Incorporating social impact on new product adoption in choice modeling: A case study in green vehicles

While discrete choice analysis is prevalent in capturing consumer preferences and describing their choice behaviors in product design, the traditional choice modeling approach assumes that each individual makes independent decisions, without considering the social impact. However, empirical studies show that choice is social – influenced by many factors beyond engineering performance of a product and consumer attributes. To alleviate this limitation, we propose a new choice modeling framework to capture the dynamic influence from social networks on consumer adoption of new products. By introducing social influence attributes into a choice utility function, social network simulation is integrated with the traditional discrete choice analysis in a three-stage process. Our study shows the need for considering social impact in forecasting new product adoption. Using hybrid electric vehicles as an example, our work illustrates the procedure of social network construction, social influence evaluation, and choice model estimation based on data from the National Household Travel Survey. Our study also demonstrates several interesting findings on the dynamic nature of new technology adoption and how social networks may influence hybrid electric vehicle adoption.     

A bi-objective user equilibrium model of travel time reliability in a road network

Travel time, travel time reliability and monetary cost have been empirically identified as the most important criteria influencing route choice behaviour. We concentrate on travel time and travel time reliability and review two prominent user equilibrium models incorporating these two factors. We discuss some shortcomings of these models and propose alternative bi-objective user equilibrium models that overcome the shortcomings. Finally, based on the observation that both models use standard deviation of travel time within their measure of travel time reliability, we propose a general travel time reliability bi-objective user equilibrium model. We prove that this model encompasses those discussed previously and hence forms a general framework for the study of reliability related user equilibrium. We demonstrate and validate our concepts on a small three-link example.     

Development of an indicator to assess the spatial fit of discrete choice models

Discrete choice models are increasingly implemented using geographical data. When this is the case, it may not be sufficient to project market shares accurately, but also to correctly replicate the spatial pattern of choices. Analysts might then be interested in assessing the results of a model’s fit relative to the spatial distribution of the observed responses. While canonical approaches exist for the exploratory spatial analysis of continuous variables, similar tools have not been widely implemented for discrete choice models, where the variable of interest is categorical. For this reason, despite recent progress with spatial models for discrete outcomes, there is still not a simple and intuitive tool to assess the quality of the spatial fit of a discrete choice model. The objective of this paper is to introduce a new indicator of spatial fit that can be applied to the results of discrete choice models. Utility of the indicator is explored by means of numerical experiments and then demonstrated by means of a case study of vehicle ownership in Montreal, Canada.     

Exploring the influence of built environment on tour-based commuter mode choice: A cross-classified multilevel modeling approach

Understanding travel behavior and its relationship to urban form is vital for the sustainable planning strategies aimed at automobile dependency reduction. The objective of this study is twofold. First, this research provides additional insights to examine the effects of built environment factors measured at both home location and workplace on tour-based mode choice behavior. Second, a cross-classified multilevel probit model using Bayesian approach is employed to accommodate the spatial context in which individuals make travel decisions. Using Washington, D.C. as our study area, the home-based work (Home-work) tour in the AM peak hours is used as the analysis unit. The empirical data was gathered from the Washington-Baltimore Regional Household Travel Survey 2007–2008. For parameter estimation, Bayesian estimation method integrating Markov Chain Monte Carlo (MCMC) sampling is adopted. Our findings confirmed the important role that the built environment at both home location and work ends plays in affecting commuter mode choice behavior. Meanwhile, a comparison of different model results shows that the cross-classified multilevel probit model offers significant improvements over the traditional probit model. The results are expected to give a better understanding on the relationship between the built environment and commuter mode choice behavior.     

Consistent formulation of network equilibrium with stochastic flows

Traffic flows in real-life transportation systems vary on a daily basis. According to traffic flow theory, such variability should induce a similar variability in travel times, but this “internal consistency” is generally not captured by existing network equilibrium models. We present an internally-consistent network equilibrium approach, which considers two potential sources of flow variability: (i) daily variation in route choice and (ii) daily variation in origin–destination demand. We particularly aspire to a flexible formulation that permits alternative statistical assumptions, which allows the best fit to be made to observed variability data in particular applications. Joint probability distributions of route—and therefore link—flows are derived under several assumptions concerning stochastic driver behavior. A stochastic network equilibrium model with stochastic demands and route choices is formulated as a fixed point problem. We explore limiting cases which allow an equivalent convex optimization problem to be defined, and finally apply this method to a real-life network of Kanazawa City, Japan.     

