Towards a conjoint-based, context-dependent model of task allocation in activity settings: Some numerical experiments

Abstract.

 None of the currently developed activity-based models of transport demand explicitly models task allocation among household members. To fill this gap, the present paper suggests to complement activity-based models of activity scheduling with a context-dependent model of task allocation. That is, it is assumed that the allocation of tasks within households is partly based on such contextual variables as the amount of time a member has to spend on mandatory activities and car availability. In particular, the paper advocates a conjoint-based approach, based on an assignment task as opposed to the traditional ranking, rating or choice response formats. By definition, an assignment task involves a combinatorial explosion of choice alternatives, implying that additional operational decisions to estimate the context-dependent model are required. This study presents the results of various numerical experiments, conducted to better understand the impacts of those decisions on the degree of bias in the parameter estimates of the choice model. The results of these simulations indicate that under particular assumptions, the development and estimation of a conjoint-based, context-dependent model of task allocation within households is feasible.