity analysis on different parameters, including the maximum distance a user is available to walk and the minimum amount of anticipation requested to users for booking requests, showing their impact on the performance of the systems.

Dynamic pricing of free-floating carsharing networks with sensitivity to travellers’ attitudes towards risk

Free-floating carsharing (FFCS) systems are characterised by volatile fleet distribution as well as customers’ heterogeneous price sensitivity and spatiotemporal flexibility. There is thus an opportunity for operators to employ dynamic pricing to manage various aspects of fleet allocation: which customer is provided which vehicle, at what time and price, and the agreed pick-up and drop-off location. While there are emerging examples of dynamic pricing in FFCS, there is as yet no general framework for the interaction of consumer and operator behaviours in this context, most particularly consumer response to the inherent risks and uncertainties in the journey characteristics noted above. In this study, we propose a choice-based framework for modelling the supply/demand interaction, drawing on behavioural models of decision-making in risky choice contexts and empirical stated-choice data of user preferences in a dynamically priced FFCS market. In addition to the ‘spot market’ mechanism of dynamic pricing, the proposed framework is capable of evaluating operator strategies of allowing (at an agreed price) customers to make guaranteed advance reservations. We demonstrate that this approach allows the system operator to set an optimal pricing strategy regardless of whether user risk preferences are risk-seeking or risk-averse. We also demonstrate the applicability of the proposed framework when the operator seeks to maximise revenue (as with a private operator) vs social welfare (as with a public operator). In the case study which employs empirical user preferences, we show that users’ risk preferences have a relatively small impact on revenue, however the impacts are much larger if there is a mismatch between users’ actual risk preferences and the system operator’s assumptions regarding users’ risk preferences.

Integration of Free Floating Car Sharing Systems in Rail Stations: A Web Based Data Analysis

In the last decades, car sharing has been a tool for city planners to reduce private car traffic and pollution in big urban areas. The emergence of the ICTs (Information and Communication Technologies), together with the development of the collaborative economy, has allowed for the birth of the new Free-Floating Carsharing (FFCS): A more flexible type of carsharing, in which electric cars can be used. Little research has been devoted using real FFCS flows data, to the FFCS impacts on user behavior and even on the public transport system thus far. Furthermore, in big metropolitan areas, central rail stations should promote modal interchanges, including new modes of electric FFCS systems. The aim of this paper is to design a web-based platform to collect and analyze FFCS demand on the surrounding areas of rail stations and makes a proposal to provide these systems with electrical recharging energy obtained from the regenerative braking of high-speed trains. This case study includes Atocha and Chamartín Central Stations in Madrid (Spain). Scientific evidence shows a high demand of FFCS cars at central rail stations and a trip profile with a short time duration linked to the closest districts of rail stations.

Feature-based selection of carsharing relocation modes

One-way and free-floating carsharing systems must be rebalanced to achieve a high service level, and thus generate benefits for users and society. In practice, vehicles can be relocated with multiple different modes (e.g., by truck or by driving them), but a single mode is sufficient in many instances. Obviously, a single mode is preferred from a computational standpoint: The routing problems are less complex since less synchronization is necessary, and thus solve much faster. It remains an open question which features drive the decision on the best mode, and if operators can decide a priori whether hybridization of several modes is beneficial/necessary, and which modes one should hybridize among. We build a classifier based on linear regression which predicts the costs for all individual modes. The advantage of this approach is that cost estimates (i) can be used as a feature in other approaches, and (ii) allow operators to estimate the necessary budget upfront. However, cost estimates cannot be used directly to determine key drivers for modal choice. We, thus, use logistic regression and decision trees for determining the best mode. These approaches are better at determining relevant features that explain which mode is preferred in an instance. We find that the most important features to decide between modes are vehicle and truck costs per kilometer as well as their velocities, and the average number of vehicles that shall be relocated per day (that is, the imbalance of the system). In most instances, the decision is between driving vehicles to rebalance them and rebalance staff by biking, or loading vehicles onto a truck. Hybridization proves useful in ≈20% of all instances, and a simple rule-based classifier is able to predict correctly that hybridization is necessary in most instances.

