Search

Deep neural networks (DNNs) frequently present behaviorally irregular patterns, significantly limiting their practical potentials and theoretical validity in travel behavior modeling. This study proposes strong and weak behavioral regularities as novel metrics to evaluate the monotonicity of individual demand functions (a.k.a. law of demand), and further designs a constrained optimization framework with six gradient regularizers to enhance DNNs' behavioral regularity. The proposed framework is applied to travel survey data from Chicago and London to examine the trade-off between predictive power and behavioral regularity for large vs. small sample scenarios and in-domain vs. out-of-domain generalizations. The results demonstrate that, unlike models with strong behavioral foundations such as the multinomial logit, the benchmark DNNs cannot guarantee behavioral regularity. However, gradient regularization (GR) increases DNNs' behavioral regularity by around 6 percentage points (pp) while retaining their relatively high predictive power. In the small sample scenario, GR is more effective than in the large sample scenario, simultaneously improving behavioral regularity by about 20 pp and log-likelihood by around 1.7%. Comparing with the in-domain generalization of DNNs, GR works more effectively in out-of-domain generalization: it drastically improves the behavioral regularity of poorly performing benchmark DNNs by around 65 pp, indicating the criticality of behavioral regularization for enhancing model transferability and application in forecasting. Moreover, the proposed framework is applicable to other NN-based choice models such as TasteNets. Future studies could use behavioral regularity as a metric along with log-likelihood in evaluating travel demand models, and investigate other methods to further enhance behavioral regularity when adopting complex machine learning models.

We investigate preferences for COVID-19 vaccines using data from a stated choice survey conducted in the US in March 2021. To analyse the data, we embed the Choquet integral, a flexible aggregation operator for capturing attribute interactions under monotonicity constraints, into a mixed logit model. We find that effectiveness is the most important vaccine attribute, followed by risk of severe side effects and protection period. The attribute interactions reveal that non-pecuniary vaccine attributes are synergistic. Out-of-pocket costs are independent of effectiveness, incubation period, and mild side effects but exhibit moderate synergistic interactions with other attributes. Vaccine adoption is significantly more likely among individuals who identify as male, have obtained a bachelor’s degree or a higher level of education, have a high household income, support the democratic party, had COVID-19, got vaccinated against the flu in winter 2020/21, and have an underlying health condition.

Various measures have been proposed and validated to assess environmental motivation and explain peoples’ consumer behavior. However, most of the measures are rather complex, sometimes comprising dozens of items. In order to overcome the associated response burden, the goal of our research is to validate a much simpler measure of environmental motivation, namely the measure of Climate Change-Stage of Change. To do so we analyze data from a discrete choice experiment in which drivers decide to purchase a car with different levels of CO2 emissions and we also measure their environmental motivation with three alternative measures. The results show that environmental motivation assessed with Climate Change-Stage of Change explains the choices in the experiment as well as with more complex measures. Our findings have substantial implications for researchers as they may be able to assess climate-relevant motivation – a significant factor for many consumer choices – with a single question.

Travel demand forecasting models play an important role in guiding policy, planning, and design of transportation systems. There is no shortage of literature critiquing the accuracy of model forecasts (see, for example, Pickrell, 1989; Wachs, 1990; Pickrell, 1992; Flyvbjerg, Skamris Holm, and Buhl 2005; Richmond, 2005; Flyvbjerg, 2007; Bain, 2009; Parthasarathi and Levinson, 2010; Welde and Odeck, 2011; Hartgen, 2013; Nicolaisen and Driscoll, 2014; Schmitt, 2016; Odeck and Welde, 2017, and Voulgaris, 2019), not to mention several high-profile lawsuits (Saulwick 2014, Stacey 2015, Rubin 2018). Many researchers and practitioners feel more can be done to advance rigorous travel analysis methods for the public good (see, e.g., zephyrtransport.org). Motivated by these critiques, a two-day, NSF-funded workshop was held at UC Berkeley in the Spring of 2017 to engage in a fundamental review of the state of the art in travel demand modeling, to discuss the future of the field, and to propose new directions and processes for advancing the science.Travel demand forecasting is an inherently practical enterprise. While academics drive the fundamental research, the users of travel demand models and forecasts are typically government agencies and transport operators that use the models to inform long-range investment, funding, and planning decisions. Private firms play a key role in assisting the agencies in both development and application of the models, and, more recently, high-tech firms have entered the development fray. While all of these actors have important roles in advancing the science of the field, in this report we focus our attention primarily on the academic side of the enterprise, consistent with the orientation of the funding agency (NSF), and in order to make the task manageable. That said, other sectors are represented in various parts of this report as they interface with academics or play particularly central roles in our proposals for advancing the science.

