ArticlePDF Available

Abstract and Figures

The growth of car-sharing services as a new and more sustainable way of transportation is shifting the private mobility from ownership to service use. Despite the emerging importance of this type of mobility and the large number of papers present in the scientific literature, to the best of our knowledge no extensive and structured analysis has been performed to classify the research and determine the mainstreams. Aim of this study is to introduce a taxonomy and analyze the different aspects of car-sharing, including the different car-sharing services and the research questions considered in the papers. We analyze and classify 137 papers, covering the last fifteen years of research and deriving an insight of the mainstreams. Finally, we deeply study the trends and research perspectives of the literature, showing the unbalancing between the literature related to the operational level and the economic, business development and customer validation aspects.
Content may be subject to copyright.
Car-sharing services: an annotated review
Francesco Ferreroc, Guido Perbolib,d, Mariangela Rosanob, Andrea Vescoa
aIstituto Superiore Mario Boella, Turin, Italy
bICT for City Logistics and Enterprises - Politecnico di Torino, Turin, Italy
cLuxembourg Institute of Science and Technology, Luxemburg
dCIRRELT, Montreal, Canada
The growth of car-sharing services as a new and more sustainable way of transportation is shifting the
private mobility from ownership to service use. Despite the emerging importance of this type of mobility
and the large number of papers present in the scientific literature, to the best of our knowledge no extensive
and structured analysis has been performed to classify the research and determine the mainstreams. Aim
of this study is to introduce a taxonomy and analyze the different aspects of car-sharing, including the
different car-sharing services and the research questions considered in the papers. We analyze and classify
137 papers, covering the last fifteen years of research and deriving an insight of the mainstreams. Finally,
we deeply study the trends and research perspectives of the literature, showing the unbalancing between
the literature related to the operational level and the economic, business development and customer
validation aspects.
Keywords: Car-sharing, Taxonomy, Optimization, Business models.
1. Introduction
In the last years the growth of car-sharing services as a new and more sustainable way of transportation
is shifting the private mobility from ownership to service use. The basic idea of car-sharing is quite
simple: share the usage of a vehicle fleet by members for trip making on a per trip basis. Although
the first shared used vehicles system can be traced back to 1948 in the city of Zurich (Switzerland),
motivated by economic reasons, in the following years other attempts of public car-sharing systems were
not successful. In the 1980s, several successful car-sharing programs were started, with a consolidation
in the early 1990s, thanks to a new awareness of the citizens and a real burst due to a more pervasive
diffusion of ICT and mobile services in 2000s. Car-sharing increases mobility for community members
to reach destinations otherwise inaccessible by public transit, walking or biking, while increasing the
citizens’ awareness about the social and environmental impact of using private cars. It encourages and
supports multi-modal communities by providing an additional transportation option. From the point of
view of building a sustainable city, the vehicles used in car-sharing are typically fuel efficient and lead to
positive effects in reduction of urban emissions and city congestion [111].
Recently, also car producers started to enter directly in the market, as Daimler, BMW and FCA
group, are directly involved in car-sharing operations with the scope of finding new channels to market
the produced cars [9, 147, 144, 146, 145]. Presently, some large companies start to exist Worldwide, as
Zipcar with over 900,000 members and 11,000 vehicles [167] and Car2Go with 2,000,000 members and
14,000 cars in several countries, including China [162, 68]. So, the market is growing fast and with this
increasing demand also the demand of better understanding and control of the system increases. In fact,
car-sharing is not just a matter of business or fleet optimization, but creates a complex system made by
different actors, including citizens, public authorities and municipalities, companies. The system becomes
complex for the strong links between the actors, as well as for the implications on the governance of a
city when a large car-sharing service is introduced, as the integration with the existing public transport
network and the policies for letting different companies to compete in the same urban area.
Despite the emerging importance of this type of mobility and the large number of papers present
in the scientific literature, to the best of our knowledge no extensive and structured analysis has been
Preprint submitted to October 23, 2017
performed to classify the whole research field and determine its mainstreams. In fact, partial visions and
state of the art reviews of car-sharing exist, but there is a lack in terms of global vision. Existing works
can be mainly split in two groups: reviews considering the technical and modeling aspects [80, 95] and
papers dealing with the business perspectives of car-sharing obtained by surveys [145, 143]. Thus, just
a little has been done to give a holistic vision of the topic and to classify in a more general way the
Aim of this study is to fulfill this gap, presenting a taxonomy considering all the multi-facet aspects
of car-sharing. Our taxonomy provides a framework for classifying papers published in the various
academic disciplines and it becomes a guide for the researcher who chooses to study car-sharing with
an interdisciplinary view, providing numerical analysis and a way for characterizing the papers in the
literature. The educator can use this taxonomy to introduce the subject in a comprehensive way, letting
a newcomer to have a depiction of the wide spectrum of possible research lines. Finally, institutional
managers and stakeholders can use it in developing strategies for sharing mobility.
More in detail, in this paper we want to answer to the following questions:
Can we find a series of keywords/axes such that we can categorize any paper dealing with car-sharing
Can we organize the papers in a taxonomy and, by means of that taxonomy, highlight the research
mainstreams and future directions?
Can we see any lack in current frameworks in the literature, in terms of global view of the car-sharing
service, current trends and future paths?
To answer to our research questions, we structured this paper as follows. We first introduce a tax-
onomy for papers dealing with car-sharing services, considering in it the different aspects, including the
specifications of the different car-sharing services and the research questions considered in the papers
(Section 2). Second, we analyze and classify 137 papers appeared in the last fifteen years, giving an
insight of the mainstreams (Section 3) and making our study based on the largest database of works from
the literature. Third, we deeply study the trends and research perspectives of the literature (Section
4), with a specific focus on four issues related to car-sharing services: the analysis of the user behaviors
(Section 4.1), the forecast of the service demand (Section 4.2), the use of optimization tools for the design
and the management of the service (Section 4.3) and the business development and its effect in driving
the research (4.4). We demonstrate that there is an imbalance between the literature related to the
operational level and the user behaviors and business development.
2. Search methodology
To build the taxonomy we followed the three-steps method described by Bailey [5, 4]. We begin with
the empirical analysis of a database of papers. Then, the second stage clusters the information obtained
by the first stage, while, the third stage envisions in a mental concept of the cluster by generating a name
or label for the cluster. We first looked to the best practices defined by other taxonomies in the literature
[12, 126, 58]. To select the papers, we used scientific refereed journals and refereed conference proceedings
as a source for the car-sharing literature. To retrieve the papers, we used the Scopus database because,
concerning the transportation field, this bibliographic database contains articles from all major journals
dealing with transportation and management. Furthermore, the largest part of these journals are also
listed in the Excellence in Research for Australia (ERA) 2012 Journal List by the Australian Research
Council [52], a known list of scientific journals.
We restricted our search to the papers dated from 2001 until the end of 2016 to consider the most recent
papers (last extraction of the database, March 2017). After a screening consisted of an in-depth analysis
of abstracts, main topics, and results of papers, we selected 137 of these. This additional phase was
needed to remove, for example, the papers that considered specific aspects as the technologies for the
charging stations of electric vehicles, powertrains optimization and other technical features used in the
car-sharing services, but not pertinent with a topology of car-sharing services. We also decided to not
consider peer-to-peer papers, because we want to focus on the core car-sharing business, where the fleet
is owned by the car-sharing company. Moreover, we focused our analysis on car-sharing services based
on urban areas because they represent the most settled business model and they cover the mostly part
of the literature.
2.1. Taxonomy dimensions
We organized our taxonomy according to a double level classification. This choice is due to our
willingness to limit the complexity of the overall classification. Due to the large number of factors
affecting the car-sharing, our taxonomy does not meant to be jointly exhaustive and mutually exclusive
in its axes and subcategories. For example, a methodology can be applied to different problems and
specifications. However, our analysis intended to give an overview of literature patterns and trends in
this field. In fact it is known that classifications too much refined becomes difficult to use and to maintain,
preventing a practical usage of them [133]. Thus, we organized the information by axes, splitting each
axis in categories when needed (see Figure 1).
The taxonomy describes articles according to five main axes, which constitute the first level of our
taxonomy: Mode,Engine,Optimization objective,Time Horizon and Methodologies. The first two axes
concern the car-sharing service specifications, while the remaining three inform on the research problem
and the role of the research in the paper under consideration. Then, each axis is organized in a set of
categories (second level of the taxonomy). This section presents each axis in turn, briefly describing its
object and scope.
Research problem
Free floating
Not Applicable
Service specifications
Fully thermic
Two-way (station based)
One-way (station based)
Stochastic optimization
Combinatorial optimization
Operational and real time
Time horizon
Statistical analysis
Figure 1: Resume of the taxonomy axes
2.1.1. Mode
Mode identifies the different ways in which a car-sharing service can be provided.
Two-way (station based): in the Two-way [120] mode the available cars are parked in pick-up
stations, which are defined parking lots by the service provider or local administration and the
journey must start and finish in the same space. Thus, this operational model does not consider
the intermediate parking, which are the stops that the customer may plan for personal needs. The
set of parking lots is predefined.
One-way (station based): the One-way [120] mode is similar to the previous one, but in the
one-way case the parking lot in which the journey finish can be different from the parking lot in
which it started. The set of parking lots is predefined.
Free-floating: the Free-floating [60] mode is the last one came to the market; the cars are freely
parked in public spaces within the operational area (i.e., the area served by the car-sharing com-
pany), and the journey can start and finish in any point in this area.
Not Applicable: This section considers the papers analyzing car-sharing services without reference
to any specific mode.
2.1.2. Engine
This dimension is used to classify the engine type of the cars involved in the car-sharing service.
Fully thermic: in this section are contained papers that analyze fleets with fully thermic engines.
Thus, these fleets are composed by vehicles powered by traditional (i.e., fossil derived) fuels such
as gasoline or diesel.
Green: in this section are contained papers that analyze fleets of green vehicles adopted by car-
sharing companies, environmentally aware. In particular, these vehicles has less-polluting engines,
as electrical, hybrid, plug-in, natural gas, liquefied petroleum gas (LPG), as well as hybrid.
2.1.3. Optimization objectives
The Optimization objectives dimension classifies the analyzed papers according to the component of
the car-sharing service (i.e., technical management of the physical assets as infrastructure and fleets as
well as the business management) that is subjected to optimization. This dimension does not consider
the specific objective function in the operating model.
Business and service: papers in this category deals with the identification of service features. In
details, we consider here all the papers dealing with the business models and the definition of the
car-sharing service, including the identification of the user behaviors and the demand estimation.
Infrastructure: papers analyzing the optimal design and location of the car-sharing facilities , as
the parking stations and, for electrical vehicles, the charging stations.
Fleet management: papers analyzing the operations for the management of the fleet, as deter-
mining the fleet size and defining the car relocation strategies.
2.1.4. Time horizon
This axis considers the interval of time for which the decisions remain valid.
Design (strategic): strategic decisions that players (i.e., car-sharing company management) must
keep in account in the designing of the service, including fleet type definition, user behavior, pricing
policies, market place (green fields and urban areas) and demand identification.
Planning (tactical): this category considers all the papers dealing with service planning deci-
sions, including fleet size definition, location of facilities (parking stations, e-charging stations, car
maintenance facilities), urban areas boundaries, management of uncertainty of local demand.
Operational and real time: operational and day-by-day decisions related to the service provided:
operative car maintenance, refueling, car washing, relocation strategies to balance the system,
avoiding stations with an excess of vehicles, or empty stations.
2.1.5. Methodologies
The Methodologies axis groups the papers according to the scientific approach used by the authors.
Simulation:simulation aims to imitate the operation of the real-world processes. These method-
ologies are adopted to estimate the demand of the service to help the decision making process for
new operators. Simulations are mainly based on real data referring the transport behavior of the
dwellers of a specific area.
Combinatorial optimization:combinatorial optimization consists in modeling, analyzing and
solving a decision problem finding the optimal objective function according to a set of constraints
when the data involved in the problem under study can be considered as deterministic. It is used
in car-sharing services for the management of the fleet (e.g., relocation of cars) as well as for the
design and the planning of the service.
Stochastic optimization:stochastic optimization methods are used to solve decision problems
where the data are affected by uncertainty. In the car-sharing services case, uncertainty often affects
the service demand or the flows of vehicles between different parking stations or within the service
Statistical analysis:statistical analysis methods are mainly adopted for analyzing data sets
deriving from real observation (such as data sets provided by operators) or from surveys and focus
groups; these methods are used to analyze the state of the art in context where at least one operator
is working.
3. Literature analysis
In the following, we apply our taxonomy to the 137 selected papers and we use it to derive a first
numerical analysis by grouping these papers in two categories: Journals (94 papers) and Proceedings (43
Although the plethora of journals that cover topics on car-sharing services is quite broad, it is in-
teresting to note that more representative ones are Transportation Research Record and Transportation
Research Part A-F. Indeed, they contain 46% of papers collected in journals and 31% of the total. More-
over, a large part of literature derives from proceedings (e.g. IEEE international conferences, Procedia
- Social and Behavioral Sciences and Transportation research Procedia). 31% of papers refer to this
category and they are usually papers describing real projects and applications of car-sharing services.
Thus, the proceedings remain the main sources to have a better insight of the real-world projects, while
journals mainly present the more theoretical and general results. In contrast, 28 journals are present
with one and four with two works only. This unbalance is due, in our opinion, to the multidisciplinarity
of the research studies in car-sharing. For more details, the interested reader can refer to [57].
