Search

• Propose a duopoly evolutionary game model for the dynamic ride-hailing market. • Capture bounded rationality, strategic learning, and dynamic competition between platforms. • Identify the existence of stationary points and derive simplified stability conditions. • Establish two extended models to capture the service aggregator and mixed electric/fuel fleets. • Provide a fresh perspective for exploring dynamic ride-hailing competition and market evolution. The platform competition in a real-world ride-hailing market is complex and dynamic. Existing research mainly relies on static competitive equilibrium, lacking an understanding of the dynamic evolution process of platform competition. This paper develops a generic duopoly evolutionary game model to capture dynamic decision adjustments and the evolution of market conditions. Bounded rationality and strategic learning are integrated into the modeling to dissect the interaction and competition between two platforms. The existence of stationary points where the system states remain unchanged over time is identified, and simplified stability conditions are derived. A pure strategy pair is found stable when constituting a strict Nash equilibrium and corresponding to positive payoffs. We establish two extended models to capture the nascent service aggregator and mixed electric/fuel fleets and compare them with the basic model. The model is calibrated employing the ride-hailing data of New York and is applicable for analyzing the market evolution under different states. The impacts of potential demand drop, aggregated services, and mixed fleets on platform decision and driver-passenger choice evolution are unveiled. It is found that the platforms possess self-recovery capability when facing a deteriorated external environment. The decision-making of the small platform exhibits a more substantial inertia and volatility since the large platform's decision can easily influence the market supply and demand. In the initial stage, the service aggregator slightly harms the large platform, but in the long run, it benefits both platforms. Besides, both platforms reduce prices while inflating commission ratios under mixed fleets. This study provides a fresh perspective for exploring the real-world dynamic ride-hailing competition and market evolution. [ABSTRACT FROM AUTHOR]

The timely transportation of goods to customers is an essential component of economic activities. However, heavy-duty diesel trucks used for goods delivery significantly contribute to greenhouse gas (GHG) emissions within many large metropolitan areas, including Los Angeles, New York, and San Francisco. To reduce GHG emissions by facilitating freight electrification, this paper proposes Joint Routing and Charging (JRC) scheduling for electric trucks. The objective of the associated optimization problem is to minimize the cost of transportation, charging, and tardiness. A large number of possible combinations of road segments as well as a large number of combinations of charging decisions and charging durations leads to a combinatorial explosion in the possible decisions electric trucks can make. The resulting mixed-integer linear programming (MILP) problem is thus extremely challenging because of the combinatorial complexity even in the deterministic case. Therefore, a Surrogate Level-Based Lagrangian Relaxation (SLBLR) method is employed to decompose the overall problem into significantly less complex truck subproblems. In the coordination aspect, each truck subproblem is solved independently of other subproblems based on the values of Lagrangian multipliers. In addition to serving as a means of guiding and coordinating trucks, multipliers can also serve as a basis for transparent and explanatory decision-making by trucks. Testing results demonstrate that even small instances cannot be solved using the off-the-shelf solver CPLEX after several days of solving. The SLBLR method, on the other hand, can obtain near-optimal solutions within a few minutes for small cases, and within 30 min for large ones. Furthermore, it has been demonstrated that as battery capacity increases, the total cost decreases significantly; moreover, as the charging power increases, the number of trucks required decreases as well. • Present a formulation of a joint heavy-duty vehicle fleet routing & charging problem. • Use surrogate level-based Lagrangian relaxation approach to solve the problem. • The proposed method obtains near-optimal solutions for realistic testing cases. [ABSTRACT FROM AUTHOR]

• Detecting outliers (anomalies) is critical in safety-related applications like ATM. • Metroplex airspace is most congested and highly-coordinated of all components of ATM. • ATM data capturing technologies enables data-driven approach to improving safety and throughput. • The anomaly detection algo must be interpretable, and it must evolve with and adapt to airspace. • This is best achieved through ingesting both operational, and real-time surveillance data. The airspace system is a complex dynamical system with complicated controlled interactions between its constituent subsystems – terminal airspace, en-route airspace, and ground. Of these, air traffic management in the multi-airport (metroplex) terminal airspace is one of the most complicated subsystems to manage, especially due to the interactions between proximal airports. Analyzing anomalous behaviors in the metroplex is emerging as a key problem in understanding air traffic management complexity and safety. Although physics-based approaches have been studied in-depth for this application, newfound interest has been observed to use recorded time-series air traffic surveillance and airport operations datasets for this purpose. In this paper, we propose a machine learning-based anomaly detection algorithm that generates mathematical models to detect anomalies in metroplex operations. Several machine learning algorithms have been developed to detect anomalies using only air traffic surveillance data, but there is a significant scope of improvement by including airport operational characteristics as well, since integrating such closely-controlled metroplex operational datasets allows the developed models to effectively detect anomalies. The key contribution of this paper is in allowing anomaly detection models to recursively update so as to adapt to changes in metroplex operations. The proposed algorithm is demonstrated with real air traffic surveillance and airport operations datasets at LaGuardia, John F. Kennedy, and Newark airports, thereby detecting anomalies for operations in the New York metroplex. [ABSTRACT FROM AUTHOR]