A polynomial-time algorithm for user-based relocation in free-floating car sharing systems

Free-floating car sharing (FFCS) systems are a promising concept to reduce the traffic volume in cities. However, spatial and temporal mismatches of supply and demand require a relocation of rental cars in order to avoid low degrees of utilization. Here, especially user-based relocation strategies seem to be promising to increase utilization in a cost-efficient manner. However, a thorough optimization-based assessment of user-based relocation strategies for FFCS systems is still missing.In this paper, we introduce an integer program that optimizes the assignment of user-based relocation strategies in FFCS fleets. We develop a graph representation that allows to reformulate the problem as a k-disjoint shortest paths problem and propose an exact algorithm to solve large-size instances. We show that this algorithm can solve real-world instances within a few milliseconds as well as instances with up to 100,000 customers and 10,000 vehicles in a few minutes.Furthermore, we present a case study based on real-world data and derive managerial insights on user-based relocation strategies. Our results reveal an upper bound on the benefit of user-based relocation strategies and demonstrate that the employment of such strategies can increase the number of fulfilled rental requests by 21%, while increasing the operator’s revenue by 10%.

Does free-floating carsharing reduce private vehicle ownership? The case of SHARE NOW in European cities.

During the last decade, the use of free-floating carsharing systems has grown rapidly in urban areas. However, little is known on the effects free-floating carsharing offerings have on car ownership in general. Also the main drivers why free-floating users sell their cars are still rarely analysed. To shed some light on these issues, we carried out an online survey among free-floating carsharing users in 11 European cities and based our analysis on a sample of more than 10,000 survey participants. Our results show that one carsharing car replaces several private cars - in optimistic scenarios up to 20 cars. In Copenhagen (followed by Rome, Hamburg, and London) one carsharing car replaces about two times more private cars than in Madrid, the city with the lowest number. The main non-city specific influencing factor of shedding a private car due to the availability of the free-floating carsharing services seems to be the usage frequency of the service. The more kilometres users drive with these cars, the more likely it becomes that they sell a private car (or they sell their car and, therefore, use this service more often). Further memberships of bikesharing and other carsharing services, users that live in larger buildings as well as users that own several cars are more likely to reduce their number of cars, too. Finally, our findings are highly valuable for carsharing operators and (transport) policy makers when introducing free-floating carsharing systems in further cities. According to our results, all 11 cities show a reduced private car fleet due to members' access to free-floating carsharing.

Predicting Car Availability in Free Floating Car Sharing Systems: Leveraging Machine Learning in Challenging Contexts

Free-Floating Car Sharing (FFCS) services are currently available in tens of cities and countries spread all over the worlds. Depending on citizens’ habits, service policies, and road conditions, car usage profiles are rather variable and often hardly predictable. Even within the same city, different usage trends emerge in different districts and in various time slots and weekdays. Therefore, modeling car availability in FFCS systems is particularly challenging. For these reasons, the research community has started to investigate the applicability of Machine Learning models to analyze FFCS usage data. This paper addresses the problem of predicting the short-term level of availability of the FFCS service in the short term. Specifically, it investigates the applicability of Machine Learning models to forecast the number of available car within a restricted urban area. It seeks the spatial and temporal contexts in which nonlinear ML models, trained on past usage data, are necessary to accurately predict car availability. Leveraging ML has shown to be particularly effective while considering highly dynamic urban contexts, where FFCS service usage is likely to suddenly and unexpectedly change. To tailor predictive models to the real FFCS data, we study also the influence of ML algorithm, prediction horizon, and characteristics of the neighborhood of the target area. The empirical outcomes allow us to provide system managers with practical guidelines to setup and tune ML models.