3.1. Mode analysis
The results of the mode analysis are shown in Figure 2. Almost 50% of the analyzed papers are
referred to one way mode, 19% are referred to free floating mode and 19% are referred to two way mode,
while 15% of the papers have not been classified according to this dimension. Furthermore, in recent
years there has been a growing interest in the electric car-sharing. In particular, one can notice how the
research is moving from the infrastructure design of the first papers to a majority of the papers dealing
with the customer behaviors and the customer acceptance. This shift in terms of papers’ topic is mainly
due to the change of maturity level of the electric car-sharing solutions, which are presently considered
as the most promising business line of the next decade [159, 62].
Recently, [79] proposed a hybrid car-sharing system, which combines two-way and one-way modes.
The authors applied it to the case study of Boston and showed the benefits and the profitability of
the system. The improvement of existing station-based mode with free floating mode has been recently
studied in [35]. Due to the limited usage of the hybrid systems, they are not considered as a separate
Figure 3 shows how the focus of the papers have changed during the last years. The chart shows
an increasing of the interest in car-sharing in the recent years, since 79% of the selected papers were
published from 2011 to 2016. The first result is that the one way mode was the first to be analyzed by
the selected papers, while free floating mode was analyzed from 2011 with a peak in 2014. Since free
floating mode has been studied to meet the growing demand for flexibility from the users of the service,
the development of this service was made possible thanks to new IT infrastructures and services such
as car positioning in the service area, mapping of the available cars and applications for the final users.
Thus, this type of car-sharing mode started to be considered by researchers and practitioners only when
the related IT technologies reached a sufficient maturity level.
3.2. Engine analysis
56% of the analyzed papers are related to fully thermic engines. Further analysis (Figure 4) show
that the interest in green engines (mainly in electric or hybrid vehicles) increased in the recent years.
The chart shows also an increase in recent years of the number of publications related to fully thermic
engines, due to the increased development of free floating car-sharing mode.
According to chart in Figure 5, fully thermic and green engines have been studied mainly referred to
one way mode, while free floating mode with green engines (25% of overall green engines publications)
are mainly referred to hybrid vehicles. Accordingly to the increasing awareness of public stakeholders
to low-emission and eco-friendly solutions, there are more publications related to green engines than
publications referred to fully thermic ones in the one way and free floating modes, which are also the
most recently introduced modes.
3.3. Optimization objectives analysis
According to Figure 6, 60% of the papers are related to business and service optimization, 28% are
related to fleet management optimization and 12% are related to infrastructure optimization. The chart
in Figure 7 shows an increased interest in business and service optimization in the last years, with 45% of
the papers published from 2011, and 22% of these were published in 2013 and 2014. Infrastructure and
fleet management optimization were investigated in the recent years because of their importance for green
car-sharing services (infrastructure) and free floating car-sharing services (fleet management). Business
and service optimization has been analyzed for all the different service modes, but predominantly for
one-way mode, with 51% of the papers analyzing business (see Figure 8). In this case, the analysis is
referred to behavioral and demand estimation issues. Furthermore, infrastructure and fleet management
optimization is mainly referred to the one way car-sharing mode. In fact, for this mode is very important
to plan the correct location of the stations (in particular for electric/hybrid vehicles) and the relocation
and maintenance strategies to balance the number of cars in the different stations and cope the users’
needs. Finally, Figure 9 presents the relation between optimization objectives and engine type. Business
and service optimization is mainly referred to fully thermic engines, while infrastructure optimization
is an important aspect for the green vehicles, with the focus on the optimal location for the charging
3.4. Time horizon analysis
59% of the selected papers are related to the design of a car-sharing service, while planning and
operational and real time analysis represent respectively the 16% and the 25% of papers selected (see
Figure 10). Analyzing the trend over time of publications (see Figure 11), an increased interest in the
design and strategical planning emerges from 2011, with many papers analyzing the user’s behavior and
the demand estimation to identify the market places. One way mode is the mostly analyzed, with a
prevalence of operational and real time analysis (see Figure 12). In particular, the interest in operational
and real time matters is justified by the need of solid relocation strategies and accurate location of the
parking stations. Finally, concerning the relationship between time horizon and engine type (Figure 13),
all the issues in the time horizon dimension are mostly investigated with fully thermic engine vehicles.
Furthermore, green engine vehicles are more analyzed according to the service design aspects (e.g. user
behavior). This is logic by considering the recent introduction of green car fleets in commercial car-sharing
3.5. Methodologies analysis
As shown in Figure 14, 61% of the analyzed papers use a statistical analysis tool, while combinatorial
optimization and stochastic programming method have a limited diffusion. This trend seems to be in
contrast with the general need of limiting the costs of the service operations. It may be linked to a
more general lack of linking between business models, business development, customer discovery and
validation, and operational models (see Subsection 4.4). In recent years, the use of statistical analysis
tools increased, accordingly to the increase of the number of publications (see Figure 15). Crossing
methodologies with the service mode (see Figure 16), statistical analysis is used for all the different
service modes, while simulation and optimization (both stochastic and combinatorial one) are mainly
used for one way car-sharing service. Simulation tools are largely used in studies related to fully thermic
engines, while green engines are mainly analyzed using optimization tools (see Figure 17). This behavior
can be explained with the need, in the case of electric cars, of a more accurate phase of design of the
service and planning of the infrastructures location due to the need of charging stations. Considering
methodologies and optimization objectives (Figure 18), while statistical analysis is mainly used in studies
related to business and service optimization (usually based on time series data collection), infrastructure
optimization is largely analyzed using stochastic optimization tools. Fleet management optimization is
mainly analyzed with both simulation and optimization tools. Finally, design issues are mainly faced
with statistical analysis tools; stochastic optimization is largely used for planning issues, while simulation
is the main tool for the analysis of operational and real time problems (Figure 19).
4. General trends and research perspectives
The aim of this section is to present a picture of the general trends and the research perspectives
emerging from the literature. In Subsection 4.1 the focus is on the users behavior analysis, used mainly
to determine how a car-sharing service could meet the customers mobility needs or the changes in users
behavior consequent the introduction of a car-sharing service in a specific area. These studies are mainly
aimed to quantify or at least estimate the positive effects (on environment, quality of life, land usage and
traffic congestion) of the car-sharing services. Sub-section 4.2 presents the situation of the analysis of
the demand, a key factor for the car-sharing operators to estimate the potential demand of the service in
a specific area necessary to justify investments. For addressing this issue, it is important to accurately
collect the primary data (Subsection 4.2.1); the two data collection methods mainly used are time series
(in which historical data were analyzed to understand the underlying structure of the phenomenon)
and qualitative and quantitative surveys (also including direct interviews, focus groups, brain storming
sessions). The collected data are then analyzed (Subsection 4.2.2), most commonly by Logit models
(used in estimating the parameters of a qualitative response model) and logistic regression. Sub-section
4.3 analyzes the papers according to their optimization objectives: business and service optimization,
infrastructure optimization and fleet management. Finally, sub-section 4.4 focuses on one of the biggest
lacks in the literature: the absence of studies related to the business models linked to car-sharing services,
their business development process, as well as the value proposition and customer segmentation.
4.1. Users behaviors and factors affecting the adoption
Since the early years car-sharing has entered in the market, several researches are focusing on the
impact of car-sharing on the urban mobility. They investigate mainly the characteristics of the services
and the impact on users travel patterns an behaviors to estimate the potential demand and to investigate
the main drivers of adoption [7, 8, 49, 62, 93, 104, 114, 168].
In [8, 100, 104, 119] the authors carry on an overview of the car-sharing service (through experts inter-
views), focusing on the main involved actors and drivers of adoption. The most relevant factors impacting
the growth of the system are related to parking policies, technology, vehicles, fuels and insurance.
Several publications examine mobility behavior of members and potential members of car-sharing
services. In different studies [39, 109, 110, 116, 144, 152, 165] the authors use stated preference methods
to investigate the awareness and the acceptance of the service among car-sharing members. Then, by
regression models, they identify the correlations between membership and social and demographic factors.
Similarly, in [41] a binomial regression is used to model a spatial diffusion of car-sharing membership in
Quebec City from 1996 to 2008, discovering how socio-economic factors such as education, motorization
and family structure affect heavily the membership rate in the covered area.
Cohen et al. show how the adoption of car-sharing services is greater in high-density neighborhoods,
where public transportation is more efficient and the usage of the private car is less frequent due to City
regulations and restrictions [40]. In [116] Morency et al. estimate the factors affecting awareness and
acceptance using both linear and logistic regression models, focusing on multimodal mobility patterns.
Shaheen and Cohen present a method for the estimation of the potential market in Germany, analyzing
potential users with objective and subjective criteria [144]. Unlike other studies, they focus on the
satisfaction level towards other transportation means. Using a Logit model, the study reveals that bus
travelers are more attracted from car-sharing models, while people traveling with high frequency and
through longer distances are less interested. In [3] the authors introduce an innovative transportation
concept in which personalized services are provided in real time to the customers, allowing them to
select the best one from a list of travel options. A more recent study individuates the presence of a latent
demand for car-sharing in a specific area, with the consequence that the increase of the number of supplied
vehicles combined with a marketing campaign could lead to an increase of car-sharing membership [152].
In [109] and [110] users are asked about their willingness to adopt different hypothetical service plans, with
different pricing policies and vehicle distribution, examining the service characteristics and evaluating the
economic utility of each plan. In the last years, the growth of the performances of electric vehicles, the
need of an electric mobility trajectory [46] and the increased attention to environmental issues shift the
attention towards studies related to electric car-sharing systems, mainly focusing on propensity of the
users and barriers to adopt these vehicles [43, 46, 51, 65, 69, 91, 94, 112, 137, 141]. Kumar and Bierlaire
investigate the potential demand of these type of services in an academic community, with the result
that electric vehicles could be chosen for short distances [94]. Furthermore, the research shows that the
main factors of adoption of car-sharing services are related to cost reduction and traffic congestion, and
respondents are willing to pay an additional cost for the use of an electric vehicle. Some studies, including
[137], consider a specific age target, using factor analysis and ordered Logit models to investigate their
satisfaction about current travel patterns and to evaluate the willingness to join electric car-sharing
services. In [69] the authors employ surveys on expectation and attitudes of users before and after using
the services, and then use qualitative methods to identify the motivations leading to successful adoption
of hybrid and electric car-sharing services. The results show that electric car-sharing is perceived as part
of an integrated transport system for short trips, with the consequence that the issue of range (also called
Range Anxiety) have low relevance. In [109] and [110], and more recently in [43], authors investigate
about economic utility of round-trip car-sharing services (which include also electric vehicles) employing
both multinomial Logit and mixed Logit. The result is a reluctance to adopt fully electric vehicles,
particularly for long distances, even if the distance is lower than the range of the vehicle, maybe because
of uncertainty in predicting travel patterns. Similar studies, considering on the impact of electric vehicles
in free floating car-sharing services, are presented in [106, 150].
Several studies [42, 48, 50, 66, 85, 89, 96, 108, 132, 135] focus on the behavior of current users
of car-sharing services all over the world, mainly investigating on drivers of usage, changes in travel
behavior before and after joining the service and membership duration. These papers analyze datasets
from car-sharing operators or surveys. In [25, 42, 45, 48, 66, 85] and [115] the authors tried to predict
the optimal location for the stations to maximize the integration of car-sharing service with other public
transportation means. Genikomsakis et al., through a Geographic Information System (GIS), analyze
regions already covered by the service, correlating factors like parking pressures, population density, age of
the neighborhoods and car-sharing service level (defined as the vehicle availability) with census data [66].
A similar methodology is recently used in [25]. A GIS-based analysis carried on in several regions of the
US finds that transportation characteristics are stronger than demographic information as indicators for
car-sharing success. The aim of the study is to present a tool to identify neighborhoods and factors, as low
vehicle ownership and high rate of one-person households, in which car-sharing could positively operate;
the same tool has then been tested in Austin (Texas). In [85] Kato et al. find out how the success of
car-sharing services is strictly correlated with factors like the size of car-sharing stations, seasonal impact,
age of the vehicles and multimodal transport network, i.e., presence of different transportation modes
nearby the car-sharing stations. Similar results are obtained in [42], where the role of other transportation
modes on car-sharing services penetration is investigated, and in [48], where, through a logistic regression
and a duration model, the authors quantify the positive correlation with the number of stations and the
role of parking costs on the likelihood of vehicles renting. In [135] the authors model and forecast users
membership duration and usage patterns, finding out a positive correlation between vehicle availability
and frequency of use and between personal car ownership rate and membership duration. [50] and [89]
identify the changes in transport behavior before and after joining car-sharing service. In particular Khan
et al. show how in areas with population density up to 10000 persons/square kilometers the introduction
of car-sharing leads to a slight decrease in public transport usage, an increase of cycling and walking
and an average driving reduction [89]. Furthermore, [88] investigates, through a literature review, how
car-sharing services can address health problems. Results show that car-sharing services can reduce car
ownership and change travel behavior, with potential positive impacts on health related to the adoption
of more active transport modes. This paper also introduces one of the major issues of car-sharing services:
the impact of car-sharing in household car ownership. Two opposite trends emerge: some studies state
that car-sharing could contribute to a reduction of total number of cars mileage as well as a reduction of
car ownership, while other studies affirm that car owners could not forego private car and non-car owners
may use car-sharing instead of other public transport modes. [74] and [90] explicitly consider this aspect.