Bike-sharing systems are a rapidly developing mode of transportation and provide an efficient alternative to passive, motorized personal mobility. The asymmetric nature of bike demand causes the need for rebalancing bike stations, which is typically done during nighttime. To determine the optimal starting inventory level of a station for a given day, a User Dissatisfaction Function (UDF) models user pickups and returns as non-homogeneous Poisson processes with piece-wise linear rates. In this paper, we devise a deep generative model directly applicable in the UDF by introducing a variational Poisson recurrent neural network model (VP-RNN) to forecast future pickup and return rates. We empirically evaluate our approach against both traditional and learning-based forecasting methods on real trip travel data from the city of New York, USA, and show how our model outperforms benchmarks in terms of system efficiency and demand satisfaction. By explicitly focusing on the combination of decision-making algorithms with learning-based forecasting methods, we highlight a number of shortcomings in literature. Crucially, we show how more accurate predictions do not necessarily translate into better inventory decisions. By providing insights into the interplay between forecasts, model assumptions, and decisions, we point out that forecasts and decision models should be carefully evaluated and harmonized to optimally control shared mobility systems. • Predictive and prescriptive performance does not always align. • Predictive models should be evaluated on inventory decision quality. • Variational Poisson RNN formulated for pickup and return rate prediction. • Experiments with synthetic and real-world data demonstrate solution approach. • The misalignment between prediction and decision objectives is quantified. [ABSTRACT FROM AUTHOR]

• Introduces the concept of the Near-on-Demand Mobility service. • Proposes algorithms for large-scale request assignment and vehicle rebalancing. • Assesses the impact of design aspects on its performance using data from China and United States. • Develops a simulation platform facilitating the evaluation of a wide range of TNC designs. Mobility-On-Demand (MoD) services have been transforming the urban mobility ecosystem. However, they raise a lot of concerns for their impact on congestion, Vehicle Miles Traveled (VMT), and competition with transit. There are also questions about their long-term survival because of inherent inefficiencies in their operations. Considering the popularity of the MoD services, increasing ride-pooling is an important means to address these concerns. Shareability depends not only on riders' attitudes and preferences but also on operating models deployed by providers. The paper introduces an advance requests operating model for ride pooling, users may request rides in advance of their desired departure times. A platform with efficient algorithms for request matching, vehicle routing, rebalancing, and flexible user preferences is developed. Data from Chengdu, China, and New York, United States are used to evaluate the performance of the advance requests system. The impact of various design aspects of the system (e.g. advance requests horizon, vehicle capacity) on its performance is investigated. The sensitivity of the results to user preferences in terms of the level of service (time to be served and excess trip time), willingness to share and place requests in advance, and traffic conditions are explored. The results suggest that significant benefits in terms of sustainability, level of service, and fleet utilization can be realized when advance requests are along with an increased willingness to share. Furthermore, even "near-on-demand" (relative short planning horizons) operations can offer benefits for all stakeholders (users, operators, cities). [ABSTRACT FROM AUTHOR]

• Parcel, grocery, and food deliveries in New York City are investigated. • Factors affecting deliveries vary by the delivery type (parcel/grocery/food). • The relationship between delivery and shopping trip varies by the delivery type and trip mode. With advances in information technology, online shopping for a parcel (retail), grocery, and food (i.e. prepared meal) have become a part of daily life and they are expected to continue to grow in the future. Accordingly, shopping behaviors, i.e. shopping channels, are becoming more complex than ever. However, the current research focus is still mainly on traditional parcel delivery and little is known for food and grocery deliveries and their impacts on transportation systems. To have a better understanding of deliveries and their impacts on transportation systems, using the survey data from NYC DOT Citywide Mobility Survey in 2018, this paper investigated (a) factors affecting the three types of deliveries (retail, grocery, and food) by developing a seemingly unrelated model, and (b) the relationships between deliveries and in-store shopping trips by different modes using multiple simultaneous equations models. The results showed that factors affecting deliveries vary by the type of delivery, and the three deliveries are positively correlated by common unobserved factors. In addition, the relationships between delivery and shopping trip vary not only by delivery type but also by trip mode (driving/walking). The findings would provide valuable insights into travel demand modeling and curbside management. [ABSTRACT FROM AUTHOR]