Optimizing Relocation Cost in Free-Floating Car-Sharing Systems

This paper addresses the problem of operator-based vehicle repositioning in free-floating electric car-sharing systems (FFCSs). New algorithmic techniques are developed which derive cost-effective relocation tours by primarily minimizing relocation drivers’ walking time. Car repositioning takes place overnight and aims at complete rebalancing of the system i.e., at achieving an optimal, based on user demand, cars distribution among non-overlapping cells of the FFCS’s operating area. The battery level of the electric cars is also taken into account when deciding if and where each car will be relocated. If $k$ is the number of employees handling car relocations from oversupplied to undersupplied regions, when $k=1$ i.e., there exists a single employee carrying out a relocation tour comprising alternating phases of a car relocation followed by walking to the next to be relocated car, we define and solve the Minimum Walking Car Repositioning Problem (MWCRP) whose main objective is to minimize the walking distance in the relocation tour as the walking part of such a tour is tiresome and takes more time than the driving one. We also define and solve the $k$ -MWCRP when $k>1$ to handle the case that a team of more than one drivers is employed to take on the relocation plan. It is worth mentioning that these are the first vehicle relocation techniques lowering the relocation cost by mostly minimizing the walking cost. Performance results confirm the effectiveness of the proposed algorithms.

Electric Free-Floating Carsharing for Sustainable Cities: Characterization of Frequent Trip Profiles Using Acquired Rental Data

Free-floating carsharing systems (FFCS) have become a new type of urban sustainable mobility, much more flexible than the previous station-based carsharing but limited by on-street parking availability and managed by municipal administrations. Literature on FFCS until now mostly relies on survey-based methodologies and simulations, and little research on FFCS has been devoted to the scientific analysis of real flows using revealed web-based data. This paper contributes to the existing literature with an analysis of FFCS trips using rental data collected directly from operators’ websites, paying special attention to the most frequent trips. The added value of this research is that it provides the first analysis of the more FFCS demanding districts in the city of Madrid. The results showed that the main origin and destinations points were concentrated in low populated and high-income districts that also had good parking availability and connectivity to the public transportation network.

Understanding the spatial distribution of free-floating carsharing in cities: Analysis of the new Madrid experience through a web-based platform

In recent years the sharing economy has become established in different modes in the urban transport system, and claims to be reducing the number of cars and contributing to lower traffic pollution. Free-floating carsharing (FFCS) is a new and more flexible type of carsharing that is driving the growth of carsharing markets around the world. While there is a very extensive literature on traditional carsharing, more research needs to be done on the new FFCS trip profile in order to estimate its spillover effects on the urban transportation system. As FFCS systems are based on ICTs, new web-based methodologies (instead of traditional surveys) are the best approach to analyse them.This paper contributes to the existing literature with a spatial evaluation of the FFCS trip profile, obtaining a temporal distribution of the main flows throughout the FFCS service area. The added value of this research is that it provides the first spatial analysis of a FFCS system in Spain using rental data collected from the operators' websites. The results clearly show the prevalence of the short-distance FFCS trip that is faster than available public transport and whose origin and destination are closely dependent on parking availability.

Traveller preferences for free-floating carsharing vehicle allocation mechanisms

Free-floating carsharing (FFCS) fleets are inherently volatile spatio-temporally, which presents both a logistical challenge for operators and a service reliability issue for customers. In this study we present a stated-choice survey to investigate the attractiveness to customers of two mechanisms for managing fleet volatility: Virtual Queuing (VQ) and Guaranteed Advance Reservation (GAR). We investigate socio-demographic features and “Big Five” personality traits that are associated ceteris paribus with choosing to use the existing FFCS service model, willingness-to-pay (WTP) for VQ and GAR, and risky-choice behaviour under the uncertainty of FFCS systems. Data (n = 289; 232 employed in analysis) are sourced from existing users of a FFCS service in London, UK. Within the survey context, we found that customers are on average not willing to pay for VQ (i.e. negative WTP), however have £0.54 per journey WTP for GAR, with low-frequency FFCS users and users scoring highly on the Big Five “Conscientiousness” dimension having larger WTP for GAR. When analysing the two dimensions of uncertainty, we found that respondents exhibit risk-seeking behaviour towards price and weaker and insignificant risk-aversion towards walking time. This pattern holds across the three standard model types of nonlinear risky-choice behaviour that we investigated. The results are intended to be useful both to policymakers and carsharing operators who are likely, as the industry matures, to seek mechanisms to differentiate their service offers to better serve individual market segments with distinctive characteristics.

