Evaluating the Impacts of Parameter Uncertainty in Transportation Demand Models
Gregory S. Macfarlane 
The inherent uncertainty in travel forecasting models — arising from potential and unkown errors in input data, parameter estimation, or model formulation — is receiving increasing attention from the scholarly and practicing community. In this research, we investigate the variance in forecasted traffic volumes resulting from varying the mode and destination choice parameters in an advanced trip-based travel demand model. Using Latin hypercube sampling to construct several hundred combinations of parameters across the plausible parameter space, we introduce substantial changes to implied travel impedances and modal utilities. However, the aggregate effects of of these changes on forecasted traffic volumes is small, with a variance of approximately 1 percent on high-volume facilities. It is likely that in this example — and perhaps in others — the static network assignment places constraints on the possible volume solutions and limits the practical impacts of parameter uncertainty. Further research should examine the robustness of this finding to other less constrained networks and to activity-based travel model frameworks.
1 Introduction
The inherent accuracy and uncertainty in travel forecasting models is receiving increasing attention from the scholarly and practicing community. Given that such models are used in the allocation of billions of dollars of infrastructure financing each year, the financial risks for inaccurate or imprecise forecasts are high (Flyvbjerg et al., 2005; Voulgaris, 2019).
Transportation demand forecasting models, like other mathematical-statistical models, might be abstracted to the following basic form,
\[ y = f(X, \beta) \]where \(y\) is the variable being predicted based on input data \(X\), moderated through a specific functional form \(f()\) and parameters \(\beta\). Three general sources of error may lead a forecast value \(\hat{y}\) to differ from the “true” or “actual” value of \(y\) (Rasouli & Timmermans, 2012):
- The input data \(X\) might contain errors, due to inaccuracies in the base year, or an inaccurate projection of land use, petroleum price, or other key input variable. This was among the primary issues identified by Hoque et al. (2021) in a historical analysis of the accuracy of travel forecasts.
- The model form \(f()\) may be improperly specified. Variables that play a major role in travel behavior may not be included due to lack of information, or the unobserved error components may have a different correlation than was assumed during model development. A detailed description of specifying mode choice model variables and nesting of error structures is given by Koppelman & Bhat (2006).
- The parameter estimates \(\hat{\beta}\) of the “true” parameters \(\beta\) may have incorrect values. This may be because the parameters were estimated on an improperly specified model \(f()\), or because the estimation dataset was improperly weighted.
Of these potential sources of error, only the third is substantively addressed in classical statistics. The standard errors of the model parameter estimates in a theoretical perspective address the parameter uncertainty question to a great degree. Yet even this source of uncertainty has been largely ignored in transportation forecasts, and model development documentation often elides the variance in these values completely (National Academies of Sciences, Engineering, and Medicine., 2012). Zhao & Kockelman (2002) examined the effects of this parameter uncertainty in a trip-based model of a contrived 25 zone region, but a systemic analysis of this uncertainty in a practical model is not common.
In this research, we investigate the uncertainty in traffic forecasts resulting from plausible parameter uncertainty in an advanced trip-based transportation demand model. Using a Latin hypercube sampling (LHS) methodology, we simulate one hundred potential parameter sets for a combined mode and destination choice model in Roanoke, Virginia, USA. We then assign the resulting trip matrices to the highway network for the region and evaluate the PM and daily assigned traffic volumes alongside the variation in implied impedance and accessibility.
This paper proceeds first with a description of the model design and simulation sampling methodology in Chapter 3, followed by a discussion of the variation in mode, destination, and traffic performance measures in Chapter 4. The paper concludes in Chapter 5 with a summary of the key findings alongside a presentation of limitations and related indications for future research.