A Carpooling Recommendation System for Taxicab Services

Lecture Notes in Computer Science, 2015

Ride-sharing schemes attempt to reduce road traffic by matching prospective passengers to drivers with spare seats in their cars. To be successful, such schemes require a critical mass of drivers and passengers. In current deployed implementations, the possible matches are based on heuristics, rather than real route times or distances. In some cases, the heuristics propose infeasible matches; in others, feasible matches are omitted. Poor ride matching is likely to deter participants from using the system. We develop a constraint-based model for acceptable ride matches which incorporates route plans and time windows. Through data analytics on a history of advertised schedules and agreed shared trips, we infer parameters for this model that account for 90% of agreed trips. By applying the inferred model to the advertised schedules, we demonstrate that there is an imbalance between riders and passengers. We assess the potential benefits of persuading existing drivers to switch to becoming passengers if appropriate matches can be found, by solving the inferred model with and without switching. We demonstrate that flexible participation has the potential to reduce the number of unmatched participants by up to 80%.

This paper deals with a combinatorial optimization problem that models situations of both dynamic ridesharing and taxi-sharing. Passengers who want to share a taxi or a ride, use an app to specify their current location, destination and further information such as the earliest departure time, the latest arrival time and the maximum cost they are willing to pay for the ride. Car owners also specify their origin, destination, the leaving time and the maximum accepted delay. Taxi drivers report their location and the time they will start and end the service. All drivers need to define a price per kilometer. The problem is to compute routes, matching requests to vehicles in such a way that ride-sharing is allowed as long as some restrictions are satisfied, such as: the capacity of the vehicle, maximum trip cost of each passenger, maximum delay, etc.

SSRN Electronic Journal, 2000

Seamless integration of ride-sharing and public transit may offer fast, reliable, and affordable transfer to and from transit stations in suburban areas thereby enhancing mobility of residents. We investigate the potential benefits of such a system, as well as the ride-matching technology required to support it, by means of an extensive computational study. 1 Introduction People all around the world use private cars to travel to work. Most of these commuter trips are singleoccupant vehicle trips. In the U.S., for example, single-occupant trips represent approximately 77% of all commuter trips [Polzin and Pisarski, 2013]; similar percentages are found in Europe [European Environment Agency, 2010]. The low vehicle occupancy rates combined with the high number of trips during peak-hours often leads to severe traffic congestion in urban areas. The resulting stress and air pollution, caused by vehicle emissions, can have serious negative health effects. To reduce the negative externalities of car travel, local governments encourage the use of public transit. Unfortunately, many suburban and rural areas are not adequately served as they lack the population density to justify public transit, i.e., the public transit is not economically viable. In cities with sprawling suburban areas, the utilization of public transit to commute to work is often low, e.g., less than 5% in metropolitan areas like Houston and Atlanta [McKenzie, 2010]. To attract more riders from suburban areas to public transit, transportation agencies must find adequate solutions to accommodate the first and last mile from the riders' home to and from the transit stations. Possible solutions for a transportation agency include operating a fleet of demand-responsive feeder vehicles

arXiv (Cornell University), 2020

In this paper, we study the challenging problem of how to balance taxi distribution across a city in a dynamic ridesharing service. First, we introduce the architecture of the dynamic ridesharing system and formally define the performance metrics indicating the efficiency of the system. Then, we propose a hybrid solution involving a series of algorithms: the Correlated Pooling collects correlated rider requests, the Adjacency Ride-Matching based on Demand Learning assigns taxis to riders and balances taxi distribution locally, the Greedy Idle Movement aims to direct taxis without a current assignment to the areas with riders in need of service. In the experiment, we apply city-scale data sets from the city of Chicago and complete a case study analyzing the threshold of correlated rider requests and the average online running time of each algorithm. We also compare our hybrid solution with multiple other methods. The results of our experiment show that our hybrid solution improves customer serving rate without increasing the number of taxis in operation, allows both drivers to earn more and riders to save more per trip, and all with a small increase in calling and extra trip time.

IEEE Transactions on Parallel and Distributed Systems, 2015

In the taxicab industry, a long-standing challenge is how to reduce taxicabs' miles spent without fares, i.e., cruising miles. The current solutions for this challenge usually depend on passengers to actively provide their locations in advance for pickups. To address this challenge without the burden on passengers, in this paper, we propose a cruising system, pCruise, for taxicab drivers to find efficient routes to pick up passengers to reduce cruising miles. According to the real-time pickup events from nearby taxicabs, pCruise characterizes a cruising process with a cruising graph, and assigns weights on edges of the cruising graph to indicate the utility of cruising corresponding road segments. Our weighting process considers the number of nearby passengers and taxicabs together in real-time, aiming at two scenarios where taxicabs are explicitly or implicitly coordinated with each other. Based on a weighted cruising graph, when a taxicab becomes vacant, pCruise provides a distributed online scheduling strategy to obtain and update an efficient cruising route with the minimum length and at least one arriving passenger. We evaluate pCruise based on a real-world GPS dataset from a Chinese city Shenzhen with 14;000 taxicabs. The evaluation results show that pCruise assists taxicab drivers to reduce cruising miles by 42 percent on average.