In details, the first paper analyzes the de-motorization potential of the service with consequent impacts
on environment through surveys to different car-sharing service members in North America, showing the
different efficiency between the private car (of which age and model were asked in the surveys) and the
car-sharing service. Hildebrandt et al. investigate users travel patterns, vehicle usage and membership
duration in correlation to several characteristics of the service, stating that car-sharing users are usually
environmental friendly, and that less perceived cost savings usually lead to shorter membership duration
and frequency of usage [74]. In [24] both in deep interviews and focus groups are used to investigate
the propensity of customers to adopt a bundle of products and services, facing their concerns about the
possibility that their needs might be unsatisfied. Results show that it is important for the planners to
focus on the interaction with customers for gaining their confidence on the service. Another important
issue is to educate customers on the life cycle costs of the products, to increase their awareness on the
potential savings related to service adoption. In [33] the authors examine the car-sharing impact on car
ownership in dense urban areas, finding that car-sharing members reduce their individual transportation
cost and emissions.
In more recent years, several studies concerned the impact of the new emerging ICT and mobile
technologies on the car-sharing system [70, 74, 98]. In [74] the authors show the fundamental role of
information systems for the success of the car-sharing systems, allowing real time information on the
fleets and helping users in the localization of the available vehicles. Applying an optimization model to
a one-way car-sharing service, the authors found out a potential increase of car-sharing operators profits
due to users’ flexibility and real time information. Information systems are analyzed also for electric
car-sharing systems in [98], where mobile technology can supply the necessary infrastructure to let the
system correctly operate. Firnkorn and M¨uller consider the strategy of an automaker entering in the car-
sharing market (focusing on the specific case of Car2Go) with the scope to start a new business segment
and reach positive effects on branding [61]. After the collection of primary data (all Car2Go members in
2010 in the city of Ulm were invited to answer to an on line survey), empirical analysis are conducted
to evaluate the impact on private vehicle holding after 18 months of service. Results show that after
18 months of service Car2Go has contributed to a strong reduction of vehicle ownership, with an even
stronger potential impact estimated for the future. [138] focuses on how the effects of risk perception
of products ownership can influence the adoption of access-based services, bringing to the conclusion
that a higher usage of access-based services can increase the likelihood in ownership reduction by the
customers. Finally, Kopp et al. analyze the travel behavior of members of a free floating car-sharing
service, comparing the results with similar results from a sample of non-members. Results show evidence
that car-sharing members are more multimodal than non-members, and even the distance traveled are
lower for the car-sharing members [92].
Even if the literature is quite clustered, some trends emerge. First, the perception of the citizens
is changing over time with the diffusion of the car-sharing services. Thus, people is moving from a car
property vision to a car-as-a-service one. Second, the pervasive presence if ICT and mobile technologies
is pushing the penetration of the car-sharing services.
4.2. Demand analysis
Car-sharing systems management has a complexity directly linked to the interplay effect of demand
and supply. To optimize operating issues and to estimate the effects of car-sharing services on mobility
management it is useful an accurate model of demand and supply [80]. Due to the strong dependence
between the availability of vehicles and the number of trips, it is difficult to correctly model the car-sharing
demand. Although various car-sharing simulation models were presented across the last years, it emerges
a difficulty in representing accurately the supply side, focusing in particular on the cost-benefit analysis
necessary to justify investments. Moreover, the recently growth of free floating car-sharing systems
introduces further complexity, adding uncertainty as to the location where the vehicles can be picked up
and returned. For modeling correctly the service, a key aspect is the data collection (Subsection 4.2.1),
in which a properly recording of the variables must be ensured, and the tools developed to carry on the
analysis (Subsection 4.2.2). Data collection, i.e., the detection of the variables that are the object of study,
is a key activity. Some variables, such as age and gender can be detected easily, others are more difficult
(such as variables related to cognitive, emotional, behavioral, learning, etc). In the literature, data are
mainly collected by two methods: time series and questionnaires/focus groups/interviews/brainstorming.
Time series is a forecasting technique that is based on historical data from which the analyst tries to
understand the underlying structure of the phenomenon. The questionnaires contain both open and
closed questions. In the first case, the subject chooses between the various alternatives proposed, while
in the second may express his views freely. In focus groups a selected group of users are asked about
their opinion, perception, beliefs and attitude towards the service, with free interaction between different
group members, while in brainstorming a specific problem is addressed by a group of experts and users,
with the scope to gather a list of potential solutions. Finally, in direct interviews an operator interview
users and non-users of the service to collect the information.
4.2.1. Data collection
The most common method for collecting data to analyze them and such to capture the latent demand
is the questionnaire. It is used by Cervero [31] on a macro sample that includes business and neighborhood
car-sharing. Later, Zhou and Kockelman administer a questionnaire to the city of Austin, Texas [166].
The purpose of this study is to investigate the latent demand during the launch of the car-sharing service
in Austin. Huwer places the emphasis on the combination of public transport with car-sharing service.
Members and non-members of car-sharing service are selected randomly and interviewed [75]. Catalano
et al. publish a study reporting a stated-preference survey in Palermo [23]. The respondents can choose
from different transportation alternatives, which include private car, public transport, car-sharing and
car-pooling. Then a random utility model is estimated using the survey data. The authors infer that in
a future scenario characterized by active policies to limit private transport use the car-sharing market
could increase up to 10%. Sioui et al. use two types of survey to gather data on travel behaviors on a
typical day, widespread in Montreal, Canada, to analyze the car-sharing demand [149]. The first one is a
regional, large-scale household travel survey and an internet survey, started after the introduction of the
service, while the second one includes both former members and current members of the service. The
study makes a comparison between respondents of two types of survey with similar characteristics and
located in the same municipal sector. Herrmann et al. solve the problem of car relocation in free floating
car-sharing [73]. For this purpose, a survey is conducted among users of the Car2Go system in Hamburg,
Germany. The survey is intended for users and potential users of free-floating car-sharing service. The
same method, extending the analysis also to electric car-sharing, is used by Wappelhorst et al., with 311
persons interviewed for the first project and 280 persons in additional personal short interviews for the
second one [159]. Firnkorn and M¨uller compare the results of two methods measuring the impact of car-
sharing on other transportation modes, starting from the same sample [60]. The first method estimates
how the mobility behavior of respondents would be with the assumption of unavailability of Car2Go,
while the second one determines the respondents past mobility behavior using Car2Go. Rabbitt and
Ghosh conduct a survey in Ireland, with 2639 respondents among population of likely users concentrated
in areas of higher population densities [131]. They present a new methodology for estimating the potential
market and the impact of car-sharing system in Ireland. Accordingly to their results, all the small areas in
the Republic of Ireland are sub-divided into groups based on the viability of introduction of car-sharing
system in the area. In the same year, Ohta et al. cluster the 1095 respondents of web-based survey
into similar sized groups based on the number of cars owned and on the residential area to investigate
the effects of community size [123]. In the survey, respondents are provided with car-sharing and eco-
cars information and subsequently are asked questions about both car-sharing services and eco-cars.
Shaheen et al. [147] interviewed car-sharing actors with mail questionnaires, telephone interviews and
other internet information, to compare Canadian car-sharing demand with North American car-sharing
demand. In [61] primary data were collected by an online survey on the Car2Go members of the City
of Ulm with the aim to analyze the variations in users behavior after the introduction of the service.
Focus groups, combined with in deep interviews of car-sharing experts, are used in [24] to investigate
the propensity of the customers to adopt bundles of services as a substitute of products. Unlike other
studies, in [92] data are collected by a survey based on a GPS tracking smartphone application with the
scope to analyze the travel patterns of both members and non-members of a car-sharing service.
Even the mostly part of the survey are covering a quite limited area, the authors try to infer general
trends. At the same, no general survey exists, differently from other domains, as guiding behaviors
in automotive [20] or the passenger’s perceived quality of service in air transportation [129]. There is
actually the need of a standard survey that will be used as a basis for the comparison of the different
results geographically and would let to infer the behavioral changes over time. This is big lack in the
literature that, up to now, is far to be filled.
4.2.2. Tools of analysis
In this section we highlight the main trends in the literature related to the tools used to analyze the
data gathered in the research.
Logit and linear regressions. One of the most productive streams of research on car-sharing has
been the study of the characteristics of its users. In several works the population characteristics are
modeled using a sample of car-sharing users, analyzing these data with a Logit model. Binomial Logit
analysis is used in 2003 by Cervero [31] to control variable factors such as price of gasoline and weather,
while a matched-pair analytically approach is adopted in [139]. To understand the drivers of adoption
of an urban car-sharing program, and to establish which modeling approach was the most effective, De
Luca and Di Pace [47] investigate multinomial Logit, hierarchical Logit, cross nested Logit and mixed
multinomial Logit models. Frost and Sullivan outline the role the car-sharing sector can play in reducing
the private car usage in London to 2020. Statistical analysis (multiple linear regression and a compound
annual growth rate) shows which existing socio-demographic and neighborhood factors have most affected
car-sharing membership. Ohta [123] in his study on Japan, uses a multiple linear regression analysis,
to examine the effects of individual attributes, including gender, age, number of cars per household and
area of residence, with the purpose of finding information on behavioral intentions regarding joining a
car-sharing organization in different situations. In 2014, Schm¨oller and Bogenberger [139] analyze the
differences of the booking behavior between free floating car-sharing and hybrid car-sharing. Hybrid
car-sharing differs for the size of parking area where cars can park (e.g. 1 km2areas predefined by the
city). Free floating car-sharing includes information about time, data and coordinates of beginning and
ending of the corresponding booking, while hybrid car-sharing contain only name of the area but not the
actual coordinates. In [157] the authors apply the conceptual framework called Perceived Activity Set
(developed by the same authors) to car-sharing market in the city of London, investigating both round trip
(two way) and point to point (one way) service modes. The aim of the study is to investigate the impact
of car-sharing services on the other forms of transport. A qualitative survey is used to investigate the
opinion of respondents relating subscribing and using car-sharing services, and a following a quantitative
model is used to forecast the number of potential subscribers, the level of usage and the impact on other
forms of transport. The results suggest that the number of potential members of a one way service mode
car-sharing is between three and four times as large as the comparable number of a two way service mode.
Furthermore, the greatest reduction in overall vehicle traveled miles was found from the introducing of two
way service mode. The reason lies in the fact that two way service mode is seen as a complement of public
transport, while one way service mode is more likely seen as a substitute of public transport. Historical
data and zone categorization are used in [161] to introduce a relocation model considering both electric
and conventional vehicles, combining relocation with charging (for the electric vehicles), and refueling (for
the conventional ones). Booking data are also analyzed in [117, 140] to introduce the relocation problem
and the short time booking predictions in free-floating car-sharing services. In [17] car-sharing map data
from a large station-based car-sharing operator in France are analyzed to determine whether a station
is profitable or not for the operator. In [158] customer data are used to explain, through a regression
model, spatial variations in car-sharing activities in the proximity of particular points of interest. Another
important issue for car-sharing systems is to determine the operating area. In [142] population density,
housing rent, city center distance, and hotel and restaurant density are used as independent variables to
predict booking hot spots in a free-floating car-sharing system. The influence of the stations’ location
on potential membership is analyzed in [37] for a two-way car-sharing system. The importance of the
collaboration between public and private sector is addressed in [155] to provide recommendations for both
sectors, starting from the results of expert interviews conducted with governments and private operators.
Simulation. Ciari et al. in [36] and again, the following year, the same authors in [35] propose an
Activity-Based Microsimulation Approach called MATSim, based on two features of car-sharing system:
the access to rental car and the time dependent fee. Starting from micro level, model can determinate
macro behavior of the system. A test was carried on in Greater Zurich where the car-sharing stations are
276. Instead, the second paper use the multi-agent simulation MATSim to evaluate different car-sharing
scenarios for the city of Berlin. The likelihood of the whole representation is guaranteed by the fact
that the artificial population is based on census data and on travel diaries surveys. The first scenario
considers only station-based car-sharing available, while the second scenario suggests that there is a peak
around 8 AM, and in the third scenario seems that the addition of free floating to traditional car-sharing
does not affect the latter. In 2013 Weikl and Bogenberger presented a new integrate two-step model
for optimal vehicle positioning and relocation and apply an optimization algorithm for finding the best
relocation strategy in case of deviation [160]. The empirical basis of this work are real historical vehicle
data of a real-life free-floating car-sharing system in Munich, Germany. The historical data consists of
the geo-referenced start and end locations of the conducted trips, booking times and booking duration,
satisfied and unsatisfied booking requests (online requests/searches by mobile phone or Internet which
did not lead to a booking), trip distances and parking duration. The relocation problem is also addressed
in [38, 82, 122] to present models to simulate different relocation strategies, with the aim to maximize
the profitability of the service. A comparison between different relocation strategies, also involving
autonomous vehicles, is presented in [30].
To predict the potential market demand for the car-sharing operator impact, [106] developed a sim-
ulation model based on Stated Preference experiments in order to obtain the necessary insights for this
newly proposed transport mode. In 2012, an agent-based scheduling and energy management system
was used to optimize the utilization of the energy produced locally and of the batteries in an electric
car-sharing fleet [64]. In 2014 Lopes et al. apply an agent based model to represent the daily operation
of a hypothetical car-sharing program operating in the city of Lisbon [101]. Some indicators were chosen
to evaluate the performance of the system, both from an economic perspective and from an operational
point of view. [67] combines technology road-map and system dynamics simulation to evaluate the envi-
ronmental, social and economic perspectives of car-sharing in Korea for a long-term period of 50 months.
