Abstract
Transportation simulation models can be used to calculate social and environmental indices of transportation system performance. The outputs of the simulation models are subject to random error, and this error is transmitted to these indices. Random error is a consequence of inaccuracies in calibration and input data, and simplified model theory. The error in the indices can often be estimated prior to model execution, thus allowing the analyst to determine whether the complexity of the model is appropriate for his particular needs. Examples of how noise pollution, air pollution, vehicle-miles-traveled, and vehicle-hours-traveled indices are affected by changing the zone size of an assignment model illustrate how knowledge of index error may be used to design a transportation simulation.
Get full access to this article
View all access options for this article.