Long-term vehicle reservations in one-way free-floating carsharing systems: A variable quality of service model

Reservations in daily services can improve user satisfaction, and give additional information about the demand patterns to the operators. However, providing reservations to carsharing clients is difficult. While carsharing is especially convenient if it is allowing one-way trips and vehicle drop-off anywhere in the service area (called free-floating), this flexibility increases management complexity because of vehicle stock imbalance. Most of the commercial providers of free-floating carsharing offer reservations under highly restrictive terms, for example only up to 30 min in advance. In this paper, we propose an innovative reservation enforcement technique that allows substantially longer reservation times while keeping the system profitable and achieving high service quality. A simple way to enforce reservations is locking vehicles until the departure time of a client. However, it comes at the cost of idling vehicles that could be used by other users and decreasing the revenue. Our approach, called relocations-based reservation enforcement method (R-BR) combines vehicle locking and relocation movements. It locks vehicles only a short time before the trip departure if a suitable vehicle is close enough due to the natural trip patterns. If no such vehicle is available, a car is relocated from another place. Further, we propose a variable quality of service (QoS) model in which the guaranteed radius around the user within which the reserved vehicle will be placed, and the maximum allowed reservation time before the departure depends on the zone of trip departure. A simulation-based optimization is used whereby the carsharing operation is simulated and optimized using an iterated local search (ILS) metaheuristic for adjustment of service level parameters. The proposed technique is tested on a set of artificial problem examples and a case study of a simulated working day in the Lisbon Municipality, Portugal. Results show that the proposed R-BR method is substantially better than the simple vehicle locking when the constant QoS approach is used and that the devised ILS metaheuristic can further increase the system performance, especially with high trip volumes.

Student behaviour towards Free-Floating Carsharing: First evidences of the experience in Madrid

Abstract Mobility in large cities is changing. Congestion and space problems and the growing concern for the environment make the citizen begin to abandon the private vehicle and look for alternatives of transportation. More flexible carsharing systems, such as Free-Floating Carsharing (FFCS), have broken into the markets of European and US cities, experiencing a surprising growth in demand. Smartphone applications enable the user to search and book a car close to his geographical position within a service area defined by each FFCS company. Users can take and leave the cars freely in public parking spaces inside the service area. Flexibility is, therefore, the main attraction and the most revolutionary feature of FFCS systems. While there is an extensive literature about the impacts of traditional station-based carsharing systems on public transportation and sustainability, little work has been done on FFCS systems. This paper contributes to the limited existing literature with an analysis of the behaviour of university students towards FFCS systems, taking Madrid (Ciudad Universitaria Campus) as case study. University students are a likely user population group, so their mobility patterns and transportation preferences have been studied. The methodology has been based on two survey campaigns conducted with a one-year time slot (2016: 186 valid questionnaires; 2017: 213 valid questionnaires). The results clearly show how the quality attributes of FFCS depend strongly on the maximum distance and time accepted by each user to walk and wait in the search of an available car.

An Explanatory Model Approach for the Spatial Distribution of Free-Floating Carsharing Bookings: A Case-Study of German Cities

When the first free-floating carsharing operators launched their business, they did not know if it would be profitable. They often started in highly populated cities without performing extensive target group analysis, and were less concerned about fleet management. Usually, there are two main datasets that can be used to find areas that would have a high demand for free-floating carsharing: booking data, for measuring the actual demand; and land use and census data for describing the activities performed in different areas in a city. In this paper, we aim to use this information to help predict the demand of free-floating carsharing systems. We use booking data provided by DriveNow for Berlin in 2014 and contextual information about the type of activity each neighborhood has. Using Berlin as a case study, we apply a negative binomial statistical model to explain the number of bookings. From the results, we conclude that free-floating carsharing is predominantly successful in areas with more affluent citizens who are open to trying new and sustainable technologies. Other important determinants that result in a high number of carsharing bookings are the area’s centrality and parking lot availability. The statistical model for Berlin was then transferred to Munich and Cologne, two other cities in Germany with similar population sizes. A comparison between the estimated demand categories and actual bookings shows satisfying results, but also non-negligible local conditions influencing the spatial demand for bookings.

