Framework for Generating 7-Day Activity Schedules Considering Household Interactions

This paper introduces a novel framework for generating activity schedules in the context of households over a 7-day period. By combining an activity generation model based on the Multiple Discrete Continuous Extreme Value (MDCEV) approach with a constraint satisfaction optimization approach, the proposed framework addresses the limitations of existing studies that focus on single-day activities. The MDCEV model estimates utility parameters for different activities and income levels based on data from the German Mobility Panel, providing a foundation for understanding activity preferences. The schedule frame and fine-tuning models then generate realistic schedules, considering work hours, joint activities, and leisure time. Although the model is in its prototype phase, it already demonstrates promising results and can be integrated into agent-based travel demand models like mobiTopp and MATSim. Future work will involve further calibration and exploration of chore allocation within households, the impact of flexible work arrangements, and the influence of non-travel activities such as online shopping on activity patterns and travel behavior. Overall, this research contributes to a more holistic understanding of household interactions and provides valuable insights for travel demand modeling and urban planning.