Comparison of Two Transportation Network Equilibrium Modeling Approaches

Existing methodologies for predicting traffic flows on transportation networks involve a sequential process, often with four stages: trip generation, trip distribution, modal split, and traffic assignment. Although the sequential approach has been applied to hundreds of transportation studies throughout the world for the last four decades, and is still being used in practice today, it has an inherent weakness. That is, its predictions need not be internally consistent. This deficiency has motivated attempts to predict all four steps simultaneously. The objective of this paper is to compare the simultaneous approach, using a Simultaneous Transportation Equilibrium Model (STEM) and the conventional sequential approach, using the Texas Demand Model developed and implemented by the Texas Department of Transportation (TxDOT). This was achieved by applying STEM to the urban transportation network of Tyler, Tex., and then comparing the accuracy of the resulting traffic network flows with the “official” TxDOT estimates. The application results showed that predictions of STEM were consistently better than those of the TxDOT by an average improvement of 25%. These results are certainly encouraging in terms of future application of STEM methodology to other real-world large-scale transportation networks.