Booking Behavior of Free-Floating Carsharing Users

In recent years, the services of free-floating carsharing (FFCS) systems have gained popularity, especially in urban areas. Some FFCS operators are now including battery electric vehicles (BEVs) in their fleets. This addition is especially beneficial for cities and their inhabitants, as BEVs are locally emission free and move more quietly. However, how FFCS customers react to the new technology has not yet been empirically analyzed. By combining the app call and booking data of an FFCS operator in Munich, Germany, this study assessed the preferences of customers when booking a specific car and evaluated the attractiveness of the fleet’s BEVs. With the consideration of the spatial availability encountered when the mobile phone app was opening, the study found that the distance to the nearest available car was the most important decision criterion for customers. If a car had been available within 200 m, the probability that customers booked a car would have been twice as great compared with the case that the nearest vehicle had been 500 m away (100 m ≈ 0.06 mi). In addition, the majority of the customers booked the nearest available car. Nevertheless, certain qualities of vehicles made users walk to a car that was farther away. The study identified popular vehicle models and found that customers preferred to use the fleet’s BEVs, independent of the battery level. BEVs were used for almost the same distances as internal combustion engine vehicles and were suitable for most use cases: more than 80% of the FFCS trips covered a distance of at most 20 km (≈12.4 mi).

Integrated Relocation Model for Free-Floating Carsharing Systems

This paper presents an integrated relocation model for free-floating car-sharing (FFCS) systems. A historical data analysis conducted monthly generates demand indicators. A zone categorization identifies zones with a historical vehicle shortage or surplus and enables the estimation of the optimal vehicle distribution for the target time period. Given the current vehicle positions, the model identifies vehicle imbalances. Five macroscopic and microscopic steps using optimization and rule-based methods recommend relocations and service trips on an individual vehicle level. This process results in detailed staff operation plans. To assess the impact of relocations on the system's operation, the model was applied to an FFCS system in Munich, Germany. Three real-world field tests conducted at stages of the model's development exhibited different degrees of automation. The evaluation showed promising results. All tests had positive impacts on the main measures of success. The earnings per vehicle were increased by up to 18%. The mean idle time per trip end of all vehicles in the feet was decreased by up to 18%. The difference in idle times between the relocated vehicles and neighboring vehicles that were not relocated was decreased by up to 31%. Most important, the profit of the operator was between 4.7% and 5.8% higher than without relocations. The automation of the whole model in the last field test led to slightly lower but still very positive impacts while facilitating the whole relocation process for the operator.

Enabling sustainable transportation: The model-based determination of business/operating areas of free-floating carsharing systems

Free-floating carsharing systems with (and without) electric vehicles are emerging and there is limited knowledge where such concepts will succeed, regardless of their environmental benefits. Within this knowledge gap, the definition of the business/operating area is a key aspect. Inside this area a carsharing vehicle can be picked-up and dropped-off everywhere at any time. This research develops a model that helps to determine business areas a-priori by predicting inner-city booking hot-spots. The approach is based on modeling success factors as independent variables, namely population density, housing rent, city center distance, and hotel/restaurant density. With the help of these variables predictions for geographic zones in cities are made that prove to be helpful for designing operating areas for free-floating carsharing systems – a key challenge for providers' success when approaching new cities.

A practice-ready relocation model for free-floating carsharing systems with electric vehicles – Mesoscopic approach and field trial results

This paper introduces a relocation model for free-floating Carsharing (FFCS) systems with conventional and electric vehicles (EVs). In case of imbalances caused by one-way trips, the approach recommends profit maximizing vehicle relocations. Unlike existing approaches, two types of relocations are distinguished: inter zone relocations moving vehicles between defined macroscopic zones of the operating area and intra zone relocations moving vehicles within such zones. Relocations are combined with the unplugging and recharging of EVs and the refueling of conventional vehicles. In addition, remaining pure service trips are suggested. A historical data analysis and zone categorization module enables the calculation of target vehicle distributions. Unlike existing approaches, macroscopic optimization steps are supplemented by microscopic rule-based steps. This enables relocation recommendations on the individual vehicle level with the exact GPS coordinates of the relocation end positions. The approach is practice-ready with low computational times even for large-scale scenarios.To assess the impact of relocations on the system’s operation, the model is applied to a FFCS system in Munich, Germany within three real world field tests. Test three shows the highest degree of automation and represents the final version of the model. Its evaluation shows very promising results. Most importantly, the profit is increased by 5.8% and the sales per vehicle by up to 10%. The mean idle time per trip end is decreased by 4%.