Dynamic vanpool services: Passenger preferences, operations modeling and simulation-based quantification of impacts
Final Report Abstract
Ridesplitting on-demand services like dynamic vanpooling have recently emerged, showing invaluable market potential. However, they have had both successes and failures, posing uncertainties for all stakeholders, including the operators, policymakers, and service users. Therefore, this project investigates the fundamental issues and characteristics of dynamic vanpooling by developing tools, methods, and experiments specific to the ridesplitting service concept. On the one hand, the project contributes to providing robust models to significantly improve service planning, operational management, and impact assessment, while on the other hand, its empirical findings help better our understanding of the factors related to its user adoption, sustainable operations and the social and environmental benefits. The project research is divided into six different contribution areas. First, we quantify the passengers’ preferences towards dynamic vanpooling by developing discrete choice models that help investigate the factors affecting the user travel preferences for dynamic vanpools and estimate their perceived value of time (VOT) for attributes like walking/waiting and in-vehicle travel times. Then, we develop a scheduling algorithm that uses a multiple scenario approach and tabu search method to optimize the dynamic vanpooling service as a dynamic and stochastic dial-a-ride problem (DARP). The algorithm can efficiently optimize the service fleet while considering fleet prepositioning and stochasticity in both requests and time-dependent travel times. The third contribution integrates the developed scheduling algorithm with the microscopic traffic model SUMO using its online interfacing module TraCI alongside the development of necessary supply and demand enhancements to model the ridesplitting service behaviour. The developed tool is only one of its kind, allowing detailed modelling of traffic congestion dynamics, link-based service operations, and incorporation of stochastic information from the microscopic traffic model in dynamic DARP optimization. The fourth research part contributes to exploring and quantifying the dynamic vanpooling performance and impacts using the simulation platform with the case study of the Munich inner-city region. It explores ranges of multiple supply and demand variables to further develop and understand their relationships with passenger serviceability, occupancy, and related benefits. This research part also includes a comprehensive review of the foreseen impacts of shared autonomous vehicles (SAVs) and a survey-based impact analysis study of a shared-mobility service platform named Jetty, operating in Mexico City. From here, the project research scope was further extended towards developing network and service demand methods, where the fifth research part contributes to developing dynamic demand estimation methods that explicitly solve the non-linearity and dimensionality issues of large-scale networks. The methods help improve both the solution convergence and quality of dynamic origin-destination demand matrices, allowing more realistic traffic congestion modelling and accurate seed demands for service mode choice. The last research part contributes to developing a market equilibrium (ME) model and a utility-based compensation pricing method. The former caters explicitly to ridesplitting characteristics and allows analytical modelling of the service market states under wider ranges of supply-demand conditions. Whereas the latter compensates service passengers upon their experienced trip utilities to help improve trip certainty and service adoption as well as an alternative to provide smart subsidies. To conclude, our scientific findings significantly contribute toward a better understanding of the fundamentals of dynamic vanpooling service. From the operators’ perspective, the developed robust models can significantly improve service planning and operational management, while from the lawmakers’ perspective, both our models and experimental evaluations help understand user adaptability and service impacts. The project work also leaves many interesting future research directions in relation to service supply and demand modelling, pricing, and impacts evaluation.
Publications
- “PC–SPSA: Employing dimensionality reduction to limit spsa search noise in dta model calibration,” IEEE Transactions on Intelligent Transportation Systems, vol. 21, no. 4, pp. 1635–1645, 2019
M. Qurashi, T. Ma, E. Chaniotakis, and C. Antoniou
(See online at https://doi.org/10.1109/TITS.2019.2915273) - “The value of prepositioning in smartphone-based vanpool services under stochastic requests and time-dependent travel times,” Transportation Research Record, vol. 2673, no. 2, pp. 26–37, 2019
D. Li, C. Antoniou, H. Jiang, Q. Xie, W. Shen, and W. Han
(See online at https://doi.org/10.1177/0361198118822815) - “Incorporating trip chaining within online demand estimation,” Transportation Research Part B: Methodological, vol. 132, pp. 171–187, 2020, 23rd International Symposium on Transportation and Traffic Theory (ISTTT 23)
G. Cantelmo, M. Qurashi, A. A. Prakash, C. Antoniou, and F. Viti
(See online at https://doi.org/10.1016/j.trb.2019.05.010) - “Modeling autonomous dynamic vanpooling services in sumo by integrating the dynamic routing scheduler,” in SUMO User Conference, 2020
M. Qurashi, H. Jiang, and C. Antoniou
(See online at https://doi.org/10.5281/zenodo.4955078) - “Shared autonomous vehicle services: A comprehensive review,” Transportation Research Part C: Emerging Technologies, vol. 111, pp. 255–293, 2020
S. Narayanan, E. Chaniotakis, and C. Antoniou
(See online at https://doi.org/10.1016/j.trc.2019.12.008) - “The sustainability of shared mobility: Can a platform for shared rides reduce motorized traffic in cities?” Transportation Research Part C: Emerging Technologies, vol. 117, p. 102 707, 2020
A. Tirachini, E. Chaniotakis, M. Abouelela, and C. Antoniou
(See online at https://doi.org/10.1016/j.trc.2020.102707) - “Identifying and quantifying factors determining dynamic vanpooling use,” Smart Cities, vol. 4, no. 4, pp. 1243–1258, 2021
K. Tsiamasiotis, E. Chaniotakis, M. Qurashi, H. Jiang, and C. Antoniou
(See online at https://doi.org/10.3390/smartcities4040066) - “Clustered tabu search optimization for reservation-based shared autonomous vehicles,” Transportation Letters, vol. 14, no. 2, pp. 124–128, 2022
S. Su, E. Chaniotakis, S. Narayanan, H. Jiang, and C. Antoniou
(See online at https://doi.org/10.1080/19427867.2020.1824309) - “Dynamic demand estimation on large scale networks using principal component analysis: The case of non-existent or irrelevant historical estimates,” Transportation Research Part C: Emerging Technologies, vol. 136, pp. 103–504, 2022
M. Qurashi, Q.-L. Lu, G. Cantelmo, and C. Antoniou
(See online at https://doi.org/10.1016/j.trc.2021.103504) - “Microscopic modeling, optimization and demand estimation for autonomous mobility-on-demand ridesplitting,” Ph.D. dissertation, Technical University of Munich, 2022
M. Qurashi
