Household activity scheduling: choice-set generation and parameter estimation

Activity-based travel models that represent both household and individual decision-making involve intricate behavioural interactions and rely on complex simulation tools. As a result, calibrating their parameters presents significant methodological challenges. In this paper, we propose a novel maximum likelihood estimation procedure that explicitly accounts for choice sets defined at the household level. We apply the method to data from the UK National Travel Survey. Beyond illustrating and validating the approach, our results demonstrate that models calibrated at the household level produce daily schedule distributions that more closely reflect observed data than those based solely on individual-level modelling.