Stein W. Wallace | Searcharxiv

arXiv2025

Assessing Autonomous Mobility-on-Demand Services and the Impacts of Operational Strategies: A Case Study of Chengdu, China

The Autonomous Mobility-on-Demand (AMoD) service is emerging as a potential alternative to on-demand urban mobility, but its operational performance relative to traditional street-hailing services and the effectiveness of related operational strategies remain unclear. This study presents a simulation framework integrating a graph theory-based trip-vehicle matching mechanism and uses historical street-hailing operations data to simulate AMoD services in Chengdu, China. The operational performance of these two urban mobility modes is evaluated using three key performance indicators: average passenger waiting time (APWT), average deadheading mileage (ADM), and average deadheading energy consumption (ADEC). We further evaluate the impacts of four operational strategies on simulated AMoD performance: vehicle repositioning, fleet size management, geofencing, and request rejection. Simulation results indicate that, under the same historical trip demand, fleet-size constraints, and road network as the observed street-hailing system, the simulated AMoD service is estimated to have lower values of APWT, ADM, and ADEC by 73.3% to 83.4%, 75.0%, and 74.0%, respectively, reflecting the potential operational gains associated with centralized dispatch in simulation settings. These differences are most pronounced during early-morning low-demand hours and in remote areas such as airports.

math.OC↗

arXiv2020

On scenario construction for stochastic shortest path problems in real road networks

Stochastic shortest path computations are often performed under very strict time constraints, so computational efficiency is critical. A major determinant for the CPU time is the number of scenarios used. We demonstrate that by carefully picking the right scenario generation method for finding scenarios, the quality of the computations can be improved substantially over random sampling for a given number of scenarios. We study a real case from a California freeway network with 438 road links and 24 5-minute time periods, implying 10,512 random speed variables, correlated in time and space, leading to a total of 55,245,816 distinct correlations. We find that (1) the scenario generation method generates unbiased scenarios and strongly outperforms random sampling in terms of stability (i.e., relative difference and variance) whichever origin-destination pair and objective function is used; (2) to achieve a certain accuracy, the number of scenarios required for scenario generation is much lower than that for random sampling, typically about 6-10 times lower for a stability level of 1\%; and (3) different origin-destination pairs and different objective functions could require different numbers of scenarios to achieve a specified stability.

eess.SP↗