A Generic Data-Driven Recommendation System for Large-Scale Regular and Ride-Hailing Taxi Services

Abstract

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 pick-ups, 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.Electronics 2020, 9, 648 2 of 24 a means of transportation and enhance the efficiency of both services for the benefits of both customers and drivers.In regular taxi services, traditional ways for taxi drivers to find potential customers include driving around the city and waiting at some 'hot spots', e.g., taxicab stands. For the first option, taxi drivers usually follow an intuition-based trajectory hoping to find customers as soon as possible, while for the second option, most of the drivers will target the same hot spots since based on their personnel experience, they know when and where customers will be gathered. In the latter case, regular taxi drivers may be subject to an unfair competition since the number of taxis is higher than the demand or vice versa. Hence, traditional solutions for customer hunting are usually exhaustive and inaccurate. On the other hand, for the ride-hailing taxi services, although a central server is dedicated to manage the requests of customers and allocate them to drivers, similar problems that face regular taxi services still exist. Customers' requests might still be raised far away from drivers' locations and high vacant distances are accumulated, resulting in huge and redundant fuel consumption. In Portland, the average waiting times are estimated to be around six and ten minutes for regular and ride-hailing taxi services, respectively, according KGW News [2]. Therefore, it is recommended to enhance the efficiency of such transportation services by tackling the offer/demand problem in both taxi categories.Thanks to the spread of on-board and infrastructure-based sensors [3], collecting and sharing data have become very com...

Demand Prediction of Ride-Hailing Pick-Up Location Using Ensemble Learning Methods

“…In attempting to optimize the number of pick-ups whilst minimizing waiting time for taxi services, [14] developed a ride-hailing recommendation system. This is completed in 3 phases.…”

Advances in Public Transport Platform for the Development of Sustainability Cities

“…Hence, references [6,7] focused on offering improved usability and services based on multi-modal door-to-door passenger experiences to increase engagement. Other examples can be found in reference [8], where recommendation systems are designed to improve the passengers' experience and the drivers' profit. Finally, other approaches focused on educating the general public about this topic [9].…”

Leveraging Intelligent Transportation Systems and Smart Vehicles Using Crowdsourcing: An Overview

“…We illustrate a benefit of mobile and automated sensing crowdsourcing for ride-share recommendation systems in Figure 6. The proposed recommendation system utilizes automated sensing and mobile-crowdsourcing techniques to improve traditional ride-hailing and regular curbside ride-sharing services by introducing collaboration among drivers, improving quality of service by reducing customer wait time and wasted searching for new fares [25,26]. In this comparison, we evaluate the performance of the proposed recommendation system for both regular and ride-hailing taxi services.…”