AN EVALUATION OF TRAFFIC ASSIGNMENT PREDICTIONS USING ROUTE CHOICE ANALYSIS AND ERROR DECOMPOSITION

A statistical model has been developed to decompose the overall estimation error in link loads predicted with the usual assignment procedures. This decomposition is derived by comparing load estimates from assignment models (all-or-nothing, equilibrium, stochastic multiple route) and levels of detail (fine, medium, coarse). The absolute and relative contribution of the potential error sources to the total error-sum-of squares has been estimated for the primary roads of the car network of Eindhoven (the Netherlands). Special attention is given to the bias in the load predictions. From the empirical findings it can be learned that assignment load outcomes are very sensitive to the assignment model type as well as of spatial detail, whereas the trip input (if unbiased) is less important. An explanation of some findings is given using a detailed analysis of route choice predictions. (Author/TRRL)