Electronics, 2020

Modern taxi services are usually classified into two major categories: traditional taxicabs and ride-hailing services. For both services, it is required to design highly efficient recommendation systems to satisfy passengers' quality of experience and drivers' benefits. Customers desire to minimize their waiting time before rides, while drivers aim to speed up their customer hunting. In this paper, we propose to leverage taxi service efficiency by designing a generic and smart recommendation system that exploits the benefits of Vehicular Social Networks (VSNs). Aiming at optimizing three key performance metrics, number of pickups , customer waiting time, and vacant traveled distance for both taxi services, the proposed recommendation system starts by efficiently estimating the future customer demands in different clusters of the area of interest. Then, it proposes an optimal taxi-to-region matching according to the location of each taxi and the future requested demand of each region. Finally, an optimized geo-routing algorithm is developed to minimize the navigation time spent by drivers. Our simulation model is applied to the borough of Manhattan and is validated with realistic data. Selected results show that significant performance gains are achieved thanks to the additional cooperation among taxi drivers enabled by VSN, as compared to traditional cases.

Advances in Spatial and Temporal …, 2011

Taxicab service plays a vital role in public transportation by offering passengers quick personalized destination service in a semiprivate and secure manner. Taxicabs cruise the road network looking for a fare at designated taxi stands or alongside the streets. However, this service is often inefficient due to a low ratio of live miles (miles with a fare) to cruising miles (miles without a fare). The unpredictable nature of passengers and destinations make efficient systematic routing a challenge. With higher fuel costs and decreasing budgets, pressure mounts on taxicab drivers who directly derive their income from fares and spend anywhere from 35-60 percent of their time cruising the road network for these fares. Therefore, the goal of this paper is to reduce the number of cruising miles while increasing the number of live miles, thus increasing profitability, without systematic routing. This paper presents a simple yet practical method for reducing cruising miles by suggesting profitable locations to taxicab drivers. The concept uses the same principle that a taxicab driver uses: follow your experience. In our approach, historical data serves as experience and a derived Spatio-Temporal Profitability (STP) map guides cruising taxicabs. We claim that the STP map is useful in guiding for better profitability and validate this by showing a positive correlation between the cruising profitability score based on the STP map and the actual profit ability of the taxicab drivers. Experiments using a large Shanghai taxi GPS data set demonstrate the effectiveness of the proposed method.

IAEME PUBLICATION, 2018

Taxi drivers always look for strategies to locate passengers quickly and increase their profit margin. Passenger-seeking strategies are primarily empirical and substantially vary among taxi drivers. From the history of taxi data, the topper-forming taxi drivers can earn 25 percent more than the ones with mediocre seeking strategy in the same period. A better strategy helps taxi drivers earn more with less effort and reduces fuel consumption and carbon emissions. Examining the influential factors in passenger-seeking strategies and algorithms to guide taxi drivers to passenger hot spots with the right timing is interesting. With the abundant availability of history taxicab traces, the existing methods of doing taxi business have been radically changed. The approach presents generic insights into the dynamics of taxicab services to maximize the profit margins for the concerned parties and the important metrics such as trip frequency, hot spots, and taxi mileage, and provides valuable insights towards more efficient operation strategies. We analyze these metrics using techniques like Newton’s polynomial interpolation and Gamma distribution to understand their dynamics.

2023

This study examines how efficient shared ride services, such as Uber Pool and Lyft Line, are within New York City with respects to cost effectiveness and consumer experience. A critical analysis using a data-driven approach of whether sharing rides are economical and the drawback of taking long routes. This study is crucial in balancing the tradeoff between affordability and quality, which can be used inform future planning including businesses, user satisfaction as well as urban transport. It seeks to answer the core question: Are the advantages of lower rates in co-rides worth the possibility risks associated with prolonged routes? CCS CONCEPTS • Applied computing~Transportation • Information systems~Data analytics • Computing methodologies~Machine learning

Proceedings of the 14th Workshop on Mobile Computing Systems and Applications - HotMobile '13, 2013

Ride-sharing on the daily homework 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 assumption 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 mobility 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 simplifying assumption that any commuter can share a ride with any other as long as their routes overlap. We show that simple 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% reduction 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%.