This work represents the first attempt to simulate the demand and behavior changes of users, which are
expected to be a major barrier for the car-sharing business. In [67] three scenarios of car-sharing services
are developed and analyzed by the authors as an illustrative example of a scenario-planning approach to
develop technology road-maps. Using system dynamics each strategic model for technology road-map is
then transferred to the operational viewpoint.
Neural network. In 2007, Xu et al. [163] propose an evolutionary neural network to address the
problem of forecasting net flow of car-sharing systems. The forecast is made according to the data
obtained by real case of Singapore car-sharing. For the forecast, it’s important to choose an appropriate
neural network structure, including the number of hidden layers, and the number of nodes in each hidden
layer. The common approach is to fix the structure and then use genetic algorithm to search the global
minimum and back propagation method to speed up the convergence around the local minimum.
The literature shows a large usage of Logit model, with all its variations (binomial Logit, multino-
mial Logit, hierarchical Logit, cross nested Logit and mixed multinomial Logit), used in estimating the
parameters of a qualitative response model. Logistic regression is used to refer specifically to problems
in which the dependent variable is binary, while problems with more than two categories are referred to
multinomial logistic regression, or, if the multiple categories are ordered, to ordinal logistic regression.
Another methodology heavily used is the simulation-based approach, while only one paper adopts neural
networks [163]. It is not clear if this trend is due to the specific expertise of the authors or by a clear
predominance of Logit models over the remaining methods. Thus, a possible and suggested research line
might be to better explore new tools, as neural networks, or hybridized simulation-optimization methods,
that proved their efficacy in several applications [67, 129, 163].
4.3. Optimization in car-sharing services
As stated in Section 3, the use of optimization tools (i.e., combinatorial and stochastic optimiza-
tion, and simulation) is limited in car-sharing services. Design of the service, infrastructures and fleet
management are the main objectives to be optimized in a car-sharing system.
To support planners of car-sharing services in business and service design, Geum et al. proposed a sce-
nario building approach that integrates technology roadmapping and system dynamics. This framework
aims to evaluate scenarios, reflecting the results on technology roadmap, and the long term behavior of
the system [67]. Other papers address business and service objectives using simulation approaches. Ciari
et al. developed an activity-based microsimulator [36] and, more recently, an agent-based simulation [35]
to estimate the travel demand and to evaluate how different policies can affect it in station-based and
free floating modes. New car-sharing modes has been studied recently mixing standard modes [35, 79].
A hybrid system that integrate one-way and an existing two-way mode is studied by Jorge et al.. They
present an optimization method to design and evaluate the hybrid system. When trips connect high de-
mand generator nodes (e.g. airports), the hybrid system increases profitability and, enabling relocation
operations, improves the demand served.
Infrastructure objectives mainly refers to find the optimal location of car-sharing stations and their
capacity. Kumar and Bierlaire optimize the locations of future stations around the city of Nice, France, by
a linear regression and an optimization model [94]. In [45], three Mixed Integer Programs (MIPs) are used
for determining the optimal locations for stations and selection trips of a one-way car-sharing operator in
Lisbon, Portugal. A MIP model is also used in [136] to manage in real time the relocation and maintenance
operations in a one-way car-sharing system. De Almeida Correia and Antunes minimize the cost of
relocation explicitly considering car imbalance issues in MIPs. MIP is also the base of the equilibrium
network design model formulated by Nair and Miller-Hooks. With the aim to determine the optimal
configuration, in terms of location and capacity of stations, the model maximizes the revenue of the
operator [118]. The same problem is studied by Rickenberg et al.. The authors propose an optimization
model that finds the best location and size of stations given a users’ demand and preferences. The
model is than integrated in a Decision Support System (DSS) [134]. Simulation-optimization approach
is used for the same problem in [81]. The simulation considers uncertainty on trips and tests a vehicle
relocation policy at time. Most of the studies analyze location problems in general and do not consider
political issues that limit the possible location of new stations [14, 99]. Focusing on user’s perceptions,
[77] proposed an approach to help local authorities at selecting new sites for existing electric car-sharing
services. The approach is based on the modeling of the preferences of the potential subscriber of the
Fleet management optimization includes the planning of the fleet size, the relocation strategies of the
vehicles, pricing and parking policies. At a tactical level, optimization tools are used to investigate the
optimal fleet size (number of cars) of a car-sharing system. In [56], the authors addressed the problem
of determining the optimal fleet size of an electric car-sharing service. The system is modeled as a
discrete event system in a closed queuing network, considering the specific requirements of the electric
vehicles utilization. In [101] the authors solve the same problem by an agent based simulation to asses
the potential of one-way car-sharing system. The model determines the fleet size, the location of stations
and position of vehicles, as well as the relocating operation, scheduling and pricing policies. It also
incorporates the uncertainty on the demand and the road network of Lisbon. Nourinejad and Roorda
consider the cooperation of tactical and operational decisions in fleet sizing. They combined two integer
programming models to plan the fleet size for a given demand and to schedule relocating operations
that maximize the profit of one-way, two-way and hybrid car-sharing systems [120]. Relocation strategies
answer to imbalance issues in one-way mode aiming to reduce management costs and to provide flexibility
of the service. Operator-based relocations are evaluated in [86], while a comparison between operators
based and users based relocation strategies is presented in [27]. The same authors evaluate a scenario in
which fully automated vehicles can move among different stations when relocation is required in [29]. The
development of these technologies could mainly impact positively on the efficiency of the transport system,
optimizing landscape usage and increasing the transportation safety [1]. Simulation based optimal vehicle
assignment and relocation at the stations have been proposed. [156] presented grouping of users to balance
the system, while [73] address the same problem using a discrete-event simulation model. Kaspi et al. use
an agent-based simulator to evaluate the improvement in terms of quality and flexibility of one-way mode
with parking reservation policies [84]. Optimization methods have been used for relocation problems
[19, 55, 54, 18]. Bruglieri et al. use a MIP for the relocation of 30 electric vehicles in Milan reducing the
average working time for the operations. A multi-stage stochastic linear programming model has been
proposed by Fan to handle with uncertain demand, while Boyacı et al. present a multi objective model
with electric vehicles charging requirements. From a resources view-point, Kek et al. propose a DSS based
on an simulation-optimization environment to evaluate the effects of different relocation strategies and
to determine manpower and operating parameters. Marouf et al. solve the vehicle distribution problem
focusing on automatic parking and platooning of the vehicles for reducing the manpower [107]. As cited
above, relocation strategies have been broadly studied in literature, showing the high costs related to
imbalance issues. Dynamic pricing policies can improve the profit of operators more than perfect balance
scenario [83]. Jorge et al. propose a strategy based on using the clients’ behavior to improve the balance of
vehicles within the network by means of dynamic pricing policies for trips, decreasing the price when a trip
increases the balance of the system. The optimal prices are defined with a non-linear optimization model.
In [22] an optimization problem is addressed in order to maximize the total number of satisfied demands
with a limited number of relocation operations. From this analysis emerges how the fleet management
is the objective most studied in literature from an optimization view-point. Moreover, [72] presents a
relocation strategy in which the relocation staff is moved with a shuttle, and the objective is to maximize
the number of relocation operations minimizing the travel duration of the shuttle.
Although the uncertainty is a key factor in optimizing and planning car-sharing operations, it is
usually surrogated by simulation-based methods and decisions are driven by deterministic and combina-
torial programs. Several papers presented scenario approaches and forecasting techniques to handle with
dynamic and stochastic demand in the car-sharing environment. Other sources of uncertainty, mainly
related to the network, are only partially considered (e.g. road congestion). They should be explicitly
considered in optimization tools, in particular for business, design and strategic decisions. Just a few
papers make use of more sophisticated methods, as stochastic programming or non-linear approximations
of the uncertainty should be considered [153, 154]. Moreover, the optimization models in the literature
privilege operational aspects, disregarding the business development and management ones. In particular,
no study considering the tariffs and their link to the operation issues emerges from our analysis. This is
a big lack typical of a research community closed in its domain and with almost no cross-contamination
with other domains. This is a big thread for the car-sharing industry. In fact, the typical outcome is or a
clear separation between academics and industry, or a failure of the business model due to an erroneous
business development schema.
4.4. Business models, business development and economics
To our knowledge, [13] is the only paper that partially analyze the business of sharing mobility services,
in particular of car-pooling, using a framework linking business factors and service strategies. The
framework mainly focuses on service aspects and does not consider explicitly more complex aspects, as the
business model, its link to the business development model, as well as the value proposition of the different
car-sharing companies. This, in fact, impacts on the tariff schemes, as well as on the service penetration
in the market. As this lack can be partially tolerated in a pioneering phase, it must be compulsorily
considered in a more mature phase of the market [124]. As highlighted also in [127], the current tariff
schemes are rigid. On the contrary, too flexible tariff schemes risk to compromise the profitability and
the operations efficiency of the car-sharing service, as happened in the telecommunication sector. In this
context, the tariffs represent a way to fit on the one hand public policies, operating models and business
strategies defined by companies and on the other hand the users’behaviors. These customizable tariffs
and the balance between the above mentioned components push the service penetration in the market. A
first in-depth study of the link between the Business Models and the tariff schemes is recently presented
in [127]. The authors propose a comparative analysis of the Business Models of different companies
according to the GUEST Lean Business methodology [125, 128] and derive a simulation-optimization
tool able to perform a quantitative comparison of different tariff schemes. Although those first studies,
more research should be devoted in this direction, by integrating other methodologies that was already
proven to be effective in other domains as bilevel optimization and stochastic programming [71, 130, 44].
5. Conclusions
More recently, car-sharing services have become increasingly popular all over the world: old operators
have increased their fleets and approached new market places, and new operators have started their
business. Furthermore, the growing attention in environmental issues involved an increased attention in
the usage of electric and hybrid vehicles. In addition, the development of information and communication
technology (ICT) allowed the market penetration of new car-sharing models, such as free floating car-
sharing services. In fact, as stated by Hayashi et al., car-sharing companies are realizing and offering
new services through the use of ICT tools. Moreover, the introduction of ICT-enabled functionalities
compatible with mobile devices, enhances the adoption and thus, spreads the car-sharing service between
users. For example, some ICT-enabled services are the unlocking/locking of the vehicle using smart card
or mobile app, dynamic location information on maps, etc. In this paper, we introduced a taxonomy
able to categorize the existing literature and, by applying it, to derive some trends and directions for the
future research. In more detail, it emerges a gap between the literature and the business development
of the market. This gap becomes more and more evident when we look at the revenues generated by
the companies, still marginal compared to the capital in use. As stated in the introduction, a taxonomy
is only the first step. In fact, the business model and the link between the business and the operations
models, the tariff scheme, need to be integrated.
6. Acknowledgments
Partial funding for this project was provided by ”Smart Cities” strategic program of the ”Istituto
Superiore Mario Boella” and the Italian University and Research Ministry (MIUR) under the UrbeLOG
project-Smart Cities and Communities. While working to this paper prof. Guido Perboli was the leader
of the Urban Mobility and Logistics Systems initiative of the CARS@POLITO Interdepartmental Center.
The authors are grateful to Dr. Luca Gobbato and Valeria Caiati for their contribution to this analysis.
We also gratefully acknowledge the support of Giancarlo Pirani of the ”Istituto Superiore Mario Boella”.
[1] Alessandrini, A., Campagna, A., Site, P. D., Filippi, F., and Persia, L. (2015). Automated vehicles and the rethinking
of mobility and cities. Transportation Research Procedia, 5:145 – 160.
[2] Alfian, G., Rhee, J., and Yoon, B. (2014). A simulation tool for prioritizing product-service system (pss) models in a
carsharing service. Computers & Industrial Engineering, 70:59–73.
[3] Atasoy, B., Ikeda, T., Song, X., and Ben-Akiva, M. E. (2015). The concept and impact analysis of a flexible mobility
on demand system. Transportation Research Part C: Emerging Technologies, 56:373–392.
[4] Bailey, K. (1994). Typologies and taxonomies: An introduction to classification techniques. London: Sage Publications.
Quantitative Applications in the Social Sciences.
[5] Bailey, K. (2005). Typology construction: Methods and issues. In Encyclopedia of Social Measurement. Elsevier.
[6] Balac, M., Ciari, F., and Axhausen, K. (2015). Carsharing demand estimation: Zurich, Switzerland, area case study.
Transportation Research Record, 2536:10–18.
[7] Baptista, P., Melo, S., and Rolim, C. (2014). Energy, environmental and mobility impacts of car-sharing systems.
empirical results from Lisbon, Portugal. Procedia - Social and Behavioral Sciences, 111:28 – 37.
[8] Barth, M. and Shaheen, S. (2002). Shared-use vehicle systems: Framework for classifying carsharing, station cars, and
combined approaches. Transportation Research Record: Journal of the Transportation Research Board, 1791:105–112.
[9] Barth, M., Shaheen, S., Fukuda, T., and Fukuda, A. (2006). Carsharing and station cars in Asia: overview of Japan
and Singapore. Transportation Research Record: Journal of the Transportation Research Board, 1986:106–115.
[10] Barth, M. and Todd, M. (2001). User behavior evaluation of an intelligent shared electric vehicle system. Transportation
Research Record, 1760:145–152.
[11] Barth, M., Todd, M., and Shaheen, S. (2003). Intelligent transportation technology elements and operational method-
ologies for shared-use vehicle systems. Transportation Research Record, 1841:99–108.
