An integrated model of intensity of activity opportunities on supply side and tour destination & departure time choices on demand side

Land use and transportation interactions exist at all time scales - long, medium, and short. In the long term, business location (and relocation) decisions, aggregate travel patterns and transportation infrastructure development are inter-dependent. In the medium-short term, in any neighborhood, the temporal profile of activity opportunities within a day, and destination and departure time (DDT) preferences of travelers are simultaneously determined. This paper explored these short-term interdependencies between the land-use supply and travel demand systems by developing a simultaneous model of time-of-day specific zonal employment intensity and non-mandatory tour DDT choices. The resulting model takes the form a panel linear regression model with employment intensity, as the dependent variable on the supply side and a mixed logit model with combinations of Traffic Analysis Zones (TAZs) and time periods as alternatives on the demand side. The modeling methodology accounts for possible endogeneity between the two systems and also considers importance sampling methods to reduce the computational burden due to explosion of choice alternatives in the discrete choice model component. The model was used to explore supply demand interactions in the Southern California region. The results not only underscore the importance of the proposed integrated modeling framework but also provide several useful insights into the factors that influence the temporal profile of zonal employment and its interaction with daily travel choices.