1. Understanding bike sharing use over time by employing extended technology continuance theory

    • Motivation: Facilitating users’ continuance intentions and retaining consumers are important to bike sharing service providers and governments.
    • Theory: Following extended technology continuance theory and incorporating perceived risk, we aim to identify factors that affect bike sharing services’ continuance intentions in this study.
    • Findings:
      • the extended technology continuance theory could provide a strong rationale in the investigation of continuance intention to adopt bike sharing services.
      • Perceived usefulness, satisfaction, and attitude are positively associated with continuance intention.
      • Perceived usefulness also positively impacts satisfaction and attitude.
      • Perceived risk tends to be negatively related to satisfaction.
      • Perceived ease of use is positively associated with perceived usefulness and attitude.

  2. Identification, Characterization, and Prediction of Traffic Flow Patterns in Multi-Airport Systems

    • Motivation: Efficient planning of airport capacity
    • Research gap: the dynamic and uncertain behavior of capacity-determining factors makes it difficult to estimate flow rates precisely, especially for strategic planning horizons.
    • Proposed method: a data-driven framework to identify, characterize, and predict traffic flow patterns in the terminal area of multi-airport systems toward improved capacity planning decision support in complex airspace.

  3. Scheduling the Operation of a Connected Vehicular Network Using Deep Reinforcement Learning

    • Motivation: Internet of Vehicles (IoV)
    • Research gap: satisfy the driver’s well-being and demand for continuous connectivity in the IoV era.
    • Proposed method: this paper addresses both safety and quality-of-service (QoS) concerns in a green, balanced, connected, and efficient vehicular network. Using the recent advances in training deep neural networks, we exploit the deep reinforcement learning model, namely deep Q-network, which learns a scheduling policy from high-dimensional inputs corresponding to the current characteristics of the underlying model.

  4. A longitudinal analysis of the effectiveness of California’s ban on cellphone use while driving

    • Motivation: In California, the use of handheld cellphones while driving has been prohibited since July 1, 2008.
    • Findings: interrupted time series analysis
      • The ban was found effective in reducing the cellphone usage-caused crashes in terms of both crash frequency and crash proportion.
      • crashes caused by cellphone use produce more severe outcomes than other crashes.
      • the ban motivates drivers to switch from handheld cellphones to hands-free cellphones, but in terms of crash severity, hands-free cellphone usage and handheld cellphone usage do not show significant differences.

  5. An Evaluation of HTM and LSTM for Short-Term Arterial Traffic Flow Prediction

    • Motivation: big data + machine learning
    • Research gap: leverage the increasingly large amounts of traffic volume data to improve traffic flow prediction and the detection of anomalous traffic flows
    • Proposed method: hierarchical temporal memory (HTM) for short-term prediction of traffic flows over real-world Sydney Coordinated Adaptive Traffic System data on arterial roads in the Adelaide metropolitan area in South Australia. Results are compared with those from long–short-term memory (LSTM).
      • HTM has potential as an effective tool for short term traffic flow prediction with results on par with LSTM and improvements when traffic flow distributions change.

  6. A pricing versus slots game in airport networks

    • Motivation: networks with two or three complementary airports.
    • Research work: We show that equilibrium policies involve slots when airport profits do not matter and pricing policies when airport profits matter. We further show that the equilibrium slot policies reach the first-best passenger quantities when congestion effects are absent.