2013 Session: 546

• Univariate Volatility-Based Models for Improving Quality of Travel Time Reliability Forecasting
  Abstract: The literature is rich in travel time prediction considering its importance in intelligent transportation system. Despite proliferation of advanced methodologies, modeling the uncertainty of traffic conditions is still a challenge, especially during congested situations. Travel time reliability associated with its time-dependent variation gives a way to measure the system performance and has been received extensive attention in recent years. In practice, one of the measures for travel time reliability is the identification of prediction interval, which has many potential applications in the development of systems that aimed at disseminating real time traffic information to travelers, such as the advanced traveler information systems (ATIS). From the management point of view, the prediction interval forecasts the unreliable traffic periods, enables the selection of proper strategy to avoid or release possible traffic congestion. GARCH model has been proved the ability of modeling the uncertainties in several literatures. However, it has some drawbacks in traffic forecasting. To improve the quality of travel time reliability forecasting, this paper proposes two univariate volatility models and compared their performance in generating high-quantity PIs. Travel time data collected from AVI stations located along U.S. Highway 290 in Houston, Texas is used to exam each model’s performance in travel time reliability forecasting. Study results indicate that all three models give reasonable prediction intervals that could be used to indicate the variability of future traffic conditions. The statistical analysis and forecasting results indicates that the proposed GIR-GARCH model outperformances the other two models in constructing better PIs.
  Authors: Zhang, Yanru; Sun, Ranye; Haghani, Ali; Zeng, Xiaosi
  Authors: Zhang, Yanru; Sun, Ranye; Haghani, Ali; Zeng, Xiaosi
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management; Safety and Human Factors
  Session: 546
  Paper Number: 13-0966
• Adaptive, Personalized Travel Information Systems: Bayesian Method to Learn Users’ Personal Preferences in Multimodal Transport Networks
  Abstract: Provision of personalized advice is an important objective in the development of advanced traveler information systems. In this paper, we propose a Bayesian method to incorporate learning of users’ personal travel preferences in a multi-modal routing system. Travel preferences are represented by parameters of link costs function the routing system uses to find optimal routes in an integrated multimodal network. Existing sampling-based methods to estimate posterior distributions require too much computation time for the incremental type of learning we are dealing with here. Therefore, we develop an approximation method that is based on sequential processing of parameters and systematic sampling of the parameter space. Data of repetitive travel choices of a large and representative sample of individuals are used to test the system. The results indicate that the system adapts rapidly already on the basis of a few observations from a user and that learning is effective. The efficiency of the algorithm allows the system to handle realistically sized learning problems with short response times even when many users are to be processed simultaneously. We conclude therefore that the approach is feasible and we identify problems for future research.
  Authors: Arentze, Theo A.
  Authors: Arentze, Theo A.
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management; Safety and Human Factors
  Session: 546
  Paper Number: 13-1325
• Freeway Path and Multi-Step-Ahead Travel Time Prediction Under Various Weather Scenarios
  Abstract: Accurate and reliable travel time prediction is very much needed for pre-trip planning and implementing traffic control strategies to reduce travel time and relieve traffic congestion. This research proposes an integrated model for path and multi-step ahead travel time prediction on freeways using various sources of real time traffic and weather data. The model’s performance is investigated and reported under various traffic and weather scenarios and especially under inclement weather conditions.
  Authors: Qiao, Wenxin; Haghani, Ali; Hamedi, Masoud
  Authors: Qiao, Wenxin; Haghani, Ali; Hamedi, Masoud
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management; Safety and Human Factors
  Session: 546
  Paper Number: 13-1339
• Empirical Identification and Quantification of Driver Anticipation Factor in Car-Following Behavior Modeling
  Abstract: In car-following behavior modeling, the anticipation factor represents the situation that drivers change the speed based on their predicted traffic condition, rather than the current condition. While current models address the anticipation factor as a constant, it is actually a dynamic variable in reality. This paper presents a new car-following model that incorporates the anticipation factor as a variable. The data from a field experiment reveal that the anticipation factor is a function of the drivers’ choice of braking process and the application of the advanced driving assistance system. The observed anticipation factor values are compared with the theoretical boundary, which is obtained from a linear stability analysis for the new car-following model. The comparison results show that the observed values are in the stable region, which indicates the modeling effort is consistent with the field observation. The new car-following model, in conjunction with the observed anticipation factor, is utilized in several simulation experiments in order to identify the influence of the anticipation factor on traffic flow. It is found that when the anticipation factor takes into effect on more drivers, it is easier for the traffic flow to recover to stability from local disturbances. The new car-following model is helpful for microscopic traffic simulation models and applications of Intelligent Transportation System (ITS).