[12] Benjelloun, A., Crainic, T. G., and Bigras, Y. (2010). Toward a taxonomy of city logistics projects. Procedia - Social
and Behavioral Sciences, 2:6217–6228.
[13] Beutel, M., Samsel, C., Mensing, M., and Krempels, K.-H. (2014). Business model framework to provide heterogeneous
mobility services on virtual markets. In Proceedings of the 11th International Conference on e-Business, Part of ICETE
2014 - 11th International Joint Conference on e-Business and Telecommunications, pages 145–151.
[14] Biondi, E., Boldrini, C., and Bruno, R. (2016). Optimal deployment of stations for a car sharing system with stochastic
demands: A queueing theoretical perspective. In Intelligent Transportation Systems (ITSC), 2016 IEEE 19th Interna-
tional Conference on, pages 1089–1095. IEEE.
[15] Boesch, P. M. and Ciari, F. (2015). Agent-based simulation of autonomous cars. In American Control Conference
(ACC), 2015, pages 2588–2592. IEEE.
[16] ohmov´a, K., Disser, Y., Mihal´ak, M., and ˇ
Sr´amek, R. (2016). Scheduling transfers of resources over time: Towards
car-sharing with flexible drop-offs. In Latin American Symposium on Theoretical Informatics, pages 220–234. Springer.
[17] Boldrini, C., Bruno, R., and Conti, M. (2016). Characterising demand and usage patterns in a large station-based car
sharing system. In Computer Communications Workshops (INFOCOM WKSHPS), 2016 IEEE Conference on, pages
572–577. IEEE.
[18] Boyacı, B., Zografos, K. G., and Geroliminis, N. (2015). An optimization framework for the development of efficient
one-way car-sharing systems. European Journal of Operational Research, 240(3):718–733.
[19] Bruglieri, M., Colorni, A., and Lu, A. (2014). The vehicle relocation problem for the one-way electric vehicle sharing:
An application to the Milan case. Procedia - Social and Behavioral Sciences, 111(0):18 – 27.
[20] Burzio, G., Mussino, G., Tadei, R., Perboli, G., Dell’Amico, M., and Guidotti, L. (2009). A sub jective field test on
lane departure warning function in the framework of the euroFOT project. In Human System Interactions, 2009. HSI
’09, pages 608–610. IEEE Press.
[21] Caputo, A. (2012). Insights on Italian car sharing: the case of Genoa. International Journal of Business and Globali-
sation, 9(2):207–220.
[22] Carlier, A., Munier-Kordon, A., and Klaudel, W. (2015). Mathematical model for the study of relocation strategies in
one-way carsharing systems. Transportation Research Procedia, 10:374–383.
[23] Catalano, M., Lo Casto, B., and Migliore, M. (2008). Car sharing demand estimation and urban transport demand
modelling using stated preference techniques. European Transport, pages 33–50.
[24] Catulli, M. (2012). What uncertainty? Further insight into why consumers might be distrustful of product service
systems. Journal of Manufacturing Technology Management, 23(6):780–793.
[25] Celsor, C. and Millard-Ball, A. (2007). Where does carsharing work? using geographic information systems to assess
market potential. Transportation Research Record: Journal of the Transportation Research Board, 1992(1):61–69.
[26] Cepolina, E. and Farina, A. (2012a). A new shared vehicle system for urban areas. Transportation Research Part C:
Emerging Technologies, 21(1):230–243.
[27] Cepolina, E. and Farina, A. (2012b). Urban car sharing: An overview of relocation strategies. WIT Transactions on
the Built Environment, 128:419–431.
[28] Cepolina, E. and Farina, A. (2014a). Simulation and optimization of a capillar third generation car sharing system.
In 16th International Conference on Harbor, Maritime and Multimodal Logistics Modelling and Simulation, HMS 2014,
pages 76–84.
[29] Cepolina, E. M. and Farina, A. (2014b). A methodology for planning a new urban car sharing system with fully
automated personal vehicles. European Transport Research Review, 6(2):191–204.
[30] Cepolina, E. M., Farina, A., Holloway, C., and Tyler, N. (2015). Innovative strategies for urban car-sharing systems
and a simulator to assess their performance. Transportation Planning and Technology, 38(4):375–391.
[31] Cervero, R. (2003). City carshare: First-year travel demand impacts. Transportation Research Record: Journal of the
Transportation Research Board, 1839:159–166.
[32] Cervero, R., Golub, A., and Nee, B. (2007). City carshare longer-term travel demand and car ownership impacts.
Transportation Research Record, 1992:70–80.
[33] Chen, T. D. and Kockelman, K. M. (2016). Carsharings life-cycle impacts on energy use and greenhouse gas emissions.
Transportation Research Part D: Transport and Environment, 47:276–284.
[34] Ciari, F., Balac, M., and Balmer, M. (2015). Modelling the effect of different pricing schemes on free-floating carsharing
travel demand: a test case for Zurich, Switzerland. Transportation, 42(3):413–433.
[35] Ciari, F., Bock, B., and Balmer, M. (2014). Modeling station-based and free-floating carsharing demand: A test case
study for Berlin, Germany. Transportation Research, 30.
[36] Ciari, F., Schuessler, N., and Axhausen, K. W. (2013). Estimation of carsharing demand using an activity-based
microsimulation approach: Model discussion and some results. International Journal of Sustainable Transportation,
[37] Ciari, F., Weis, C., and Balac, M. (2016). Evaluating the influence of carsharing stations location on potential
membership: a Swiss case study. EURO Journal on Transportation and Logistics, 5(3):345–369.
[38] Clemente, M., Nolich, M., Ukovich, W., and Fanti, M. P. (2015). A decision support system for the management of an
electric-car sharing system. In Systems, Man, and Cybernetics (SMC), 2015 IEEE International Conference on, pages
533–538. IEEE.
[39] Clewlow, R. R. (2016). Carsharing and sustainable travel behavior: Results from the San Francisco Bay Area. Transport
Policy, 51:158–164.
[40] Cohen, A. P., Shaheen, S., and McKenzie, R. (2008). Carsharing: a guide for local planners. Institute of Transportation
[41] Coll, M.-H., Vandersmissen, M.-H., and Th´eriault, M. (2014). Modeling spatio-temporal diffusion of carsharing mem-
bership in Qu´ebec City. Journal of Transport Geography, 38:22–37.
[42] Correia, G. H. D. A., Jorge, D. R., and Antunes, D. M. (2014). The added value of accounting for users’ flexibility and
information on the potential of a station-based one-way car-sharing system: An application in Lisbon, Portugal. Journal
of Intelligent Transportation Systems, 18(3):299–308.
[43] Costain, C., Ardron, C., and Habib, K. N. (2012). Synopsis of users’ behaviour of a carsharing program: A case study
in Toronto. Transportation Research Part A: Policy and Practice, 46(3):421–434.
[44] Crainic, T. G., Gobbato, L., Perboli, G., and Rei, W. (2016). Logistics capacity planning: A stochastic bin packing
formulation and a progressive hedging meta-heuristic. European Journal of Operational Research, 253(2):404–417.
[45] De Almeida Correia, G. H. and Antunes, A. P. (2012). Optimization approach to depot location and trip selection in
one-way carsharing systems. Transportation Research Part E: Logistics and Transportation Review, 48(1):233–247.
[46] De Lorimier, A. and El-Geneidy, A. M. (2013). Understanding the factors affecting vehicle usage and availability in
carsharing networks: A case study of Communauto carsharing system from Montr´eal, Canada. International Journal of
Sustainable Transportation, 7(1):35–51.
[47] De Luca, S. and Di Pace, R. (2015). Modelling users’ behaviour in inter-urban carsharing program: A stated preference
approach. Transportation research part A: policy and practice, 71:59–76.
[48] de Luca, S. and Pace, R. D. (2014). Modelling the propensity in adhering to a carsharing system: A behavioral
approach. Transportation Research Procedia, 3(0):866 – 875.
[49] Efthymiou, D. and Antoniou, C. (2016). Modeling the propensity to join carsharing using hybrid choice models and
mixed survey data. Transport Policy, 51:143–149.
[50] Efthymiou, D., Antoniou, C., and Waddell, P. (2013). Factors affecting the adoption of vehicle sharing systems by
young drivers. Transport policy, 29:64–73.
[51] Ensslen, A., Jochem, P., Sch¨auble, J., Babrowski, S., and Fichtner, W. (2013). User acceptance of electric vehicles in
the French-German transnational context. Selected proceedings of the 13th WCTR, Rio de Janeiro, Brazil, pages 15–18.
[52] Era2012 (2012). Excellence in research for Australia (era) 2012 national report. Technical report, Australian Govern-
[53] Fan, W. (2013). Management of dynamic vehicle allocation for carsharing systems. Transportation Research Record,
[54] Fan, W. D. (2014). Optimizing strategic allocation of vehicles for one-way car-sharing systems under demand uncer-
tainty. Journal of the Transportation Research Forum, 53(3):7–20.
[55] Fan, W. D., Machemehl, R. B., and Lownes, N. E. (2008). Carsharing: Dynamic decision-making problem for vehicle
allocation. Transportation Research Record: Journal of the Transportation Research Board, 2063(1):97–104.
[56] Fanti, M., Mangini, A., Pedroncelli, G., and Ukovich, W. (2014). Fleet sizing for electric car sharing system via closed
queueing networks. In Systems, Man and Cybernetics (SMC), 2014 IEEE International Conference on, pages 1324–1329.
[57] Ferrero, F., Perboli, G., Vesco, A., Caiati, V., and Gobbato, L. (2015). Car-sharing services part a taxonomy and
annotated review. Technical report, CIRRELT-2015-47.
[58] Figueira, G. and Almada-Lobo, B. (2014). Hybrid simulation–optimization methods: A taxonomy and discussion.
Simulation Modelling Practice and Theory, 46:118–134.
[59] Firnkorn, J. (2012). Triangulation of two methods measuring the impacts of a free-floating carsharing system in
Germany. Transportation Research Part A: Policy and Practice, 46(10):1654–1672.
[60] Firnkorn, J. and M¨uller, M. (2011). What will be the environmental effects of new free-floating car-sharing systems?
the case of car2go in ulm. Ecological Economics, 70(8):1519–1528.
[61] Firnkorn, J. and M¨uller, M. (2012). Selling mobility instead of cars: New business strategies of automakers and the
impact on private vehicle holding. Business Strategy and the Environment, 21(4):264–280.
[62] Firnkorn, J. and M¨uller, M. (2015). Free-floating electric carsharing-fleets in smart cities: The dawning of a post-private
car era in urban environments? Environmental Science & Policy, 45:30–40.
[63] Firnkorn, J. and Shaheen, S. (2016). Generic time-and method-interdependencies of empirical impact-measurements:
A generalizable model of adaptation-processes of carsharing-users’ mobility-behavior over time. Journal of Cleaner
Production, 113:897–909.
[64] Freund, D., Raab, A. F., Kuster, T., Albayrak, S., and Strunz, K. (2012). Agent-based integration of an electric car
sharing fleet into a smart distribution feeder. In Innovative Smart Grid Technologies (ISGT Europe), 2012 3rd IEEE
PES International Conference and Exhibition on, pages 1–8.
[65] Fukuda, T., Kashima, S., and Tanishita, M. (2001). A feasibility study on ev-sharing application to residential area
and its characteristics. In Vehicle Electronics Conference, 2001. IVEC 2001. Proceedings of the IEEE International,
pages 169–174.
[66] Genikomsakis, K. N., Ioakimidis, C. S., Bocquier, B., Savvidis, D., and Simic, D. (2013). Electromobility and car-
sharing/carpooling services at the University of Deusto: A preliminary exploratory survey. In 16th International IEEE
Conference on Intelligent Transportation Systems.
[67] Geum, Y., Lee, S., and Park, Y. (2014). Combining technology roadmap and system dynamics simulation to support
scenario-planning: A case of car-sharing service. Computers & Industrial Engineering, 71:37–49.
[68] Green Car Reports (2016). Daimler’s Car2Go carsharing service signs up 78,000 China users in 2
months. daimlers-car2go-carsharing-service-signs-up-78000-china-users-
in-2-months. Last accessed April 4th, 2017.
[69] Habib, K. M. N., Morency, C., Islam, M. T., and Grasset, V. (2012). Modelling users’ behaviour of a carsharing
program: Application of a joint hazard and zero inflated dynamic ordered probability model. Transportation research
part A: policy and practice, 46(2):241–254.
[70] Hayashi, T., Hidaka, K., and Teshima, S. (2014). Carsharing and IT Enabled Services. In Global Conference (SRII),
2014 Annual SRII, pages 274–280.
[71] Hayel, Y., Quadri, D., Jim´enez, T., and Brotcorne, L. (2016). Decentralized optimization of last-mile delivery services
with non-cooperative bounded rational customers. Annals of Operations Research, 239(2):451–469.
[72] Herbawi, W., Knoll, M., Kaiser, M., and Gruel, W. (2016). An evolutionary algorithm for the vehicle relocation
problem in free floating carsharing. In Evolutionary Computation (CEC), 2016 IEEE Congress on, pages 2873–2879.
[73] Herrmann, S., Schulte, F., and Voß, S. (2014). Increasing acceptance of free-floating car sharing systems using smart
relocation strategies: A survey based study of Car2go Hamburg. In Computational Logistics, pages 151–162. Springer.
[74] Hildebrandt, B., Hanelt, A., Piccinini, E., Kolbe, L., and Nierobisch, T. (2015). The value of IS in Business Model
Innovation for sustainable mobility services-the case of carsharing. In Wirtschaftsinformatik Proceedings 2015.
