A behavioral model of accessibility based on the number of available opportunities

The paper proposes a new behavioral definition of accessibility as the expected number of opportunities "available" for a subject to perform an activity, where "available" means that the opportunity i) is perceived as a potential alternative to satisfy one's needs, and ii) it can be reached given the spatiotemporal constraints of the individual's schedule. A new class of accessibility models is derived in accordance with the above definition, exploiting the strengths of both utility-based and opportunity-based models, and explicitly incorporating spatiotemporal constraints which may limit the availability of perceived opportunities. The proposed model is formulated for both active and passive accessibility, does not suffer from reflexivity issues, supports both trip-based and activity-based formulations, and scales up from individual to spatially aggregated opportunities, taking into account effectively the different ways of perceiving opportunities. Moreover, the resulting accessibility measure has a straightforward interpretation, being expressed in physical units, and is comparable across different locations. Performances of the proposed model in reproducing the active accessibility to cinemas in the Naples metropolitan area (Italy) are compared to those of a traditional isochrone-based measure and a distance-decay model. For this aim, measurements of individuals' perceived cinemas were gathered through a survey, and adopted for the calibration of both the accessibility models. Calibration results show that the proposed model outperforms both the traditional isochrone-based measure and a distance-decay model calibrated against the same dataset, better reproducing both the quantity and spatial distribution of perceived opportunities