PassengerRequestDataset

Passenger Request Dataset for Urban Transport Study in Shenzhen, China

The dataset in this repository is associated with our paper titled "Idle-vehicle Rebalancing Coverage Control for Ride-sourcing systems" which can be found at this link. The study focuses on urban on-demand mobility in Shenzhen, China. There are two primary components of the dataset: the road network and passenger requests.

Road Network Details:

The road network specifically captures characteristics from two primary districts in Shenzhen, China, namely Futian and Luohu. It's designed as a graph where intersections are depicted as nodes, and road segments as links. These segments align with their actual geographical lengths. The two key columns in this dataset are:

streets, which portrays the entire road network in graph format.

N_coord, showing the geographical coordinates of each node. Here, the first column gives the latitude and the second, the longitude.

Passenger Requests:

This section is based on synthesized data that captures passenger requests over a period of 3 hours. The parameter γ is used to create synthetic destination distributions from given origin distributions. Its value, ranging between [0,1][0,1], indicates the balance in this Origin-Destination distribution. For instance, a smaller γ value points to a greater imbalance between the two distributions.

Datasets labeled request_gamma_000, request_gamma_025, request_gamma_050, request_gamma_075, and request_gamma_100 contain details of passenger requests when γ equals 0, 0.25, 0.5, 0.75, and 1, respectively. The columns in this data are:

time, indicating when the request was made.

orig, which points to the starting node on the graph.

dest, the destination node on the graph.

Usage Instructions:

For those using Matlab, the data can be accessed with the code:

load('path/to/shzn_Streets');
load('path/to/shzn_Coord');

And if you need to fetch passenger request data for γ = 0:

load('path/to/request_gamma_000');

Visualization:

To visually represent the Shenzhen road network on Matlab, you can use the following code:

graph_shenzhen = plot(shzn_Streets, 'XData', shzn_Coord(:, 1), 'YData', shzn_Coord(:, 2), 'Marker', 'none', 'LineStyle', '-', 'LineWidth', 1);

Citation:

If you decide to utilize this dataset for your research, we'd appreciate a reference to our paper. The citation details are:

@INPROCEEDINGS{ZHU2022,
  author={Zhu, Pengbo and Sirmatel, Isik Ilber and Trecate, Giancarlo Ferrari and Geroliminis, Nikolas},
  booktitle={2022 European Control Conference (ECC)}, 
  title={Idle-vehicle Rebalancing Coverage Control for Ride-sourcing systems}, 
  year={2022},
  volume={},
  number={},
  pages={1970-1975},
  doi={10.23919/ECC55457.2022.9838069}}

This dataset is also featured in our paper, "Data-enabled Predictive Control for Empty Vehicle Rebalancing," which you can find at this link. The citation details are:

@INPROCEEDINGS{ZHU2023,
  author={Zhu, Pengbo and Ferrari-Trecate, Giancarlo and Geroliminis, Nikolas},
  booktitle={2023 European Control Conference (ECC)}, 
  title={Data-enabled Predictive Control for Empty Vehicle Rebalancing}, 
  year={2023},
  volume={},
  number={},
  pages={1-6},
  doi={10.23919/ECC57647.2023.10178140}}