[75] Huwer, U. (2004). Public transport and car-sharing–benefits and effects of combined services. Transport Policy,
[76] ICT for City Logistics and Enterprises (2016). ICE Lab mission and vision. Last accessed
April 4th, 2017.
[77] Ion, L., Cucu, T., Boussier, J.-M., Teng, F., and Breuil, D. (2009). Site selection for electric cars of a car-sharing
service. In Proceedings of EVS24: International Battery, Hybrid and Fuel Cell Electric Vehicle Symposium & Exhibition.
[78] Jian, S., Rashidi, T. H., Wijayaratna, K. P., and Dixit, V. V. (2016). A spatial hazard-based analysis for modelling
vehicle selection in station-based carsharing systems. Transportation Research Part C: Emerging Technologies, 72:130–
[79] Jorge, D., Barnhart, C., and de Almeida Correia, G. H. (2015a). Assessing the viability of enabling a round-trip
carsharing system to accept one-way trips: Application to Logan Airport in Boston. Transportation Research Part C:
Emerging Technologies, 56:359 – 372.
[80] Jorge, D. and Correia, G. (2013). Carsharing systems demand estimation and defined operations: a literature review.
European Journal of Transport and Infrastructure Research, 13(3):201–220.
[81] Jorge, D., Correia, G., and Barnhart, C. (2012). Testing the validity of the MIP approach for locating carsharing
stations in one-way systems. Procedia-Social and Behavioral Sciences, 54:138–148.
[82] Jorge, D., Correia, G. H., and Barnhart, C. (2014). Comparing optimal relocation operations with simulated relocation
policies in one-way carsharing systems. IEEE Transactions on Intelligent Transportation Systems, 15(4):1667–1675.
[83] Jorge, D., Molnar, G., and de Almeida Correia, G. H. (2015b). Trip pricing of one-way station-based carsharing
networks with zone and time of day price variations. Transportation Research Part B: Methodological, pages –.
[84] Kaspi, M., Raviv, T., and Tzur, M. (2014). Parking reservation policies in one-way vehicle sharing systems. Trans-
portation Research Part B: Methodological, 62:35–50.
[85] Kato, H., Inagi, A., and Igo, T. (2013). Potential choices of travel mode including carsharing and carsharing member-
ship: Evidences from four Japanese Cities. Journal of the Eastern Asia Society for Transportation Studies, 10(0):630–646.
[86] Kek, A., Cheu, R., and Chor, M. (2006). Relocation simulation model for multiple-station shared-use vehicle systems.
Transportation Research Record: Journal of the Transportation Research Board, 1986:81–88.
[87] Kek, A. G., Cheu, R. L., Meng, Q., and Fung, C. H. (2009). A decision support system for vehicle relocation operations
in carsharing systems. Transportation Research Part E: Logistics and Transportation Review, 45(1):149–158.
[88] Kent, J. L. (2014). Carsharing as active transport: What are the potential health benefits? Journal of Transport &
Health, 1(1):54–62.
[89] Khan, M., Machemehl, R., and Zhang, Z. (2014). The impact of land-use variables on free-floating carsharing vehicle
rental choice and parking duration. Presented at the 2014 Workshops on Big Data and Urban Informatics at the
University of Illinois, Chicago.
[90] Kim, D., Ko, J., and Park, Y. (2015). Factors affecting electric vehicle sharing program participants’ attitudes about
car ownership and program participation. Transportation Research Part D: Transport and Environment, 36:96–106.
[91] King, C., Griggs, W., Wirth, F., and Shorten, R. (2013). Using a car sharing model to alleviate electric vehicle range
anxiety. In Proceedings of the Sixteenth Yale Workshop on Adaptive and Learning Systems, pages 130–135.
[92] Kopp, J., Gerike, R., and Axhausen, K. (2015). Do sharing people behave differently? An empirical evaluation of the
distinctive mobility patterns of free-floating car-sharing members. Transportation.
[93] Krueger, R., Rashidi, T. H., and Rose, J. M. (2016). Preferences for shared autonomous vehicles. Transportation
research part C: emerging technologies, 69:343–355.
[94] Kumar, V. P. and Bierlaire, M. (2012). Optimizing locations for a vehicle sharing system. In Swiss Transport Research
Conference.(Cited on pages 2, 8, and 24.).
[95] Laporte, G., Meunier, F., and Wolfler Calvo, R. (2015). Shared mobility systems. 4OR, 13:341–360.
[96] Le Vine, S. and Polak, J. (2017). The impact of free-floating carsharing on car ownership: Early-stage findings from
London. Transport Policy.
[97] Lee, D., Quadrifoglio, L., Meloni, I., et al. (2016). Discovering relationships between factors of round-trip car sharing
by using association rules approach. Procedia Engineering, 161:1282–1288.
[98] Lee, J., Nah, J., Park, Y., and Sugumaran, V. (2011). Electric car sharing service using mobile technology. In
International Conference on Information Resources Management.
[99] Li, X., Ma, J., Cui, J., Ghiasi, A., and Zhou, F. (2016). Design framework of large-scale one-way electric vehicle
sharing systems: A continuum approximation model. Transportation Research Part B: Methodological, 88:21–45.
[100] Lindloff, K., Pieper, N., Bandelow, N. C., and Woisetschl¨ager, D. M. (2014). Drivers of carsharing diffusion in
Germany: an actor-centred approach. International Journal of Automotive Technology and Management, 14(3):217–
[101] Lopes, M. M., Martinez, L. M., and de Almeida Correia, G. H. (2014). Simulating carsharing operations through
agent-based modelling: An application to the City of Lisbon, Portugal. Transportation Research Procedia, 3(0):828 –
[102] Loukopoulos, P., Jakobsson, C., Grling, T., Schneider, C. M., and Fujii, S. (2004). Car-user responses to travel
demand management measures: goal setting and choice of adaptation alternatives. Transportation Research Part D:
Transport and Environment, 9(4):263 – 280.
[103] Lu`e, A., Colorni, A., Nocerino, R., and Paruscio, V. (2012). Green move: An innovative electric vehicle-sharing
system. Procedia-Social and Behavioral Sciences, 48:2978–2987.
[104] Mannan, M. (2001). Car sharing - an (ITS) application for tomorrows mobility. In Systems, Man, and Cybernetics,
2001 IEEE International Conference on, volume 4, pages 2487–2492 vol.4.
[105] Manzini, R., Pareschi, A., et al. (2012). A decision-support system for the car pooling problem. Journal of Trans-
portation Technologies, 2(02):85.
[106] Marco, K., Eric, K., Eric, T., and Cyrille, G. (2011). Estimating potential demand for Autolib - a new transport
system for Paris. In International Choice Modelling Conference, 2011.
[107] Marouf, M., Pollard, E., and Nashashibi, F. (2014). Automatic parallel parking and platooning to redistribute electric
vehicles in a car-sharing application. In Intel ligent Vehicles Symposium Proceedings, 2014 IEEE, pages 486–491.
[108] Martens, K., Sierzchula, W., and Pasman, S. (2011). Broadening the market for carshare? Results of a pilot project
in the netherlands. In Transportation Research Board 90th Annual Meeting, volume 11-0615.
[109] Martin, E. and Shaheen, S. (2011a). The impact of carsharing on public transit and non-motorized travel: an
exploration of North America in carsharing survey data. Energies, 4(11):2094–2114.
[110] Martin, E., Shaheen, S. A., and Lidicker, J. (2010). Impact of carsharing on household vehicle holdings. Transportation
Research Record: Journal of the Transportation Research Board, 2143(1):150–158.
[111] Martin, E. W. and Shaheen, S. A. (2011b). Greenhouse gas emission impacts of carsharing in North America.
Intelligent Transportation Systems, IEEE Transactions on, 12(4):1074–1086.
[112] Masuch, N., Lutzenberger, M., Ahrndt, S., Heßler, A., and Albayrak, S. (2011). A context-aware mobile accessi-
ble electric vehicle management system. In Computer Science and Information Systems (FedCSIS), 2011 Federated
Conference on, pages 305–312.
[113] Matsoukis, E. (2007). Congestion charging/carpooling/carsharing: Solutions for reducing traffic congestion, pollution,
and vehicle kms traveled. In Proceedings of the Greek Road Federation conference, pages 8–9.
[114] Millard-Ball, A. (2005). Car-Sharing: Where and how it succeeds, volume 108. Transportation Research Board.
[115] Monteiro, P., Carreto, C., and Peixoto, J. (2010). Comparing GPS traces in car sharing systems. In Information
Systems and Technologies (CISTI), 2010 5th Iberian Conference on, pages 1–6.
[116] Morency, C., Tepanier, M., Agard, B., Martin, B., and Quashie, J. (2007). Car sharing system: what transaction
datasets reveal on users’ behaviors. In Intelligent Transportation Systems Conference, 2007. ITSC 2007. IEEE, pages
[117] uller, J. and Bogenberger, K. (2015). Time series analysis of booking data of a free-floating carsharing system in
Berlin. Transportation Research Procedia, 10:345–354.
[118] Nair, R. and Miller-Hooks, E. (2014). Equilibrium network design of shared-vehicle systems. European Journal of
Operational Research, 235(1):47–61.
[119] Nobis, C. (2006). Carsharing as key contribution to multimodal and sustainable mobility behavior: Carsharing in
Germany. Transportation Research Record: Journal of the Transportation Research Board, 1986(1):89–97.
[120] Nourinejad, M. and Roorda, M. (2015). Carsharing operations policies: a comparison between one-way and two-way
systems. Transportation, 42(3):497–518.
[121] Nourinejad, M. and Roorda, M. J. (2014). A dynamic carsharing decision support system. Transportation Research
part E: logistics and transportation review, 66:36–50.
[122] Nourinejad, M., Zhu, S., Bahrami, S., and Roorda, M. J. (2015). Vehicle relocation and staff rebalancing in one-way
carsharing systems. Transportation Research Part E: Logistics and Transportation Review, 81:98–113.
[123] Ohta, H., Fujii, S., Nishimura, Y., and Kozuka, M. (2013). Analysis of the acceptance of carsharing and eco-cars in
Japan. International Journal of Sustainable Transportation, 7(6):449–467.
[124] Osterwalder, A. and Pigneur, Y. (2010). Business model generation: a handbook for visionaries, game changers, and
challengers. John Wiley & Sons.
[125] Perboli, G. (2016). GUEST-OR. linking lean business and OR. In Workshop of the 28th European Conference on
Operation Research, Poznan, Poland.
[126] Perboli, G., De Marco, A., Perfetti, F., and Marone, M. (2014). A new taxonomy of smart city projects. Transportation
Research Procedia, 3:470–478.
[127] Perboli, G., Ferrero, F., Musso, S., and Vesco, A. (2017). Business models and tariff simulation in car-sharing services.
Transportation Research Part A: Policy and Practice.
[128] Perboli, G. and Gentile, R. (2015). GUEST: From lean startup to lean business. a manifesto. CIRRELT Webinar
[129] Perboli, G., Ghirardi, M., Gobbato, L., and Perfetti, F. (2015a). Flights and their economic impact on the airport
catchment area: an application to the Italian tourist market. Journal of Optimization Theory And Applications, 164:1109–
[130] Perboli, G., Gobbato, L., and Maggioni, F. (2015b). A progressive hedging method for the multi-path travelling
salesman problem with stochastic travel times. IMA Journal of Management Mathematics, page dpv024.
[131] Rabbitt, N. and Ghosh, B. (2013). A study of feasibility and potential benefits of organised car sharing in Ireland.
Transportation Research Part D: Transport and Environment, 25:49–58.
[132] Rabbitt, N. and Ghosh, B. (2016). Economic and environmental impacts of organised car sharing services: A case
study of Ireland. Research in Transportation Economics, 57:3–12.
[133] Reisman, a. (2005). Transfer of technologies:a cross-disciplinary taxonomy. Omega, 33:189–202.
[134] Rickenberg, T. A., Gebhardt, A., and Breitner, M. H. (2013). A decision support system for the optimization of car
sharing stations. In ECIS, page 207.
[135] Ruhrort, L., Steiner, J., Graff, A., Hinkeldein, D., and Hoffmann, C. (2014). Carsharing with electric vehicles in the
context of users mobility needs–results from user-centred research from the BeMobility field trial (Berlin). International
Journal of Automotive Technology and Management, 14(3):286–305.
[136] Santos, G. and Correia, G. (2015). A MIP model to optimize real time maintenance and relocation operations in
one-way carsharing systems. Transportation Research Procedia, 10:384–392.
[137] Schaefers, T. (2013). Exploring carsharing usage motives: A hierarchical means-end chain analysis. Transportation
Research Part A: Policy and Practice, 47:69–77.
[138] Schaefers, T., Lawson, S., and Kukar-Kinney, M. (2015). How the burdens of ownership promote consumer usage of
access-based services. Marketing Letters.
[139] Schm¨oller, S. and Bogenberger, K. (2014). Analyzing external factors on the spatial and temporal demand of car
sharing systems. Procedia - Social and Behavioral Sciences, 111(0):8 – 17.
[140] Schm¨oller, S., Weikl, S., M¨uller, J., and Bogenberger, K. (2015). Empirical analysis of free-floating carsharing usage:
The Munich and Berlin case. Transportation Research Part C: Emerging Technologies, 56:34–51.
[141] Seign, R. and Bogenberger, K. (2013). Prescriptions for the successful diffusion of carsharing with electric vehicles.