IEEE Transactions on Intelligent Transportation Systems, 2021

The ubiquitous growth of mobility-on-demand services for passenger and goods delivery has brought various challenges and opportunities within the realm of transportation systems. As a result, intelligent transportation systems are being developed to maximize operational profitability, user convenience, and environmental sustainability. The growth of last mile deliveries alongside ridesharing calls for an efficient and cohesive system that transports both passengers and goods. Existing methods address this using static routing methods considering neither the demands of requests nor the transfer of goods between vehicles during route planning. In this paper, we present a dynamic and demand aware fleet management framework for combined goods and passenger transportation that is capable of (1) Involving both passengers and drivers in the decision-making process by allowing drivers to negotiate to a mutually suitable price, and passengers to accept/reject, (2) Matching of goods to vehicles, and the multi-hop transfer of goods, (3) Dynamically generating optimal routes for each vehicle considering demand along their paths, based on the insertion cost which then determines the matching, (4) Dispatching idle vehicles to areas of anticipated high passenger and goods demand using Deep Reinforcement Learning (RL), (5) Allowing for distributed inference at each vehicle while collectively optimizing fleet objectives. Our proposed model is deployable independently within each vehicle as this minimizes computational costs associated with the growth of distributed systems and democratizes decision-making to each individual. Simulations on a variety of vehicle types, goods, and passenger utility functions show the effectiveness of our approach as compared to other methods that do not consider combined load transportation or dynamic multi-hop route planning. Our proposed method showed improvements over the next best baseline in various aspects including a 15% increase in fleet utilization and a 20% increase in average vehicle profits.

2017

Taxis (which include cars working with car aggregation systems such as Uber, Grab, Lyft etc.) have become a critical component in the urban transportation. While most research and applications in the context of taxis have focused on improving performance from a customer perspective, in this paper, we focus on improving performance from a taxi driver perspective. Higher revenues for taxi drivers can help bring more drivers into the system thereby improving availability for customers in dense urban cities. Typically, when there is no customer on board, taxi drivers will cruise around to find customers either directly (on the street) or indirectly (due to a request from a nearby customer on phone or on aggregation systems). For such cruising taxis, we develop a Reinforcement Learning (RL) based system to learn from real trajectory logs of drivers to advise them on the right locations to find customers which maximize their revenue. There are multiple translational challenges involved in...

EAI Endorsed Transactions on Industrial Networks and Intelligent Systems, 2018

On-demand transport has been disrupted by Uber and other providers, which are challenging the traditional approach adopted by taxi services. Instead of using fixed passenger pricing and driver payments, there is now the possibility of adaptation to changes in demand and supply. Properly designed, this new approach can lead to desirable tradeoffs between passenger prices, individual driver profits and provider revenue. However, pricing and allocations-known as mechanisms-are challenging problems falling in the intersection of economics and computer science. In this paper, we develop a general framework to classify mechanisms in ondemand transport. Moreover, we show that data is key to optimizing each mechanism and analyze a dataset provided by a real-world on-demand transport provider. This analysis provides valuable new insights into efficient pricing and allocation in on-demand transport.

Proceedings of the 14th International Conference on Agents and Artificial Intelligence, 2022

In this paper, we propose a unified mechanism known as T-Balance for scheduling taxis across a city. Balancing the supplies and demands in a city scale is a challenging problem in the field of the ride-sharing service. To tackle the problem, we design a unified mechanism considering two important processes in ride-sharing service: ride-matching and vacant taxi repositioning. For rider-matching, the Scoring Ride-matching with Lottery Selection (SRLS) is proposed. With the help of Lottery Selection (LS) and smoothed popularity score, the Scoring Ride-matching with Lottery Selection (SRLS) can balance supplies and demands well, both in the local neighborhood areas and hot places across the city. In terms of vacant taxi repositioning, we propose Qlearning Idle Movement (QIM) to direct vacant taxis to the most needed places in the city, adapting to dynamic change environments. The experimental results verify that the unified mechanism is effective and flexible.

Procedia Computer Science, 2020

Ridehailing has undeniably become an important mobility alternative for trip-makers around the globe. Its advent has given rise to a broad range of challenges in policymaking, planning, and modelling. Conventional models lack the functionality and structure required to handle complex mobility services such as ridehailing. This complexity resides in service provision processes that involve interdependent features such as matching, rebalancing, dynamic pricing, and driver activity. The latter is perhaps the most critical in that the vast majority of providers currently rely on human drivers, hence have very limited control over service fleets. Despite the importance of modelling driver activity, there is a marked dearth of research in this matter, which can in part be attributed to the hype of autonomous vehicles, and in part to the lack of data due to stringent data protection policies of (often private) ridehailing providers. In this context, this paper reports data mining efforts to exploit the information available for a conventional ridehailing trip-based dataset from RideAustin. Namely, the paper contributes to the literature with procedures to synthesize unobserved vehicle locations, generate very detailed full-day driver activity logs, and identify rebalancing trips from driver logs. These unique contributions provide valuable data that allow a wide range of applications. Relevant examples involve modelling matching and rebalancing mechanisms, or drivers' choice-making situations such as entering active/inactive periods.