You have to prepare movement data as a CSV file. The file must follow the rules below.
  • Each line must contain ID, date/time, longitude, latitude and movement status (one of stay, walk, vehicle and train).
  • Each individual’s location needs to be reported once an hour at least.
  • Integer numbers are recommended for ID.
  • Both slashes (/) and hyphens (-) can be delimiter between date parts.
  • Lines must be in chronological order. Do not back to past. Order of IDs are not cared.
  • The file can contain other extra attributes
Evaluation of Results

The accuracy of generated dataset will be evaluated by three aspects with following metrics:

Link Traffic Volume (40%): Traffic volume is defined as the number of people who traverse a specific road in one hour. We provide a traffic simulator and OpenStreetMap to specify the spatio-temporal locations of each synthetic agents in the output. The results will be compared with the volume of test dataset using Root Mean Square Error (RMSE)

Fig.1 Example of Intersection over Union

where n is the total number of links in the road network, i is the link #, y_i and y ̂_i are the real and estimated traffic volume of link i.

Hourly Grid-based Population Distribution (40%): To evaluate the population distribution, target area is split into 1km grid and the population in each is compared with the test dataset. Each synthetic agent’s spatio-temporal locations are specified using the above-mentioned simulator and the results will be evaluated by RMSE.

Transport Mode Share (20%): the number of users for each transport mode is also evaluated at each hour. The results will also be evaluated by RMSE.

Your Code

Source code will also be required to be submitted, either through a publicly available repository on a Git based version control hosting service such as GitHub or BitBucket, or directly upload code files. Since our work is publicly funded, all source code is expected to be released as open-source software, utilizing some generally accepted licensing such as Apache License 2.0, GNU General Public License, MIT license, or others of similar acceptance by the Open Source Initiative.