Travel demand forecasting models play an important role in guiding policy, planning, and design of transportation systems. There is no shortage of literature critiquing the accuracy of model forecasts (see, for example, Pickrell, 1989; Wachs, 1990; Pickrell, 1992; Flyvbjerg, Skamris Holm, and Buhl 2005; Richmond, 2005; Flyvbjerg, 2007; Bain, 2009; Parthasarathi and Levinson, 2010; Welde and Odeck, 2011; Hartgen, 2013; Nicolaisen and Driscoll, 2014; Schmitt, 2016; Odeck and Welde, 2017, and Voulgaris, 2019), not to mention several high-profile lawsuits (Saulwick 2014, Stacey 2015, Rubin 2018). Many researchers and practitioners feel more can be done to advance rigorous travel analysis methods for the public good (see, e.g., zephyrtransport.org). Motivated by these critiques, a two-day, NSF-funded workshop was held at UC Berkeley in the Spring of 2017 to engage in a fundamental review of the state of the art in travel demand modeling, to discuss the future of the field, and to propose new directions and processes for advancing the science.Travel demand forecasting is an inherently practical enterprise. While academics drive the fundamental research, the users of travel demand models and forecasts are typically government agencies and transport operators that use the models to inform long-range investment, funding, and planning decisions. Private firms play a key role in assisting the agencies in both development and application of the models, and, more recently, high-tech firms have entered the development fray. While all of these actors have important roles in advancing the science of the field, in this report we focus our attention primarily on the academic side of the enterprise, consistent with the orientation of the funding agency (NSF), and in order to make the task manageable. That said, other sectors are represented in various parts of this report as they interface with academics or play particularly central roles in our proposals for advancing the science.

Travel demand forecasting models play an important role in guiding policy, planning, and design of transportation systems. There is no shortage of literature critiquing the accuracy of model forecasts (see, for example, Pickrell, 1989; Wachs, 1990; Pickrell, 1992; Flyvbjerg, Skamris Holm, and Buhl 2005; Richmond, 2005; Flyvbjerg, 2007; Bain, 2009; Parthasarathi and Levinson, 2010; Welde and Odeck, 2011; Hartgen, 2013; Nicolaisen and Driscoll, 2014; Schmitt, 2016; Odeck and Welde, 2017, and Voulgaris, 2019), not to mention several high-profile lawsuits (Saulwick 2014, Stacey 2015, Rubin 2018). Many researchers and practitioners feel more can be done to advance rigorous travel analysis methods for the public good (see, e.g., zephyrtransport.org). Motivated by these critiques, a two-day, NSF-funded workshop was held at UC Berkeley in the Spring of 2017 to engage in a fundamental review of the state of the art in travel demand modeling, to discuss the future of the field, and to propose new directions and processes for advancing the science.Travel demand forecasting is an inherently practical enterprise. While academics drive the fundamental research, the users of travel demand models and forecasts are typically government agencies and transport operators that use the models to inform long-range investment, funding, and planning decisions. Private firms play a key role in assisting the agencies in both development and application of the models, and, more recently, high-tech firms have entered the development fray. While all of these actors have important roles in advancing the science of the field, in this report we focus our attention primarily on the academic side of the enterprise, consistent with the orientation of the funding agency (NSF), and in order to make the task manageable. That said, other sectors are represented in various parts of this report as they interface with academics or play particularly central roles in our proposals for advancing the science.

The standard Gibbs sampler of Mixed Multinomial Logit (MMNL) models involves sampling from conditional densities of utility parameters using Metropolis-Hastings (MH) algorithm due to unavailability of conjugate prior for logit kernel. To address this non-conjugacy concern, we propose the application of P\'olygamma data augmentation (PG-DA) technique for the MMNL estimation. The posterior estimates of the augmented and the default Gibbs sampler are similar for two-alternative scenario (binary choice), but we encounter empirical identification issues in the case of more alternatives ($J \geq 3$)., Comment: arXiv admin note: text overlap with arXiv:1904.03647

Variational Bayes (VB) methods have emerged as a fast and computationally-efficient alternative to Markov chain Monte Carlo (MCMC) methods for scalable Bayesian estimation of mixed multinomial logit (MMNL) models. It has been established that VB is substantially faster than MCMC at practically no compromises in predictive accuracy. In this paper, we address two critical gaps concerning the usage and understanding of VB for MMNL. First, extant VB methods are limited to utility specifications involving only individual-specific taste parameters. Second, the finite-sample properties of VB estimators and the relative performance of VB, MCMC and maximum simulated likelihood estimation (MSLE) are not known. To address the former, this study extends several VB methods for MMNL to admit utility specifications including both fixed and random utility parameters. To address the latter, we conduct an extensive simulation-based evaluation to benchmark the extended VB methods against MCMC and MSLE in terms of estimation times, parameter recovery and predictive accuracy. The results suggest that all VB variants with the exception of the ones relying on an alternative variational lower bound constructed with the help of the modified Jensen's inequality perform as well as MCMC and MSLE at prediction and parameter recovery. In particular, VB with nonconjugate variational message passing and the delta-method (VB-NCVMP-Delta) is up to 16 times faster than MCMC and MSLE. Thus, VB-NCVMP-Delta can be an attractive alternative to MCMC and MSLE for fast, scalable and accurate estimation of MMNL models.

