A modified logit type parking choice model

The adoption of Logit model in transportation has led to widespread use of choice models to describe the influence of the characteristics of decision-makers and the attributes of alternatives and choice. The problem of modeling the intermediate destination choice for mixed mode trips is discussed as one example of a distribution model based on an activity chain. It is shown how such models, if based on a C-Logit distribution, can be implemented efficiently by the use of simple algebraic matrix multiplication. The main purpose of this paper is to extend the mixed mode C-Logit model for intermediate destination choice to a similar model in which an explicit capacity is associated with each intermediate destination. This leads to a more realistic approach for modeling park & ride trips, since the capacity or an intermediate zone is limited by parking, capacity it offers. we first show that this model can he expressed as a convex minimization problem. Then we look at the dual formulation and show that it can be solved efficiently with a coordinate descent method. A solution algorithm is proposed in detail.