The Carpooling Phenomenon: Sharing a Vehicle to Fight Traffic Congestion

The researcher incorporated some ideas, concepts, and studies with similar topic done by different authors and researchers in the preparation of this study. He carefully selected the related literature and studies suitable to the topic as a fortification of his investigation to make it clearer and more coherent to the readers.

Carpooling is a scheme in which people share a vehicle in order to reach common or nearby destinations. Despite its clear advantages in reducing costs, pollution, and time spent in finding a car park, there are still a few obstacles that prevent it from being the preferred way of transport.

Never Drive Alone

The carpooling phenomenon is a subject widely studied in the literature. It has been analyzed form various, very different points of view. Carpooling is the second most popular way of commuting, and maybe one of the least understood – a fact that probably explains the need for such a large corpus of studies in the literature.

J. Dahlgren et al. (1998) believed that high occupancy vehicle lanes (HOV) are not always more effective than general purpose lanes and they developed models to calculate the benefits gained for eliminating traffic congestion by adding HOV lanes, or by converting general purpose lanes into HOV.

G. Giuliano, D.W. Levine, R.F. Teal, Impact of high occupancy vehicle lanes on carpooling behavior, Transportation 17 (2) (1990) 159–177, Shows that there is no increase in ridesharing related with the introduction of new HOV lanes, despite the carpooling rate among commuters increases in some periods.

A.M. Bento, J.E. Hughes, D. Kaffine, Carpooling and driver responses to fuel price changes: evidence from traffic flows in Los Angeles,J. Urban Econ. 77 (2013) 41–56y, is a study about carpooling related with the economy world that examines carpooling and driver responses to fuel price changes. It shows that traffic flows in mainline lanes decrease when fuel prices increase and this effect is stronger when the presence of a HOV lane provides a substitute to driving alone.

The method of never drive alone (NVA) is to minimize the number of single occupancy vehicle (SOV). The use if this method is essential for the road users such as commuters. Carpooling communities can be classified into two categories: autonomous communities and non autonomous communities. Autonomous communities are being independent from the rest of the car drivers, and they are made by many good carpoolers offering and taking lift to many users on the other hand, non-autonomous communities, that being influenced by extra community car drivers, cannot be managed on their own. If a new carpooling service is to be realized, a good start point would be autonomous communities

The GRAAL (Green and Social) of Carpooling

GRAAL, optimizes a carpooling system not only by minimizing the number of cars needed at the city level, but also by maximizing the enjoyability of people sharing a trip. GRAAL computes the enjoyability within a set of users from crowd-sourced data, and then uses it on real world datasets to optimize a weighted linear combination of number of cars and enjoyability.

Trasarti et al. (2011) and He et al. (2012) stated that using fine mobility data, extract not only patterns of presence in locations, but also systematic routes, from GPS directions. These routes are used to find user-to-user matches and to provide ride-sharing recommendations.

Geo-social data is used in Elbery et al. (2013), recommending individuals to join their friends during trips by using home location models and users’ similarities. Twitter data is used in Frias-Martinez et al. (2012) as a complementary source of information for urban planning applications. Similarly, Correia and Viegas (2011) proposed a model to find compatible matches for traditional groups of users and also to find rides in alternative groups. In Bicocchi and Mamei (2014) the authors develop an application for car sharing by exploiting a clustering algorithm applied to labeled trajectories. Finally, Byon et al. (2013) introduces a Facebook-based carpooling, with Twitter-based traffic monitoring and Flickr-based incident reporting applications.

Furuhata (2013) argued that carpooling is often modeled as an optimization problem, where it finds several solutions. In Coppersmith et al. (2011), the problem is reduced to the chairman assignment problem (Tijdeman, 1980), while (Munkres,1957) use an instance of the transportation problem.

GRAAL is being described as a multiobjective model that optimizes carpooling recommendations for a weighted linear combination of number of cars used and total enjoyability. GRAAL takes Twitter data in input, as this contains information on spatio-temporal, text, and social dimensions of geo-located user tweets. GRAAL has the potential of saving a significant number of the cars needed, while keeping a high level of the total enjoyability.

Dynamic Carpooling in Urban Areas

EU CommissionVehicular (2007) discussed that traffic congestion is one of the main problems of most of our cities and towns, it degrades the quality of life, leading to a wide set of social, economic and environmental impacts. It calls for a great effort in studying and deploying innovative and ambitious urban transport modes to reach a less car-dependent life-style, which is one of the main causes of urban traffic congestion.

Nowadays, most of the carpooling services implement a “static” approach: when using such a service, the carpoolers have to post ridesharing requests and offers several hours before their desired departure time, and the shared ride has to be arranged before the trip starts. In addition to using communication technologies, dynamic ridesharing systems must establish a procedure that enables travelers to form ridesharing instantaneously. These elements are coordinated by a ridesharing provider, which can match rides using opportune route matching algorithms. This type of carpooling generally makes use of three recent technological advances: GPS navigation devices to determine a driver’s route and arrange the shared ride, smartphones for riders to request a ride from wherever they happen to be and social networks to establish trust and accountability between drivers and passengers. Dynamic ridesharing is a relatively new type of carpooling: it is a system where an automated process employed by a ridesharing provider matches up drivers and riders on a very short notice [9], which can range from a few minutes to a few hours before departure time.