Welcome!
We have developed an automated, low-cost, and generalizable approach using Google Street View images and deep learning techniques to evaluate bus stop amenities. Leveraging the latest YOLOv8 model, transfer learning, and a dynamic prediction algorithm, our approach achieves efficient detection of bus stop amenities (e.g., shelters and benches) with high accuracy and precision. Scalability and transferability tests further suggest that highly accurate feature detection results can be achieved through model fine-tuning on a small sample of local data.
Key words: Computer Vision, transit, Google Street View Image, Bus Stop Amenities
Dai, Y., Liu, L., Wang, K., L., M., Yan, X. Using Deep Learning and Google Street View Images to Assess Bus Stop Amenities. Download Preprint.
Mobility hubs are physical locations that allow travelers to seamlessly switch between various modes of transportation such as public transit, ridehailing, and micromobility. We have developed a methodology to assess the suitability of an area for establishing mobility hubs and identify potential locations. The results are validated through an crowdsourcing approach facilitated by interactive mapping.
Read more about the project here
Key words: Mobility hub, transit, new mobility, crowdsourcing
Duarte, E., Lyu, D., Zheng, A., Merlin, L., Renne, J., Hoermann, S., Yan, X. Developing and validating a multi-criteria approach for locating multimodal mobility hubs. Download Preprint.
Through a data use agreement, we got access to novel dataset (i.e., the
11 Spin post-ride survey dataset) that records thousands of transit-connecting shared e-scooter trips in Washington DC. This unique dataset allows us to address some critical research gaps in prior research on transit and micromobility integration. Specifically, we used the dataset to reveal the spatiotemporal patterns of transit-connecting shared e-scooter trips in Washington DC and to validate the buffer-zone approach commonly used in prior research to infer transit-connecting micromobility trips.
Key words: Extreme heat, transit, microclimate, heat exposure
Yin, Z., Rybarczyk, G., Zheng, A., Su, L., Sun, B., & Yan, X. (2024). Shared micromobility as a first-and last-mile transit solution? Spatiotemporal insights from a novel dataset. Download Preprint Journal of Transport Geography, 114, 103778.
To address the policy question of how e-scooters interact with existing public mobility options, we conducted a spatiotemporal analysis of e-scooters’ relationships with public transit and station-based bikeshare in Washington DC. Results suggested that the service areas of the three modes largely overlap, and most e-scooter trips could have been made by transit or bikeshare. In addition, we found that e-scooters enhance mobility services for some underserved neighborhoods. Before COVID-19, about 10% of all e-scooter trips were taken to connect with the Metrorail system
Key words: E-scooter, public transit, bikeshare
Yan, X., Yang, W., Zhang, X., Xu, Y., Bejleri, I., Zhao, X. (2021). A spatiotemporal analysis of e-scooters’ relationships with transit and station-based bikesharing. [Download Preprint]. Transportation Research Part D: Transport and Environment. 12, 103088. https://doi.org/10.1016/j.trd.2021.103088
Some transit observers envision future public transit to be integrated systems with fixed-route services running along major corridors and ridesourcing servicing lower-density areas. This paper evaluates traveler preferences for a proposed integrated transit system versus the existing fixed-route system, with a particular focus on disadvantaged travelers. Results from low-income communties in Detroit and Ypsilanti suggest that a majority of survey respondents preferred a MOD transit system over a fixed-route one. However, some women may have safety concerns, and low technology self-efficacy can be a more serious barrier for many people to adopt MOD transit.
Key words: mobility on demand, public transit, transport equity
Yan, X., Zhao, X., Han, Y., Van Hentenryck, P., Dillahunt, T. (2021). Mobility-on-demand versus fixed-route transit systems: An evaluation of traveler preferences in low-income communities. [Download Paper]. Transportation Research Part A: Policy and Practice, 148, 481-495. https://doi.org/10.1016/j.tra.2021.03.019
Electric scooters (or e-scooters) have quickly proliferated in cities worldwide, presenting a host of regulatory challenges. We analyze trip origins and destinations for shared e-scooter use at the street-segment level. Results show that street segments near tourist sites, hotels, and transit stops attract the most scooter-trip destinations. In constrast, the supply of available e-scooters is the dominant force shaping scooter-trip origins
Key words: E-scooter, street segment, public transit, travel demand
Merlin, L., Yan, X., Zhao, X. (2021). A segment-level model of shared scooter origins and destinations. [Download Preprint]. Transportation Research Part D: Transport and Environment, 92, 102709. https://doi.org/10.1016/j.trd.2021.102709
In this paper, we developed a random forest model to examine which factors are key predictors of ridesplitting adoption rate (i.e., the proportion of ridesourcing trips with ridesharing authorization) and to explore their nonlinear relationships. The random forest model resulted in additional insights that were not obtainable from the commonly applied linear regression model.
Key words: random forest, machine learning, travel demand prediction
Xu, Y., Yan, X., Liu, X., Zhao, X. (2021). Identifying key factors associated with ride-splitting rate and modeling their nonlinear relationships. [Download Preprint]. Transportation Research Part A: Policy and Practice, 144, 170-188. https://doi.org/10.10416/j.tra.2020.12.005
As ridesourcing continues to grow in popularity, being able to accurately predict the demand for it is essential for effective land-use and transportation planning and policymaking. In this research project, we show that using random forest, a machine learning model that can automatically capture nonlinear relationships and interactive effections, can significantly improve predictive accuracy compared to the traditional statistical model.
Key words: random forest, machine learning, travel demand prediction
Yan, X., Liu, X., Zhao, X. (2020). Using machine learning for direct demand modeling of ridesourcing services in Chicago. [Download Preprint]. Journal of Transport Geography, 83, 102661.
In recently years, transit agencies have started to consider integrating ridesourcing services (i.e. on-demand, app-driven ridesharing services) with public transit. In this project, we evaluated traveler preferences for this type of integrated transit system. We find that when used to provide convenient last-mile connections, ridesourcing could provide a significant boost to transit ridership.
Key words: Ridesourcing, public transit, travel behavior, mode choice
Yan, X., Levine, J., Zhao, X. (2019). Integrating ridesourcing services with public transit: An evaluation of traveler responses combining revealed and stated preference data. [Download Preprint]. Transportation Research Part C: Emerging Technologies, 105, 683-696. https://doi.org/10.1016/j.trc.2018.07.029