Mobility-on-Demand (MoD) systems are generally designed and analyzed for a fixed and exogenous demand, but such frameworks fail to answer questions about the impact of these services on the urban transportation system, such as the effect of induced demand and the implications for transit ridership. In this study, we propose a unified framework to design, optimize and analyze MoD operations within a multimodal transportation system where the demand for a travel mode is a function of its level of service. An application of Bayesian optimization (BO) to derive the optimal supply-side MoD parameters (e.g., fleet size and fare) is also illustrated. The proposed framework is calibrated using the taxi demand data in Manhattan, New York. Travel demand is served by public transit and MoD services of varying passenger capacities (1, 4 and 10), and passengers are predicted to choose travel modes according to a mode choice model. This choice model is estimated using stated preference data collected in New York City. The convergence of the multimodal supply-demand system and the superiority of the BO-based optimization method over earlier approaches are established through numerical experiments. We finally consider a policy intervention where the government imposes a tax on the ride-hailing service and illustrate how the proposed framework can quantify the pros and cons of such policies for different stakeholders., Comment: 31 pages, 11 figures. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0

In response to concerns related to climate change, and an attempt to encourage more sustainable behavior, individuals are often provided with information on greenhouse gas emissions (GHGs) of consumer items, such as personal vehicles. Currently in the US, information on vehicle efficiency is provided as grams of carbon dioxide (CO2) per mile. Previous research presenting CO2 as a mass and testing willingness-to-pay through Discrete Choice Experiment has found that such information can influence vehicle choice. However, other research has questioned whether how this information is presented might affect choice. That research argues that CO2 emission information generally lacks contextualization that allows for interpretation. As well, it argues that the type of contextualization may affect choices. That research though did not test willingness-to-pay and the strength of its influence is not clear. In addition, research exists that argues that using pro-social, as opposed to financial, contextualization might be more influential on people’s choices. Thus, the purpose of this paper is to build on these previous findings on how CO2 emissions are presented to determine whether changing how that information impacts vehicle choice with a Discrete Choice Experiment of vehicle choice analyzed using latent class modeling. No previous study has so robustly studied the influence that different framings might have on vehicle purchase. Five different methods of presenting CO2 information are tested in this experiment: CO2 emissions as grams per mile (current method), CO2 emissions as pounds per year (consistent imperial units), CO2 emissions as tons per year (yearly contextualization), an annual tax on CO2 (yearly financial contextualization), and CO2 as a percentage of the 2025 US EPA reduction target of 26% from 2005 levels (social goal contextualization). Results demonstrate that the current method results in lowest willingness to pay for CO2 emission reductions, while the soc

In response to concerns related to climate change, and an attempt to encourage more sustainable behavior, individuals are often provided with information on greenhouse gas emissions (GHGs) of consumer items, such as personal vehicles. Currently in the US, information on vehicle efficiency is provided as grams of carbon dioxide (CO2) per mile. Previous research presenting CO2 as a mass and testing willingness-to-pay through Discrete Choice Experiment has found that such information can influence vehicle choice. However, other research has questioned whether how this information is presented might affect choice. That research argues that CO2 emission information generally lacks contextualization that allows for interpretation. As well, it argues that the type of contextualization may affect choices. That research though did not test willingness-to-pay and the strength of its influence is not clear. In addition, research exists that argues that using pro-social, as opposed to financial, contextualization might be more influential on people’s choices. Thus, the purpose of this paper is to build on these previous findings on how CO2 emissions are presented to determine whether changing how that information impacts vehicle choice with a Discrete Choice Experiment of vehicle choice analyzed using latent class modeling. No previous study has so robustly studied the influence that different framings might have on vehicle purchase. Five different methods of presenting CO2 information are tested in this experiment: CO2 emissions as grams per mile (current method), CO2 emissions as pounds per year (consistent imperial units), CO2 emissions as tons per year (yearly contextualization), an annual tax on CO2 (yearly financial contextualization), and CO2 as a percentage of the 2025 US EPA reduction target of 26% from 2005 levels (social goal contextualization). Results demonstrate that the current method results in lowest willingness to pay for CO2 emission reductions, while the soc