The Propagation of Uncertainty Through Travel Demand Models
Yong Zhao and Kara Kockelman, University of Texas at Austin, October 2000, 46 pp. (167804-1)
The future operations of transportation systems involve enormous uncertainty – in both input and model parameters. This work investigates the stability of contemporary transport demand model output by quantifying variability in model input, such as zonal socioeconomic data and trip generation rates. It simulates the propagation of their variation through a series of common demand models over a simplified twenty-five-zone network. The results suggest that uncertainty may be compounded over a series of models and highly correlated across output. The propagated uncertainty varies remarkably in link flows, thus, some link flows may be more uncertain than others. The final step model, trip assignment, may reduce prior increased uncertainty through the first three steps, but generally could not lessen the uncertainty lower than the input uncertainty. Mispredictions at early stages (e.g., trip generation) in multi-stage models appear to amplify across later stages; thus, improvements to these models and their estimates are sorely needed.
Keywords: Demand Models, Planning, Sequential Modeling, Trip Generation
ENTIRE REPORT (Adobe Acrobat File – 501KBytes)