A choice experiment on alternative fuel vehicle preferences of private car owners in the Netherlands

This paper presents results of an online stated choice experiment on preferences of Dutch private car owners for alternative fuel vehicles (AFVs) and their characteristics. Results show that negative preferences for alternative fuel vehicles are large, especially for the electric and fuel cell car, mostly as a result of their limited driving range and considerable refueling times. Preference for AFVs increases considerably with improvements on driving range, refueling time and fuel availability. Negative AFV preferences remain, however, also with substantial improvements in AFV characteristics; the remaining willingness to accept is on average € 10,000–€ 20,000 per AFV. Results from a mixed logit model show that consumer preferences for AFVs and AFV characteristics are heterogeneous to a large extent, in particular for the electric car, additional detour time and fuel time for the electric and fuel cell car. An interaction model reveals that annual mileage is by far the most important factor that determines heterogeneity in preferences for the electric and fuel cell car. When annual mileage increases, the preference for electric and fuel cell cars decreases substantially, whilst the willingness to pay for driving range increases substantially. Other variables such as using the car for holidays abroad and the daily commute also appear to be relevant for car choice.     

Inconsistent choices in Stated Choice data;Use of the logit scaling approach to handle resulting variance increases

The scaling approach is a statistical estimation method that allows for differences in the amount of unexplained variation in different types of data, which can then be used together in the analysis. This approach has been mostly used in context of combining Stated Preference and Revealed Preference data, but has also been used as a method of identifying systematic differences in the variance of choices within a single Stated Preference data set, e.g. for investigation of learning and fatigue effects. This paper investigates whether a scaling approach is suitable for handling inconsistencies in Stated Choice data. Both the number of inconsistent choices, based on a test of violations of the transitivity axiom, and education are used as scaling variables. Scaling effects appear to exist due to inconsistent choices, and the amount of unexplained variance is shown to increase as the number of inconsistent choice increase. Scaling due to inconsistencies significantly improves the models and reduces the valuations of travel time. In addition, the scaling approach makes the valuations of travel time from the Stated Choice data more consistent with the valuations from Contingent Valuation data included in the same study. In spite of the fact that education is the only significant explanatory variable for the number of inconsistent choices, scaling due to education gives no significant improvement in the model.     

考虑老年人出行行为的公交票价补贴政策研究

北京市现行的老年人公交免费政策在保障老年人日常出行的同时,诱发了老年人无谓出行、高峰出行、冒乘出行等问题. 针对不同出行目的、出行距离及出发时刻下老年人出行行为偏好,基于北京市第五次综合交通调查数据,构建多场景下老年人出行方式选择MNL (Multinominal Logit)模型,分析错峰优惠、阶梯式津贴等补贴政策对多方式分担率的影响,并从公共交通运能利用率、政府补贴和老年人公交支出等方面评价政策效果. 结果表明：相对现行政策,两种政策均可有效调节老年人出行行为,前者能够缓解政府财政补贴压力,但造成老年人公交支出的上升;后者侧重保护老年人群的社会福利,对行为的引导效果更显著,但仍需大量的财政补贴投入.     

Exploration of route choice behavior with advanced traveler information using neural network concepts

A model of driver's route choice behavior under advanced traveler information system (ATIS) is developed based on data collected from learning experiments using interactive computer simulation. The experiment subjected drivers to 32 simulated days in which they were to choose between the freeway or a side road. A neural network model is used as a convenient modeling technique in this initial phase of the analysis. The results indicated that most subjects made route choices based mainly on their recent experiences. It was also demonstrated that route choice behaviors are related to the personal characteristics as well as the characteristics of the respective routes. Travel experiences have less effect on the choice of the side road compared to the freeway and the results indicate that the prediction accuracy of the model, the acceptance rate of advice, and the quality of advice are closely correlated. The model developed here was for advice consistently provided at a level of 75 percent accuracy. The paper concludes with a discussion of experimental limitations and suggestions for future research.     

Valuing reductions in environmental pollution in a residential location context

A stated preference ranking experiment is designed to estimate the willingness-to-pay (WTP) for reducing the amount of atmospheric pollution in a group-based residential location context. Important issues are the proper definition of the context and the variable metric for the environmental attribute. Sample members were asked to rank 10 options arising from variations in the attributes travel time to work and to study, rent of the house and an environmental attribute (the number of days of Alert, in terms of concentration of PM₁₀, at a dwelling’s location). Multinomial logit models based on a consistent microeconomic framework were estimated for various stratifications of the data (income, pollution sensitivity, and type of dwelling currently inhabited). From these subjective values of time and WTP were derived for reductions in the number of days of alert and hence the amount of pollutant concentration at a given location. The WTP came out at about 1% of the family income for reducing one contingence day per year; this is approximately 60% higher than an estimate reported for the city of Edmonton, Canada, but the average PM₁₀ concentration in Santiago is about six times higher.