A visit generation process for human mobility random graphs with location-specific latent-variables: From land use to travel demand

This research introduces a novel mathematical framework for understanding collective human mobility patterns, integrating mathematical modeling and data analysis. It focuses on latent-variable networks to investigate the dynamics of human mobility using stochastic models. By analyzing origin–destination data, the study uncovers scaling relations and explores the economic implications of mobility patterns, particularly regarding the income elasticity of travel demand. The mathematical analysis begins with the development of a stochastic model based on inhomogeneous random graphs, constructing a visitation model with multipurpose drivers for travel demand. Through this model, the study gains insights into the structural properties and dynamic correlations of human mobility networks, deriving analytical solutions for key network metrics: visit distribution, assortativity behavior and clustering coefficient. Empirically, the study validates the model's assumptions and reveals scaling behaviors in origin–destination flows within a region, reproducing statistical regularities observed in real-world cases. Notably, the model's application to estimating income elasticity of travel demand provides significant implications for urban and transport economics. Overall, this research contributes to a deeper understanding of the interplay between human mobility and regional demographics and economics. It sheds light on critical scaling relations across various aspects of collective human mobility and underscores the importance of incorporating latent-variable structures into mobility modeling for accurate economic analysis and decision-making in urban and transportation planning.