Most of the previous literature on combined trip distribution and assignment problems has focused on the logit and entropy distribution models. Though these models are satisfactory for many applications, they are incapable of handling situations in which the observed trip patterns are represented by both compulsory (captive) and discretionary (free) travel behavior. Consequently, the use of a dogit distribution formula in the construction of a combined trip distribution and assignment model is suggested. The new version of the combined model can itself be reformulated as an equivalent mathematical programming problem so that the equilibrium conditions on the network and the dogit destination demand functions can be derived as the kuhn-tucker conditions of the proposed programming problem. Moreover, this equivalent mathematical program turns out to be a convex programming problem with linear constraints, a great advantage from the computational aspect. Numerical experiment indicates that this behaviorally sound combined model can be used in a realistic application at areasonable cost and within a reasonable time period. This paper appears in transportation research record no. 1285, Transportation forecasting 1990.
Samenvatting