Jaeyoung Jung

Dynamic Shared-Taxi Dispatch Algorithm with Hybrid-Simulated Annealing

Taxi is certainly the most popular type of on-demand transportation service in urban areas because taxi-dispatching systems offer more and better services in terms of shorter wait times and passenger travel convenience. However, a shortage of taxicabs has always been critical in many urban contexts especially during peak hours, and taxi has great potential to maximize its efficiency by employing the shared-ride concept. There are recent successes in dynamic ride-sharing projects that are expected to bring substantial benefits arising from energy consumption and operation efficiency and thus, it is essential to develop advanced shared-taxi-dispatch algorithms and investigate the collective benefits of dynamic ride-sharing by maximizing occupancy and minimizing travel times in real-time. This article investigates how taxi services can be improved by proposing shared-taxi algorithms and what type of objective functions and constraints could be employed to prevent excessive passenger detours. Hybrid-simulated annealing HSA is applied to dynamically assign passenger requests efficiently. A series of simulations are conducted with two different taxi operation strategies. The simulation results reveal that allowing ride-sharing for taxicabs increases productivity over the various demand levels and HSA can be considered as a suitable solution to maximize the system efficiency of dynamic ride-sharing.

High-Coverage Point-to-Point Transit Study of Path-Based Vehicle Routing Through Multiple Hubs

This study focuses on the optimization and simulation modeling associated with the design of alternative transportation, the high-coverage point-to-point transit (HCPPT), which involves a sufficient number of deployed small vehicles with advanced-information supply schemes. This paper identifies the inefficiency of the existing heuristic rules for vehicle routing and proposes a new optimization approach for an HCPPT solution. A path-based model for routing through multiple hubs as opposed to a single pair of hubs is formulated to improve HCPPT operational schemes. This study also develops a simulation framework for the application of the proposed algorithm. To illustrate the system and computational performances of the proposed model, simulations are conducted with different sets of scenarios and model parameters. The path-based model shows reasonable performance over the various demand patterns in level of service and ride time index. It is also shown that, with the use of constraint-driven schemes and model parameters, the scale of the problem is reduced. The computational times are shown to be quite small, and demonstrate the viability in real-time operations.

High-Coverage Point-to-Point Transit: Electric Vehicle Operations

Electric vehicles (EVs) have substantial potential for fleet applications as energy-efficient vehicles with fewer carbon-based emissions. However, EV utilization brings other concerns related to limited vehicle ranges and battery-charging issues, and thus the recharging infrastructure needs to be carefully designed and financially supported. This study focuses on the use of an EV fleet for an innovative transportation alternative called high-coverage point-to-point transit (HCPPT), which involves a sufficient number of deployed small vehicles that respond to real-time desires for point-to-point travel of individual travelers. As large-scale fleet operation of EVs and the HCPPT system are both envisaged only as future transportation alternatives, the benefits of combining these options have to be modeled and studied. Possible infrastructure investment benefits could result for both systems when transfer-hub locations of HCPPT are used as electric-charging locations. This paper studied the performance results from the simulation of a real-world transfer-hub system and developed certain vehicle-routing schemes to handle the specifics of EV operations. Alternative ways to schedule the charging of EVs were modeled with different insertion heuristics within vehicle-routing optimization. The results showed that the battery-charging limitations of EVs do not significantly affect the efficiency of the HCPPT system and revealed the importance of details such as the number of required charging stations at the hub locations.

Simulation Framework for Modeling Large-Scale Flexible Transit Systems

Despite the prevalent use of simulation methods in a majority of areas in transportation research, transit and paratransit research has not relied heavily on simulation models, not only because traditional transit studies do not necessarily require simulation techniques but also because proper simulation packages are not available for modeling innovative transit systems. This study proposes a new type of simulation framework, targeting large-scale flexible transit systems with various vehicle operation schemes. The intent of this study is to describe the generalized concepts and the detailed architecture of the framework developed for urban transportation networks. A comprehensive process for the framework development, including considerations of simulation data conversion and user interface design, is discussed. Two real-time flexible transit applications are modeled with the proposed simulation framework: high coverage point-to-point transit and real-time shuttle service. Detailed simulation results are provided, and the importance of simulation design and operational features is addressed. The results also show that the framework can model realistic large networks several times faster than real time with standard desktop computers; these results raise the possibility of its use in real-time optimization schemes as well as for synchronous modeling with commercial road traffic simulators.

Dually sustainable urban mobility option: Shared-taxi operations with electric vehicles

ABSTRACT Electric vehicles (EVs) are energy efficient and often presented as a zero-emission transport mode to achieve long-term decarbonization visions in the transport sector. The implementation of a sustainable transportation environment through EV utilization, however, requires the addressing of certain cost and environmental concerns such as limited driving range and battery-charging issues before its full potential can be realized. Nevertheless, a specific type of use of EVs, namely in taxi services, may elicit positive public opinion, as it promises a commitment toward sustainability in urban life. In light of this, this study proposes an integrated approach that combines EV operation with a conceptual design for shared-ride taxi services. As some productivity loss may be naturally expected due to the time spent in charging, it is important to look at whether such performance loss from the passenger and system standpoints can be offset with ingenuity in operational design. In this study, an EV taxi charge-replenishing scheme that can be coupled with a real-time taxi-dispatch algorithm is designed. The proposed EV charging schemes for taxi services are studied via simulations and the effects of the limited driving range and battery-charging details are examined from a system performance viewpoint. The simulation study also reveals illustrative results on the impact of the EV taxi fleets operation on the charging system. Next, a real-time shared-taxi operation scheme that allows ride sharing with other passengers is proposed to maximize the operational efficiency. The simulation results suggest that the shared-taxi concept can be a viable option to improve on the limitations caused by EV operation. In addition, the importance of projected charging demands and queue delays at different charging locations are also addressed. Some limitations and a future research agenda are also discussed.

High coverage point-to-point transit: Local vehicle routing problem with genetic algorithms

High Coverage Point-to-Point Transit (HCPPT) is a new design of alternative transportation, which involves a sufficient number of deployed small vehicles with advanced information supply schemes. This paper focuses on Genetic Algorithms (GA) to improve system performance with real-time re-optimization and compares GA with the existing insertion heuristics for local vehicle routing in HCPPT. Two genetic operation schemes, Best Feasible Position (BFP) and Random Feasible Position (RFP), are designed. Simulations are performed with different demand levels based on OCTA (Orange County Transportation Authority) trip demands. The results show that BFP significantly improves the local routing performance in terms of both system efficiency and productivity whereas RFP shows improvement only in system efficiency compared to the insertion heuristics. This study also provides results of computational performances as well as convergence performance. In terms of computational performance, both GA approaches show viability in real-time operations.