After completing the Trip Distribution Analysis, we need to determine what transportation system each of those travelers will use. Mode choice models estimate how many people will use public transit and how many will use private automobiles. The most common form of the mode choice model is the logit model. The logit mode choice relationship states that the probability of choosing a particular mode for a given trip is based on the relative values of a number of factors such as cost, level of service, and travel time. The most difficult part of employing the logit mode choice model is estimating the parameters for the variables in the utility function. The estimation is often accomplished using one or more multivariate statistical analysis programs to optimize the accuracy of estimates of the coefficients of several independent variables.
In regions where there are several alternative modes available, the mode choice model may require a special form called the "nested" logit. This form attempts to represent the choices presented to the traveler in a more structured manner. Nesting is necessary when there are major competing alternatives within, as well as between, principal modes.