Quantifying the Potential of Ride-Sharing using Call Description Records Blerim Cici †? , Athina Markopoulou † , Enrique Frías-Martínez ? , Nikolaos Laoutaris ? UC Irvine † , Telefonica Research(Spain) ? {bcici, athina}@uci.edu, {efm, nikos}@tid.es ABSTRACT Ride-sharing on the daily home-work-home commute can help individuals save on gasoline and other car-related costs, while at the same time reducing traffic and pollution in the city. Work in this area has typically focused on technology, usability, security, and legal issues. However, the success of any ride-sharing technology relies on the implicit assump- tion that human mobility patterns and city layouts exhibit enough route overlap to allow for ride-sharing on the first place. In this paper we validate this assumption using mobil- ity data extracted from city-wide Call Description Records (CDRs) from the city of Madrid. We derive an upper bound on the effectiveness of ride-sharing by making the simpli- fying assumption that any commuter can share a ride with any other as long as their routes overlap. We show that sim- ple ride-sharing among people having neighboring home and work locations can reduce the number of cars in the city at the expense of a relatively short detour to pick up/drop off passengers; e.g., for a 0.6 km detour, there is a 52% reduc- tion in the number of cars. Smartphone technology enables additional passengers to be picked up along the way, which can further reduce the number of cars, as much as 67%. 1. INTRODUCTION Ride-sharing is an effective way to reduce the number of cars on the streets in order to address both individual and city-wide issues. On one hand, individuals are interested in reducing the cost of their car usage and save on gasoline and other usage-based costs [2]. On the other hand, cities are interested in reducing traffic and pollution and provide incentives (e.g. reserved carpooling lanes) to encourage com- muters to share rides. In recent years, a plethora of web- or smartphone-based solutions have emerged in order to fa- cilitate intelligent traffic management [18], [17], [5], and in particular ride-sharing. Systems like carpooling.com, and eRideShare.com have attracted a few million users in Eu- rope and the US but the technology hasn’t been widely adopted yet. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. ACM HotMobile’13, February 26–27, 2013, Jekyll Island, Georgia, USA. Copyright 2013 ACM 978-1-4503-1421-3 ...$10.00. Ride-sharing systems started in the US during WW-II. These early “word-of-mouth” systems required predefined rendezvous and prior familiarity among commuters, which limited the number of neighbors a person could share a ride with. More recently, web-based solutions, such as Amovens.com, allow drivers and passengers to advertise their interest in ride-sharing, thereby increasing the chances of finding a match, but still generally require predefined ren- dezvous. Using smartphones with ride-sharing apps, such as Avego.com, allows drivers and passengers to be matched opportunistically without pre-arranged rendezvous. Such systems are promising but it is still unclear whether they will be widely adopted. Work in the area has focused on technological, usability, security and legal aspects [16] [20]. Previous research has shown that ride-sharing has economic advantage over driv- ing alone, and that is more spatially flexible and less time consuming than public transportation, but it is not sure if this advantages are strong enough to entice commuters to switch to ride-sharing; privacy and flexibility are major rea- sons why the vast majority of commuters choose to drive alone. Many believe that current technology provides in- sufficient levels of security and monitoring to allow people to travel safely with strangers; others believe that it is only an unsolved bootstrapping problem that keeps the technol- ogy from booming. Most people, however, implicitly assume that human mobility patterns and the layouts of today’s cities exhibit enough route overlap for ride-sharing to take off, once the aforementioned issues are solved. In this paper, we validate this underlying assumption, which is crucial for the success of any ride-sharing system. To this end, we use home/work locations, for a large popu- lation of a city, extracted from a CDR dataset; the inferred home and work locations are used to match people in groups that share the same car. The exact potential of ride-sharing depends on many factors, not all of which can be known at the scale of our study (e.g., behavioral traits). Our approach is to focus only on quantifiable factors and mask all other unknown factors by making the simplifying assumption that ride-sharing is constrained only by the distance of end-points and time. Therefore, our quantification is actually an upper bound of the exact potential of ride- sharing that may be constrained by additional factors. Our contributions are as follows. We consider two sce- narios for ride sharing, and for each scenario we develop an efficient algorithm to do the matching, and quantify the ben- efit of ride-sharing in terms of reduction in the number of cars. Note that, the theoretical limit of car reduction is 75%