Individual-level travel data in urban areas, which describes rich information such as the time, origin, destination, and other details of each travel activity, has high research value. However, the acquisition and open sharing of this type of data have always been a challenge. Due to limitations in data collection, the transportation industry has long relied on aggregate data such as traffic flow, speed, and density to conduct research. However, this type of aggregate data obscures a lot of information and cannot accurately characterize the operation of the transportation system, nor can it support research at the individual travel level.

With the development of sensing technology and the innovation of detection equipment, it has become possible to obtain individual-level travel data. However, because identity detection data involves individual travel privacy, this type of data cannot be shared to the same extent as aggregate data, which greatly hinders the development of individual-level transportation travel research.

In this context, we proposed an individual travel generation method that balances individual travel privacy protection with data availability. They have also released a city-scale individual travel generation dataset. The method extracts multidimensional travel information and features based on individual historical travel data, and then generates individual travel data by inferring travel frequency, travel time, destination, and other information in sequence. To address data privacy issues, the generation method protects sensitive travel information, making the data open for sharing.

The generated dataset covers travel records in the city for a single week, including important travel information. It provides a data foundation for micro-level deconstruction of urban transportation operation and analysis of individual travel, promoting the transformation of transportation research from aggregate calculation to individual inference calculation.

Experiments have shown that the generated individual travel data is highly consistent with historical data at the macro level, accurately reproducing information such as travel time distribution and high-traffic areas of the city. In addition, the generated individual travel data annotates the types of travelers, depicting the differences in travel distribution among different traveler groups. At the individual level, the generated individuals are close to real individuals in terms of travel frequency, travel interval, travel entropy, and other aspects, making the generated individual travel data is valuable for individual-level analysis and research.

The data is currently available on the data open platform Figshare (https://doi.org/10.6084/m9.figshare.c.6148536.v1), filling the gap in open data for individual-level travel datasets. The open access to this generated dataset can support individual-level travel analysis in urban areas and promote the development of research methods for individual travel, such as traffic control, route planning, and travel induction research.