In Conference on Future Automotive Technology Focus Electromobility.
[142] Seign, R., Sch¨ußler, M., and Bogenberger, K. (2015). Enabling sustainable transportation: The model-based de-
termination of business/operating areas of free-floating carsharing systems. Research in Transportation Economics,
[143] Shaheen, S. A., Chan, N. D., and Micheaux, H. (2015). One-way carsharing’s evolution and operator perspectives
from the Americas. Transportation, 42(3):519–536.
[144] Shaheen, S. A. and Cohen, A. P. (2007). Growth in worldwide carsharing: An international comparison. Transporta-
tion Research Record: Journal of the Transportation Research Board, 1992(1):81–89.
[145] Shaheen, S. A. and Cohen, A. P. (2013). Carsharing and personal vehicle services: worldwide market developments
and emerging trends. International Journal of Sustainable Transportation, 7:5–34.
[146] Shaheen, S. A., Cohen, A. P., and Chung, M. S. (2009). North American carsharing. Transportation Research Record:
Journal of the Transportation Research Board, 2110(1):35–44.
[147] Shaheen, S. A., Cohen, A. P., and Roberts, J. D. (2006). Carsharing in North America: Market growth, current
developments, and future potential. Transportation Research Record: Journal of the Transportation Research Board,
[148] Shaheen, S. A., Mallery, M. A., and Kingsley, K. J. (2012). Personal vehicle sharing services in North America.
Research in Transportation Business & Management, 3:71–81.
[149] Sioui, L., Morency, C., and Tr´epanier, M. (2013). How carsharing affects the travel behavior of households: a case
study of Montr´eal, Canada. International Journal of Sustainable Transportation, 7(1):52–69.
[150] Steiner, J., Wappelhorst, S., and Graff, A. (2014). Free-floating e-carsharing: Integration in public transport without
range problems. In European Transport Conference 2014.
[151] Steininger, K. W. and Bachner, G. (2014). Extending car-sharing to serve commuters: An implementation in Austria.
Ecological Economics, 101:64–66.
[152] Stillwater, T., Mokhtarian, P. L., and Shaheen, S. A. (2009). Carsharing and the built environment. Transportation
Research Record: Journal of the Transportation Research Board, 2110(1):27–34.
[153] Tadei, R., Perboli, G., and Perfetti, F. (2017). The multi-path traveling salesman problem with stochastic travel
costs. EURO Journal on Transportation and Logistics, 6:3–23.
[154] Tadei, R., Perboli, G., Ricciardi, N., and Baldi, M. M. (2012). The capacitated transshipment location problem with
stochastic handling utilities at the facilities. International Transactions in Operational Research, 19:789–807.
[155] Terrien, C., Maniak, R., Chen, B., and Shaheen, S. (2016). Good practices for advancing urban mobility innovation:
A case study of one-way carsharing. Research in Transportation Business & Management, 20:20–32.
[156] Uesugi, K., Mukai, N., and Watanabe, T. (2007). Optimization of vehicle assignment for car sharing system. In
Apolloni, B., Howlett, R., and Jain, L., editors, Knowledge-Based Intelligent Information and Engineering Systems,
volume 4693 of Lecture Notes in Computer Science, pages 1105–1111. Springer.
[157] Vine, S. L., Lee-Gosselin, M., Sivakumar, A., and Polak, J. (2014). A new approach to predict the market and impacts
of round-trip and point-to-point carsharing systems: Case study of London. Transportation Research Part D: Transport
and Environment, 32:218 – 229.
[158] Wagner, S., Brandt, T., and Neumann, D. (2016). In free float: Developing business analytics support for carsharing
providers. Omega, 59:4–14.
[159] Wappelhorst, S., Sauer, M., Hinkeldein, D., Bocherding, A., and Gla, T. (2014). Potential of electric carsharing in
urban and rural areas. Transportation Research Procedia, 4(0):374 – 386.
[160] Weikl, S. and Bogenberger, K. (2013). Relocation strategies and algorithms for free-floating car sharing systems.
Intelligent Transportation Systems Magazine, IEEE, 5(4):100–111.
[161] Weikl, S. and Bogenberger, K. (2015). A practice-ready relocation model for free-floating carsharing systems with
electric vehicles–mesoscopic approach and field trial results. Transportation Research Part C: Emerging Technologies,
[162] Wikipedia (2016). Car2Go. Last accessed April 4th, 2017.
[163] Xu, J.-X. and Lim, J. (2007). A new evolutionary neural network for forecasting net flow of a car sharing system. In
Evolutionary Computation, 2007. CEC 2007. IEEE Congress on, pages 1670–1676.
[164] Zhang, W., Guhathakurta, S., Fang, J., and Zhang, G. (2015). Exploring the impact of shared autonomous vehicles
on urban parking demand: An agent-based simulation approach. Sustainable Cities and Society, 19:34–45.
[165] Zheng, J., Scott, M., Rodriguez, M., Sierzchula, W., Platz, D., Guo, J. Y., and Adams, T. M. (2009). Carsharing in a
university community. Transportation Research Record: Journal of the Transportation Research Board, 2110(1):18–26.
[166] Zhou, B. and Kockelman, K. M. (2011). Opportunities for and impacts of carsharing: A survey of the Austin, Texas
market. International Journal of Sustainable Transportation, 5(3):135–152.
[167] Zipcar (2015). Les villes Zipcar. Last accessed April 4th, 2017.
[168] Zoepf, S. M. and Keith, D. R. (2016). User decision-making and technology choices in the us carsharing market.
Transport Policy, 51:150–157.
47 %
19 %
19 %
15 %
Not applicable
Figure 2: Modes
Figure 3: Number of papers referring to modes of car-sharing services
Figure 4: Number of papers referring to types of engine in car-sharing services
Figure 5: Cross analysis of engine and modes
12 %
28 %
Fleet management
60 %
Business and service
Figure 6: Optimization objectives
Figure 7: Number of papers referring to the objectives of optimization
Figure 8: Cross analysis of optimization objectives and modes
Figure 9: Cross analysis of optimization objectives and types of engine
25 %
Operational and real time
16 %
59 %
Figure 10: Time horizon analysis
Figure 11: Number of papers referring to the time horizon analysis
Figure 12: Cross analysis of time horizon and modes
Figure 13: Cross analysis of time horizon and engine type
61 %
Statistic analysis
23 %
7 %
Stochastic optimization
9 %
Combinatorial optimization
Figure 14: Methodology analysis
Figure 15: Number of papers referring to the methodology analysis
Figure 16: Cross analysis of methodologies and modes
Figure 17: Cross analysis of methodologies and types of engine
Figure 18: Cross analysis of methodologies and objectives of the optimization
Figure 19: Cross analysis of methodologies and time horizons of the decisions
... Carsharing organizations operating throughout the world in various forms have sprung up. Ferrero, et al. found that 69% of the analyzed papers are referred to B2C (Business-to-Consumer) one-way mode [5]. B2C model here demonstrates that the ownership of the shared car belongs to the company. ...
... According to different research objects, the existing publication in the field of carsharing can be roughly divided into 4 aspects: the impact of carsharing system on the environment, market dynamics, daily operation management, and customer behavior [6]. By looking back on the last fifteen years of research, a research trend was found that the focus of carsharing topic had shifted from infrastructure design to dealing with the customer behaviors and the customer acceptance [5]. For carsharing, as a supplement to the transportation system, customer satisfaction is of greater significance. ...
... Transportation has been identified as a critical need by survivors of IPV in order from them to gain economic independence (Hamby et al., 2015;Resko and Mendoze, 2012). For this reason, shared mobility programs like ridesource and car-share programs may be a feasible solution to help survivors currently residing in shelter mitigate the transportation barriers they experience (Standing et al., 2019;Ferrero et al., 2018). This study sought to explore the feasibility of ridesource and car-share services as possible solutions to transportation barriers for survivors of IPV by gathering perspectives of providers who work with shelter-based survivors using scenario planning. ...
... Shared mobility services, like ridesource and car-share programs have been identified as a possible solution for transportation barriers identified by vulnerable communities (Ferrero et al., 2018;Standing et al., 2019). Ridesource programs involve individuals who are employed by a ridesource company using their personal vehicle to offer rides on an on-demand basis. ...
Introduction Survivors of intimate partner violence (IPV) experience transportation disadvantages due to financial constraints often caused by their abusive partners. Shared mobility services like ridesource and car-share programs may be a feasible solution to transportation barriers for survivors to access needed resources like employment and healthcare. Methods This study presented scenarios to providers who work with IPV shelter residents to determine if having access to an on-demand ridesource and car-share service would mitigate survivors’ transportation challenges. Ten participants were interviewed using scenario planning and an accompanying semi-structured interview guide. Participants were presented with two scenarios: 1) imagine if there was a ridesource service which operated at the shelter and 2) imagine if there was a car-share service placed at the shelter. Thematic analysis was used to develop themes around the utilization of the ridesource and car-share services, barriers and concerns that may need to be addressed, and policy recommendations that would make the program successful. Results Participants agreed that with the proper policies in place, ridesource and car-share services could greatly aid survivors in getting the transportation they need to regain their independence both socially and economically. Conclusion Exploring the potential for shared mobility services to be a viable transportation solution can aid social service providers, city planners, and shared mobility companies in understanding how to use the services to benefit not only survivors of IPV, but also other members within communities who may experience transportation disadvantage.
... The sharing of vehicles (bikes, scooters, and cars) can contribute to the solution and has been an active field of research with numerous publications [2]. Firstly, it reduces the number of vehicles owned in total, thereby reducing the demand for raw materials. ...
Full-text available
In many developed cities around the world, vehicle sharing is becoming an increasingly popular form of green transportation. While such services are associated with lower emissions and easier mobility, their management poses a significant challenge. In this paper, we examine a dataset collected in Barcelona during the months of august and september 2020 in order to investigate relocation strategies and user clustering. By proposing a neighborhood area split and relating it to user demand, we propose two different areas based on majority demand and users’ requests and provide interpretations of both. We then aim to identify groups of similar users using a variant of Recency Frequency Monetary/Duration (RFM or RFD) clustering that extends to GPS coordinates of voyages in order to differentiate scores based on economic and geographical factors; furthermore, a user-based clustering approach was used to maximize client preferences. As a result of our analysis, the sharing company may be able to make more informed decisions regarding where to focus its resources. In fact, we find that the majority of the demand is concentrated in an area that represents 7.47 percent of the city’s area. Additionally, we propose a discount-based approach in order to influence the user’s behavior in parking the vehicle where it is most needed.
... Car-sharing services are systems that give the possibility of renting a motor vehicle for a short time via a website or mobile application. There are four main types of car sharing [10,11]: ...
Full-text available
Car-sharing systems, i.e., automatic, short-time car rentals, are among the solutions of the new mobility concept, which in recent years has gained popularity around the world. With the growing interest in services in society, their demands for the services offered to them have also increased. Since cars play a key role in car-sharing services, the fleet of vehicles should be properly adapted to the needs of customers using the systems. Due to the literature gap related to the procedure of proper selection of vehicles for car sharing and the market need for car-sharing service operators, this work has been devoted to the selection of car models for car sharing from the perspective of users constantly using the systems (regular users). This paper considered the case of the Polish who are constantly using car-sharing service systems. Vehicle selection was classified as a multi-faceted, complex problem, which is why one of the ELECTRE III multi-criteria decision support methods was used for this study. This study focused on the classification of vehicles from the user’s perspective. Twelve modern and most popular car models in 2021 with internal combustion, electric and hybrid engines were considered. The results indicate that the best choice from the point of view of regular customers is large cars (representing vehicle classes C and D), with a large luggage compartment capacity, the highest possible ratio of engine power to vehicle weight, and the ratio of engine power to energy consumption. Importantly, small urban vehicles, which ideologically should be associated with car-sharing services due to occupying as little urban space as possible, were classified as the worst in the ranking. The results support car-sharing operators during the process of completing or upgrading their vehicle fleets.
His study aims to explain the problems of online transportation in big cities such as Jakarta, Medan, and Surabaya in Indonesia. The method used in this research is qualitative analysis. Furthermore, this study uses Nvivo-12 Plus Software to analyze qualitative data and present cross-tabulation and Visual analysis. There are five stages in using the Nvivo application in this study: data collection, data coding, data classification, and data presentation. The data that has been processed with Nvivo-12 Plus is then carried out with qualitative analysis. The data sources in this study were from well-known local media websites. The findings in this study indicate that the highest level of problems for online transportation is DKI Jakarta, with seven problems: congestion problems, fictional orders, violence, tariff problems, licensing, quotas, and zoning, Ranked second in Surabaya with problems of thought order, tariffs, permits, quotas, zones, and permits, Then the third rank is Medan City, with licensing problems, quota problems, and the controversy over the Minister of Transportation Regulation No. 108 of 2017. This study only analyzes the problems that arise due to online transportation by comparing them in three big cities. Further research is needed regarding the government's response in providing solutions to the problems caused by the arrival of online transportation.
Carsharing is an important element in the transition to a more sustainable transport system. In contrast to the widely studied business-to-consumer (B2C) market segment, studies on Business-to-business (B2B) carsharing and its impacts are limited. This paper analyses the factors that drive and hinder organisations from using B2B carsharing for their employees’ work trips. The paper takes a case study approach, analysing B2B carsharing use for local work-trips at seven employers in Gothenburg, Sweden, based on interviews with employers, property owners, and carsharing operators. Our results indicate that carsharing services can contribute to reducing the employers’ costs for work-trips, ensuring the sustainability and safety of the employee’s work-trips, as well as increasing the employer’s workplace attractivity. However, carsharing services are currently used only to a very limited extent. Obstacles that limit a greater use of carsharing services are employers’ lack of data on the employees’ work travel patterns, little economic incentives for employers, as well as parking management and travel policies that favour the use of the private car.
