Yufeng Zhang

Modular autonomous transit systems (MATS) present a promising direction for the advancement of urban mobility due to their remarkable ability to flexibly allocate capacity across both spatial and temporal dimensions. Although scholars have extensively explored optimizing the coupling and decoupling of modules and creating adaptable service strategies for MATS, the capacity for modules to overtake has largely been neglected, which could reduce MATS efficiency. In this paper, we introduce a novel operational strategy allowing certain modules to detach from their scheduled trip and, by bypassing stops, create an unscheduled express service, thus leading to an en-route express service. The potential overtaking actions of both decoupled and non-decoupled modules, due to skipping stops, and passengers transfer within the module, add significant complexity to the model. To address this, we develop a mixed integer nonlinear programming (MINLP) model with an objective to minimize the total cost for both passengers and the operator of the transit system, and we determine optimal decoupling/coupling strategies and schedules for the en-route express services. To enhance computational efficiency, we recast the original nonlinear model into a mixed integer quadratic program (MIQP) and introduce an outer approximation (OA) algorithm to solve it effectively. The results of our illustrative and large-scale experiments reveal that the proposed OA algorithm significantly enhances computational efficiency compared to CPLEX solvers. Compared to two benchmark systems with fixed capacity buses—local (all-stop) service and stop-skipping service, the proposed en-route express service reduces the total cost by 15.1% and 12.9%, and lowers the average passenger service time cost by 19.7% and 33.8%, respectively, underscoring the advantages of the en-route express service for MATS. These findings contribute to the development of more efficient MATS operations by introducing an en-route decoupling strategy that leverages overtaking capabilities to create adaptive express services. The work highlights the significant potential and importance of developing urban mobility systems that are more adaptable and responsive to urban travelers, while optimizing the utilization of available resources.

As app-based ride-hailing (or e-hailing) services have achieved great success worldwide, the traditional taxi industry faces an unprecedented crisis. To gain a deeper understanding of these two distinct modes that both offer door-to-door mobility services, in this study, we use Chicago as a case to comparatively analyze the travel patterns of the app-based ride-hailing and traditional taxi services and understand how socio-demographic and urban land use attributes affect the usage of two services in temporal and spatial dimensions. We employ the local regression models named geographically and temporally weighted regression (GTWR) models that can capture the spatiotemporal nonstationarity between variables as the primary analysis tool. Four GTWR models that predict the average hourly taxi or ride-hailing trip volume on weekdays or weekends for each community area are carefully constructed with R^2 all above 0.98. The nonstationarity test shows all variables exhibit extra local variations, indicating the necessity of using the GTWR models to explore the attributes’ spatiotemporal impacts. Spatial error models (SEM) and ordinary least square (OLS) models, global regression models, are also built as comparisons and benchmark references. The differences between global and local regression models are illustrated through clustering analysis. To compare and contrast the usage patterns of ride-hailing and taxi services, we analyze the temporal-spatially varying coefficients of GTWR models. Interpretations are provided for our interesting findings, including the strong trip generation power of the population in the South Side area, the attitude of white people towards taxi services is positive but ambiguous towards ride-hailing services, young people’s opposing behaviors towards taxi services on weekdays and weekends, and the opposite relationship between transit and taxi or ridehailing in certain areas. Based on these key findings, we offer some planning and operational suggestions, such as developing mobile apps, implementing price-related strategies, strengthening partnerships with local businesses, building multi-modal transportation, clarifying customer segmentation, providing auxiliary services, etc.

Reliable shortest path (RSP) problem reflects the variability of travel time and is more realistic than standard shortest path problem which considers only the average travel time. This paper describes an algorithm for solving the mean-standard deviation RSP problem considering link travel time correlations. The proposed algorithm adopts the Lagrangian substitution and covariance matrix decomposition technique to deal with the difficulty resulting from non-linearity and non-additivity of the Mixed Integer Non-Linear Program (MINLP). The problem is decomposed into a standard shortest path problem and a convex optimization problem whose optimal solution is proved and the Lagrangian multipliers ranges are related to the eigenvalues of the covariance matrix to further speed up the algorithm. The complexity of the original problem is notably reduced by the proposed algorithm such that it can be scaled to large networks. In addition to the sub-gradient Lagrangian multiplier updating strategy integrated with projection, a novel one based on the deep-cut ellipsoid method is proposed as well. Numerical experiments on large-scale networks show the efficacy of the algorithm in terms of relative duality gap and computational time. Besides, there is evidence showing that, though having longer computational time, the ellipsoid updating method tends to obtain better solutions compared with the sub-gradient method. The algorithm outperforms the existing one-to-one Lagrangian relaxation-based RSP algorithms and the exact Outer Approximation method in the literature.