  Authors: Liu, Hao; Wei, Heng; Yao, Zhuo; Ai, Qingyi
  Authors: Liu, Hao; Wei, Heng; Yao, Zhuo; Ai, Qingyi
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management; Safety and Human Factors
  Session: 546
  Paper Number: 13-2360
• Measuring User Awareness at Signalized Intersections
  Abstract: Drivers have limited awareness of changes in trip attributes or the performance of the traffic system. Due to non-utilitarian behavior and perceptual biases a distinctive amount of changes go unnoticed or are valued incorrectly, which makes drivers indifferent to changing traffic conditions to a certain extent. Quantifying user awareness and understanding the probability of behavioral response to changes is valuable input for road operators and traffic engineers designing traffic management measures. This paper presents the results of a field study on user awareness at signalized intersections. The study focused on the ability of drivers to observe and rightly value differences in the timing of traffic lights. Measurements of actual waiting times were compared with perceived waiting times derived from interviews. Results show that drivers’ perception of waiting time was on average fairly accurate, but widely variable, and that waiting times were systematically underestimated. Remarkably, the classification of deviations from the average waiting time showed that the vast majority of the respondents considered their waiting time ‘normal’ or shorter than they were used to. Although in terms of representativeness field studies are believed to be of great importance to perception studies, the selected approach for this study did not provide the expected data. Therefore it was not possible to provide definite answers related to user awareness at signalized intersections. Advantages, disadvantages and lessons learned are discussed in the paper and have been incorporated in a follow up study as much as possible.
  Authors: Vreeswijk, Jaap; van Berkum, Eric; van Arem, Bart
  Authors: Vreeswijk, Jaap; van Berkum, Eric; van Arem, Bart
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management; Safety and Human Factors
  Session: 546
  Paper Number: 13-2612
• Network-Sensitive Transport Modeling Framework for Evaluating Impacts of Network Disruptions on Traveler Choices Under Varying Levels of User Information Provision
  Abstract: There is considerable interest in the application of active traffic demand management (ATDM) and traveler information systems strategies to mitigate the adverse impacts of congestion and network disruptions. Such strategies and user information provision systems not only impact network performance through the modification of traveler route choices, but also through changes in the entire range of activity-travel choices such as activity generation, destination choice, mode choice, and time of day choice. The simulation of the impacts of alternative strategies on network performance therefore calls for the development and application of integrated modeling frameworks capable of reflecting the entire slate of activity-travel pattern adjustments that may occur in response to changes in network conditions and user information provision. This paper describes an integrated modeling framework wherein an activity-based travel demand model and a dynamic traffic assignment model are tightly coupled together with continuous information exchange between the models along the continuous time axis. The framework is enhanced to accommodate the possible impacts of alternative user information provision strategies on traveler choices and applied to a subregion in the Greater Phoenix metropolitan area to demonstrate the sensitivity of the model to network disruptions under alternative information provision scenarios. Model results are consistent with expectations and show that impacts of network disruption are substantially mitigated in the presence of traveler information systems. Further, the model results show that workers – who have more rigid work schedules and locations – are more greatly impacted by network disruptions than non-workers who do not have such constraints. Integrated modeling tools such as that described in this paper offer promise for evaluating emerging operational and policy strategies aimed at influencing traveler choices.
  Authors: Konduri, Karthik Charan; Pendyala, Ram M.; You, Daehyun; Chiu, Yi-Chang; Hickman, Mark D.; Noh, Hyunsoo; Gardner, Brian; Waddell, Paul; Wang, Liming
  Authors: Konduri, Karthik Charan; Pendyala, Ram M.; You, Daehyun; Chiu, Yi-Chang; Hickman, Mark D.; Noh, Hyunsoo; Gardner, Brian; Waddell, Paul; Wang, Liming
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management; Safety and Human Factors
  Session: 546
  Paper Number: 13-3366
• Route choice model and algorithm for dynamic assignment in overcrowded bus networks with real-time information at stops
  Abstract: The paper presents a route choice model and algorithm for dynamic assignment in congested, i.e. overcrowded, transit networks where it is assumed that passengers are supported with real-time information on carrier arrivals at stops. If the stop layout is such that passenger congestion results in First-In-First-Out (FIFO) queues, a new formulation is devised for calculating waiting times, total travel times and route splits. Numerical results for a simple example network show the effect of congestion and information on route choice, both in terms of total travel time to the destination and route split. Moreover, it is shown that while the provision of information does not lead to a remarkable decrease in total travel time, with the exception of some particular instances, it changes the travel behaviour of passengers that seem to be more averse to queuing at later stages of their journey and, thus, prefer to interchange at less congested stations. This result suggests that information can help to achieve a more equilibrate use of the network.
  Authors: Trozzi, Valentina; Gentile, Guido; Kaparias, Ioannis; Bell, Michael G.H.
  Authors: Trozzi, Valentina; Gentile, Guido; Kaparias, Ioannis; Bell, Michael G.H.
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management; Safety and Human Factors
  Session: 546
  Paper Number: 13-4413
• Empirical Identification and Quantification of Driver Anticipation Factor in Car-Following Behavior Modeling
  Authors: Yao, Zhuo
  Keywords: poster presentation; poster design; poster template
  Authors: Yao, Zhuo
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Operations and Traffic Management; Safety and Human Factors
  Session: 546
  Paper Number: 13-2360
• Route choice model and algorithm for dynamic assignment in overcrowded bus networks with real-time information at stops
  Authors: Kaparias, Ioannis
  Authors: Kaparias, Ioannis
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Operations and Traffic Management; Safety and Human Factors
  Session: 546
  Paper Number: 13-4413
• Freeway Path and Multi-Step-Ahead Travel Time Prediction Under Various Weather Scenarios
  Authors: Qiao, Wenxin
  Authors: Qiao, Wenxin
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Operations and Traffic Management; Safety and Human Factors
  Session: 546
  Paper Number: 13-1339