Global adoption of electric vehicles (EVs) faces many challenges such as range anxiety, high cost of EVs, and inadequate charging infrastructure. EV-sharing platforms resolve such concerns by setting up an optimal configuration for charging infrastructure and optimizing the charging decisions for depleted EVs. These platforms manage the vehicles’ flow to different charging stations and decide when and to what energy level the depleted vehicles should be recharged. Station-based platforms are one of the mainstream vehicle sharing systems where the customer picks-up and drops-off the vehicle at the designated stations. If a vehicle’s battery energy level falls below a threshold after completing the customer trip, it is charged either partially or fully at the charging station. This study addresses various operational and strategic decisions (such as the number of chargers, vehicle repositioning, and partial charging policy) for a one-way station-based EV-sharing platform using a stylized three-stage analytical framework. We use vehicle dynamics to model the EV powertrain and regenerative braking under different traffic conditions and simulate them using AVL CRUISE™. We model the platform operations using an open queuing network and provide a mixed-integer non-linear optimization program using inputs from the queuing network and vehicle dynamics simulation. We also provide a bound-based heuristic to solve this NP-hard optimization problem. We generate various managerial insights for an efficient implementation of the partial charging policy for EV-sharing platforms. The increase in the partial charging probability (the fraction of depleted vehicles charged partially) reduces the effective charging demand, resulting in fewer chargers and a higher profit. On the other hand, if we increase the target battery energy level for partial charging, the platform’s profit decreases due to higher effective charging demand dominating the benefits of lower charging frequency of vehicles.
Full-text available
Short-term, automated car rental services, i.e., car sharing, are a solution that has been improving in urban transportation systems over the past few years. Due to the intensive expansion of the systems, service providers face increasing challenges in their competitiveness. One of them is to meet the customer expectations for the fleet of vehicles offered in the system. Although this aspect is noted primarily in the literature review on fleet optimization and management, there is a gap in research on the appropriate selection of vehicle models. In response, the article aimed to identify the vehicles best suited for car-sharing systems from the customer’s point of view. The selection of suitable vehicles was treated as a multi-criteria decision-making issue; therefore, the study used ELECTRE III—one of the multi-criteria decision-making methods. The work focuses on researching the opinions of users who rarely use car-sharing services in Poland. The most popular car models in 2021, equipped with internal combustion, hybrid, and electric engines, were selected for the analysis. The results indicate that the best suited cars are relatively large, spacious, and equipped with electric drive and represent the D segment of vehicles in Europe. In addition, these vehicles are to be equipped with a powerful engine, a spacious boot, and a fast battery charging time. Interestingly, small city cars, so far associated with car sharing, ranked the worst in the classification method. In addition, factors such as the warranty period associated with the quality of the vehicles, or the number of car doors, are not very important to users. The results support car-sharing operators in the process of selecting or modernizing a fleet of vehicles.
The current revolutions of automation, electrification, and sharing are reshaping the way we travel, with broad implications for future mobility management. While much uncertainty remains about how these disruptive technologies would exactly impact demand for future mobility and enhancement of transportation supply, it is clear that innovative demand management is equally important as smart supply technology development in solving worsening traffic problems in big cities. In this work, we will discuss the significances, opportunities, and challenges of demand management in the era of smart transportation. Innovative ways of travel demand management for road transportation, public transit, and smart mobility are described, including tradable travel credit schemes for road congestion mitigation, revenue-preserving and Pareto-improving strategies for peak-hour transit demand management, and a novel reward scheme integrated with surge pricing in a ride-sourcing market.
Conference Paper
Full-text available
Within the framework of the project CROME (Cross-border Mobility for Electric Vehicles) there is an accompanying scientific research on a fleet test of about 100 Battery Electric Vehicles (BEV) in the French-German border region taking place. A user acceptance study is accomplished with the focus on transnational trips. The observed BEV are predominantly company fleet vehicles and are used by several persons. This increases the potential number of BEV users taking part in the accompanying research activities of the fleet test significantly. During the survey period cross-border mobility with BEV has hardly been possible due to different standards concerning hardware and software components, especially concerning components of the charging infrastructure. The idea of CROME is to demonstrate seamless cross-border mobility between France and Germany and to give recommendations to the European standardization process on infrastructure components. Key findings of the first online questionnaire filled out by BEV users and fleet managers indicate that the acceptance for BEV is highest for people that live in communities with less than 5,000 inhabitants, with two or more cars in the household, a higher annual mileage and a high commuting distance.
Full-text available
The paper considers an important new and growing business in sustainable transportation, car-sharing services. This is, to our knowledge, the first comprehensive analysis of car-sharing services from the business model point of view. Specifically, we apply and introduce a standard and reproducible way to compare the business models of car-sharing companies. Our analysis results show that a crucial issue in defining car-sharing services is the creation of customized tariff plans. Thus, as a second contribution of our paper, we introduce a specific solution based on Monte Carlo simulation. This tool simulates the existing price and tariff policies or the introduction of new ones for different profiles of car-sharing users, according to different mobility needs and the traffic congestion of the urban area. As an example, we use our methodology to provide an in-depth description of the situation in the city of Turin, Italy.
Full-text available
To meet the binding annual Green House Gas (GHG) emission targets according to the European Union (EU) Effort Sharing Decision by 2020, transport related CO2 emissions are required to be reduced in Ireland. Internationally Car Sharing (CS) has been identified as a means of reducing car dependency and travel related CO2 emissions while still allowing users the benefits of car access. Rabbitt & Ghosh (2013) established that CSS adoption would be beneficial to Dublin & the benefits may extend to Ireland. This study extended the work by providing a detailed framework of evaluating economic and environmental impacts of joining CSS for both individuals and the collective society. The study also expanded the estimation of travel behaviour changes from the users in Dublin city to the potential users in the entire country of Ireland. The analysis identified that car owners who travel predominantly on alternative modes, could make significant travel cost and CO2 emission savings through joining CSS. The long-term benefits included a slower growth rate of car-ownership and in turn generating significantly high CO2 savings of 84 kt for Dublin and up to 229 kt for Ireland with some policy and financial support.
Full-text available
To maximize private vehicle usage efficiency and alleviate urban congestion, many studies and actual field operations on vehicle sharing have been done since the mid ‘90s. The classic carsharing system, which is known as a round-trip, is operated out of fixed stations so that customers can pick up and drop off the vehicle at the same station. Although many private carsharing companies offer one-way or free-floating trips currently for customers’ convenience, studying characteristics of round-trip carsharing is still significant in making cost-beneficial, fuel-efficient, and less-congested driving environments in urban areas. The main objective of this research is a comprehensive analysis for discovering relationships between critical factors of round-trip carsharing operations based on the city of Cagliari, Italy, by analysing data retrieved from the local carsharing providing company, PlayCar, with the association rules approach. This paper investigates round-trip carsharing behaviour characteristics from various angles, including demand analysis of reservation by hourly and daily manner, geographic analysis, and connectivity to public transportations. The association rules technique was used to discover the relationships between the characteristics and understand their attributes.
Car-sharing offers an environmentally sustainable, socially responsible and economically feasible mobility form in which a fleet of shared-use vehicles in a number of locations can be accessed and used by many people on as-needed basis at an hourly or mileage rate. To ensure its sustainability, car-sharing operators must be able to effectively manage dynamic and uncertain demands, and make the best decisions on strategic vehicle allocation and operational vehicle reallocation both in time and space to improve their profits while keeping costs under control. This paper develops a stochastic optimization method to optimize strategic allocation of vehicles for one-way car-sharing systems under demand uncertainty. A multi-stage stochastic linear programming model is developed and solved for use in the context of car-sharing. A seven-stage experimental network study is conducted. Numerical results and computational insights are discussed.
Conference Paper
We consider an offline car-sharing assignment problem with flexible drop-offs, in which n users (customers) present their driving demands, and the system aims to assign the cars, initially located at given locations, to maximize the number of satisfied users. Each driving demand specifies the pick-up location and the drop-off location, as well as the time interval in which the car will be used. If a user requests several driving demands, then she is satisfied only if all her demands are fulfilled. We show that minimizing the number of vehicles that are needed to fulfill all demands is solvable in polynomial time. If every user has exactly one demand, we show that for given number of cars at locations, maximizing the number of satisfied users is also solvable in polynomial time. We then study the problem with two locations A and B, and where every user has two demands: one demand for transfer from A to B, and one demand for transfer from B to A, not necessarily in this order. We show that maximizing the number of satisfied users is NP-hard, and even APX-hard, even if all the transfers take exactly the same (non-zero) time. On the other hand, if all the transfers are instantaneous, the problem is again solvable in polynomial time.
Free-floating carsharing is a rapidly growing urban mobility service. It has emerged at commercial scale more recently than traditional ‘round-trip’ carsharing, and at present its growth trajectory is steeper. The evidence base regarding its impacts on sustainable transport indicators is, however, less well-developed. This issue is topical for a variety of reasons, including the importance of public policy to the success of this form of carsharing. The research objective of this study is to establish the early-stage impact of free-floating carsharing on private car ownership. We report findings from a point in time three months following the initiation of a free-floating carsharing service in London (UK). We investigate characteristics of FFCS users that are associated with having one's car ownership impacted, as well as the distinction between deterrence of increased car ownership and sale/disposal of a previously owned private car. We find that 37% (n=347; 95% confidence interval:±5%) of users indicate that free-floating carsharing has impacted their ownership of private cars. Of this 37%, a large majority (83%) indicated that the mechanism of impact was that they decided not to buy a car that they otherwise would have purchased. 11% reported that they had disposed of a car in the past three months, and 6% stated that they will sell a private car within the next three months. The average income and education level of users are both higher than for the general population. Within the population of service-users, multivariate analysis demonstrates that, net of confounding effects, heavier (more-frequent) service-users are more likely to indicate impacts on car ownership, and that being highly-educated and higher-income than other users were both (independently) associated with maintaining one's car ownership level. Additional findings are presented that relate car ownership impacts to further demographic characteristics as well as behavioural indicators. Our findings should be interpreted to pertain to the ‘early adopter’ cohort of FFCS users, and as free-floating carsharing services mature and grow further research will be needed to ascertain how user characteristics, behaviour and impacts are evolving.
Conference Paper
Car sharing holds a promise of reducing traffic congestion and pollution in cities as well as of boosting the use of public transport when used as a last-mile solution in a multimodal transportation scenario. Despite this huge potential, several problems related to the deployment and operations of car sharing systems have yet to be fully addressed. In this work, we focus on station-based car sharing and we define an optimization problem for the deployment of its stations. The goal of this problem is to find the minimum cost deployment (in terms of number of stations and their capacity) that can guarantee a pre-defined level of service to the customers (in terms of probability of finding an available car/parking space). This problem combines insights from queueing theory (used to model the stochastic demand for cars/parking spaces at the stations) with a variant of the classical set covering problem. For its evaluation, we use a trace of more than 100,000 pickup and drop-off events at a free-floating car sharing service in The Netherlands, which are used to model the input demand of the car sharing system. Our results show that the proposed solution is able to strike the right balance between cost minimisation and quality of service, outperforming three alternative schemes used as benchmarks.
Transforming urban mobility requires integrating public with private services into a single transportation system. Local governments and private companies face the challenge of how to coordinate themselves. An emblematic example is one-way carsharing (shared use of a fleet of vehicles that are typically free-floating throughout an urban area). Surprisingly, good practices for public and private players innovating together remain relatively undocumented. This paper proposes a systematic and balanced public-private approach to foster transportation innovation management. We review both public policy and business management literature and build a framework to help governments and companies collaborate (organizational structures, project management processes, and profitability assessment tools). We use this framework to examine both public and private experiences through a case study analysis with five one-way carsharing services in Europe (Paris, Munich) and the United-States (San Francisco, Portland, Seattle). For each we conducted expert interviews with the government and the private operator. This paper provides recommendations for both sectors. First, public and private players should have specific organizations, separated from the core business. Second, they should co-manage innovation since pilot projects lack certainty and require risk management. Third, a new approach to value emphasizing the role of project learning and capability building is necessary.
Carsharing, as an alternative to private vehicle ownership, has spread worldwide in recent years due to its potential of reducing congestion, improving auto utilization rate and limiting the environmental impact of emissions release. To determine the most efficient allocation of resources within a carsharing program, it is critical to understand what factors affect the users’ behavior when selecting vehicles. This study attempts to investigate the importance of users’ attributes and fleet characteristics on choice set formation behavior in selecting vehicles using a Spatial Hazard Based Model (SHBM). In the SHBM model, “distance to a vehicle” is considered as the prospective decision criteria that carsharing users follow when evaluating the set of alternative vehicles. This variable is analogous to the duration in a conventional hazard-based model. In addition, user socio-demographic attributes, vehicle characteristics, land use type of the trip origin, etc., collected from the Australian carsharing company GoGet are utilized to parameterize the shape/scale/location parameter of the hazard function. A number of forms of parametric SHBMs are tested to determine the best fit to the data. The accelerated failure time model with a Log-logistic distribution was found to provide the best fit. The estimation results of the coefficients of the parameters can provide a starting point for carsharing organizations to optimize their pod locations and types of cars available at different pods to maximize usage.