An urban system-based travel demand forecasting technology framework

The Journal of Geo-Information Science published an online article on research led by Professor Pengjun Zhao (College of Urban and Environmental Sciences, Peking University; School of Urban Planning and Design, Peking University Shenzhen Graduate School) recently. The research team analyzed the systemic characteristics of urban transportation, proposed an integrated land-population-housing-transportation simulation technology framework, and developed an urban system-based travel demand simulation and prediction technology. The paper elaborates on the theoretical foundations, model construction, platform development, and applications of this technology.

The urban system-based transportation demand simulation and prediction technology encompasses sub-modules such as transportation demand distribution, mode share and route assignment, land use simulation, population and employment distribution, real estate prices, and carbon emissions to reflect the complete urban system. It incorporates variables like generalized travel cost, locational accessibility, real estate prices, job-housing relationship coefficients, and land-use mix to capture inter-submodule interactions and time-lag effects. Core algorithms for submodules were designed to enable urban system simulation and prediction. By leveraging spatiotemporal big data, the technology achieves fine-grained predictions at a kilometer grid scale and establishes a comprehensive workflow for data acquisition, cleaning, management, computation, and database construction. The accuracy of simulated results for transportation demand, congestion patterns, land use, and population distribution exceeds 85%.

The scale, distribution, mode structure, and flow of passenger travel demand are outcomes of socioeconomic activities and their spatial interactions across locations. The inherent complexity of socioeconomic systems necessitates that travel demand prediction must stem from the urban system itself to overcome the current technical limitations of "isolated transportation-centric approaches." This research addresses the challenge of real-time integrated simulation and prediction of land use, population, and employment distribution, achieving systematic and unified modeling of land-population-housing-transportation interactions. It provides innovative methodologies and technological tools to advance urban transportation planning.

For more details, please refer to the original article:

Construction and application of an urban system-based travel demand forecasting technology framework

https://www.sciengine.com/JGIS/doi/10.12082/dqxxkx.2024.240313（If you want to see the English version of the full text, please click on the “iFLYTEK Translation” in the article page.）

---

---