Modeling Travelers’ Acceptance Of Traffic Information Systems:an Integrated Choice And Latent Variable Approach

China’s economy has experienced a spectacular growth in the last four decades.This rapid growth has also resulted in many traffic-related problems such as traffic congestion.In order to mitigate the effects of traffic congestion,the local authorities have adopted many solutions,including the provision of travel information to drivers through Advanced Traveler Information Systems(ATIS).Such systems can help travelers make better decisions,improve network performance and reduce traveler stress.However,their effectiveness depends highly on travelers’acceptance.Many methods have been adopted to investigate travelers’acceptance of ATIS.Recent studies adopted models of user acceptance of information technology to predict and explain drivers’acceptance of traffic information systems.Among these frameworks,the most commonly used is the Technology Acceptance Model(TAM).However,TAM is too general and does not consider drivers’response in specific traffic conditions or choice scenarios.This study aims to develop a comprehensive framework that integrates TAM into a route switching model of travelers’response to ATIS.In addition to the parsimonious TAM constructs(perceived usefulness,perceived ease of use and behavioral intention),the model is extended with latent variables that are specific to travelers’response to ATIS(perception of information quality,attitude towards route diversion and familiarity with the network).In line with the hierarchical relationships inherent in TAM,the model takes into account the causal relationships among the latent variables.All hypothesized relationships are based on strong theoretical and empirical background.The obtained framework is then incorporated into a route switching model to form a hierarchical integrated choice and latent variable(ICLV)model for traveler’s response to ATIS.The modeling process starts with the parsimonious TAM and gradually incorporates the other latent variables.In the model development,the proposed framework is compared with less complex and more familiar models.In order to demonstrate the superiority of the ICLV model over the discrete choice model(DCM),the obtained framework is compared with a binary logit model and mixed logit model with random parameter.In order to demonstrate the need to account for the causal relationships among latent variables,the proposed ICLV model framework is compared with a simpler ICVL model in which the latent variables are directly added to the utility function without any causal relationships.The models are calibrated using stated preference data collected by the author from road users in Dalian,China.The results show that the ICLV model has better explanatory and predictive power than the traditional DCM.In addition,the latent variables are among the strongest determinants of travelers’response.Furthermore,the hierarchical ICLV model has better explanatory power than the ICLV model that does not account for causal relationships.Regarding the causal relationships among latent variables,travelers’perception of information quality has a positive effect on perceived usefulness,perceived ease of use and attitude towards route diversion.Familiarity with the road network has a positive effect on the attitude towards route diversion and a negative impact on the perceived usefulness of ATIS.Perceived ease of use and the attitude towards route diversion have positive effects on perceived usefulness and behavioral intention.Perceived usefulness also positively affects behavioral intention.Attitude towards route diversion and behavioral intention have a direct and positive impact on route switching behavior.Route choice attributes(travel time saving,guidance on the best routing strategy and the number of signalized intersections)have direct effects on route switching behavior while individual characteristics(age,gender,income,etc.)indirectly affect route switching behavior through their impact on travelers’attitudes and perceptions.The main contributions of this study are:1)The development of a comprehensive framework which combines TAM(latent variable model)and observed variables(individual characteristics and SP scenarios).This could help explain travelers’ acceptance in more specific scenarios compared to earlier TAM studies.The resulting framework is a hierarchical ICLV model for travelers’ route switching behavior in response to travel information systems.Such framework can help strengthen the existing literature regarding the acceptance of information systems.2)Although hierarchical ICLV model(with causal relationships)has been used in travel behavior studies,its application in route choice behavior or travelers’response to traffic information is rare.This study addresses this limitation in the existing literature by applying a hierarchical ICLV model to investigate travelers’route switching behavior in response to Variable Message Sign(VMS).The use of such model helps better explain travel behavior compared to the ICLV model without causal relationships.3)From a practical point of view,the results from the proposed framework helped develop policies that aim to improve the design of traffic information systems.In this regard,the results from the hierarchical ICLV model are valuable in understanding the cognitive process leading to the formation of choices(acceptance of travel information and route choice)and how it can be used to design ATIS that are more suitable for travelers’ needs.