Ridesharing, a shared service that uses the information and knowledge matching, can efficiently utilize scattered social resources to reduce the demand for vehicles in urban road networks. However, car ridesharing has the problems of low capacity and high cost, and it cannot satisfy demands for recurring, long-distance, and low-cost trips. In this paper, we formally define the bus ridesharing problem and propose a large-scale bus ridesharing service to resolve this problem. In our proposed model, the rider can use an online bus-hailing service to upload his or her trip demand and wait to be picked up when it gathers enough people. The provider assigns drivers to riders after integrating the matched ride requests. To maximize ridesharing's success rate, we developed both exact algorithms and approximate algorithms to optimize the ride-matching service. A real-life dataset that contains 65,065-trip instances extracted from 10,585 Shanghai taxis from one day (Apr 1, 2018) is used to demonstrate that our proposed service can provide higher cost performance and on-demand bus services for every ride request. Meanwhile, it reduces the number of vehicles used by 92% and 96% and the amount of oil used by 87% and 92% compared with car ridesharing and no ridesharing, respectively.INDEX TERMS Ridesharing, bus pooling, capacitated clustering problem, location-allocation problem.
An optimal meeting point query is used to determine a location in a spatial region to build a new facility that minimizes the sum of the (weighted) road distances from all clients. This problem has been studied in previous work with the assumption that all clients and facilities reside in Euclidean space or along road networks. However, due to the limitations of geographic information system technologies, it is difficult to return an exact geographic location to answer the optimal meeting point query based on a set of arbitrary coordinates. This issue results in various problems, such as positioning and measurement errors, in practical use. In this paper, it is aimed to identify the optimal meeting point in road networks for clients and facilities residing in non-Euclidean spaces. Two efficient heuristic solutions are proposed based on approximate and adaptive query processing techniques by using randomized adaptive search and random direction search methods, respectively, to rapidly converge to the global optimum in the geographic coordinate system. Extensive experiments based on real datasets demonstrate that our proposed method achieves a 32.11% improvement over the state-of-the-art approach.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.