Traffic congestion in urban areas presents a major challenge to efficient transportation systems. Recent advancements in traffic management provide promising solutions, with perimeter control emerging as a technique to tackle network-wide congestion. However, it is crucial to identify geographically connected homogeneously congested areas for effective implementation. This research explores the application of clustering techniques, particularly geographical self-organizing maps (GeoSOM), to identify spatially connected and homogeneously congested areas within transportation networks. While GeoSOM has found applications across various domains, its adaptation to transportation networks for congestion clustering is novel. This study introduces and implements an adaptation of the GeoSOM algorithm tailored for the large-scale urban environment of downtown Los Angeles. Its performance is assessed through a comparative evaluation with two other clustering algorithms, namely DBSCAN and K-means. The results demonstrate that GeoSOM surpasses other clustering algorithms, exhibiting improvements of up to 43% in traffic density variance, up to 61% in the spatial quantization error, and 15% in the quantization error. This finding demonstrates that the proposed clustering algorithm is effective in identifying a spatially homogeneous congested area within a large-scale transportation network.

Traffic estimation using probe vehicle data is a crucial aspect of traffic management as it provides real-time information about traffic conditions. This study introduced a novel framework for traffic density estimation using Temporal Convolutional Network (TCN) for time series data. The study used two datasets collected from a three-leg intersection in Greece and a four-leg intersection in Germany. The model was built to predict the density in an approach of the signalized intersection using features extracted from the other approaches. The results showed that the highest accuracy was achieved when only probe vehicle data was used. This implies that relying solely on probe vehicle data from two approaches can effectively predict traffic density in the third approach, even when the Market Penetration Rate (MPR) is low. The results also indicated that having Signal Phase and Timing (SPaT) information may not be necessary for high accuracy in traffic estimation and that as the MPR increases, the model becomes more predictable.

The transportation system has evolved into a complex Cyber Physical System (CPS) with the introduction of wireless communication and the emergence of connected travelers and Connected Automated Vehicles (CAVs). The talk will discuss the challenges associated with multi-modal transportation system optimization and modeling, large-scale integrated modeling of the transportation and communication systems, some research in the area of multi-objective CAV optimization, some research in CAV-enabled traffic signal control, and the modeling and optimization of battery electric vehicles and hybrid electric vehicles.

This research develops and evaluates an Eco-Cooperative Adaptive Cruise Control system in proximity of traffic signals (ECO-CACC-I) that estimates the spatiotemporal back of queue propagation in real-time, using loop detector and probe vehicle data, while considering the inherent uncertainties in actuated traffic signal timings. The system uses probe vehicle information as well as information from the control logic to enhance the real-time queue length estimates and does not require historical queue data. Through comprehensive simulation experiments involving a single vehicle approaching a traffic signal, as well as simulations of a signalized intersection covering various market penetration levels (MPLs) and varying demand levels, the fuel savings achieved by the ECO-CACC-I system are quantified. The developed queue estimation algorithm provides a significant improvement over the traditional use of shockwave theory alone. In addition, results of the ECO-CACC-I system demonstrate average fuel savings of up to 10.57% and 18.89%, respectively without and with consideration of the queueing process from the perspective of an individual vehicle. Furthermore, an average fuel saving of 16.5% is achieved for the isolated intersection simulations at a 100% MPL. While significant marginal savings are achieved from a single-vehicle perspective when considering the queue, this is not the case for a network-wide perspective, where the consideration of the queue process only results in a 2% overall enhancement in terms of fuel savings. This research quantifies the benefits and the limitations of ECO-CACC-I when considering surrounding traffic by incorporating real-time queue predictions for deployment in real-world traffic environments.

This paper introduces a robust green light optimal speed advisory (GLOSA) system for fixed and actuated traffic signals considering a probability distribution. These distributions represent the domain of possible switching times from the signal phasing and timing (SPaT) messages. The system finds the least-cost (minimum fuel consumption) vehicle trajectory using a computationally efficient$\text{A}^{\ast} $algorithm incorporated within a dynamic programming (DP) procedure to minimize the vehicle’s total fuel consumption. Constraints are introduced to ensure that vehicles do not collide with other vehicles, run red indications, or exceed a maximum vehicular jerk for passenger comfort. Results of simulation scenarios are evaluated against empirical comparable trajectories of uninformed drivers to compute fuel consumption savings. The proposed approach produced significant fuel savings compared to an uninformed driver behavior amounting to 37% on average for deterministic SPaT and 30% for stochastic SPaT data. A sensitivity analysis was performed to understand how the degree of uncertainty in SPaT predictions affects the optimal trajectory’s fuel consumption. The results present the required levels of confidence in these predictions to achieve savings in fuel consumption. Specifically, the study demonstrates that the proposed system can be within 85% of the maximum savings if the timing error is (±3.3 seconds) at a 95% confidence level. Results also emphasize the importance of more reliable SPaT predictions as the time to green decreases relative to the time the vehicle is expected to reach the intersection given its current speed.

Bike sharing systems (BSSs) are becoming an important part of urban mobility in many cities given that they are sustainable and environmentally friendly. BSS operators spend great efforts to ensure bike and dock availability at each station. Measuring the quality of service (QoS) of each station and/or the entire system is critical for efficient system operations. The traditionally-known QoS measure reported in the literature is based on the proportion of problematic stations, which are defined as those with no bikes or docks available to users. This measure neither exposes the spatial dependencies between stations nor does it discriminate between stations in the BSS. Hence, we propose a novel QoS measure, namely the Optimal Occupancy, in which: 1) the temporal variations in arrival and pick up rates at individual stations are considered; 2) the discriminative property of the Optimal Occupancy is demonstrated using Analysis of Variance (ANOVA) procedures; and 3) geo-statistics, which have not been used before, are applied to explore the spatial Optimal Occupancy variations and model variograms for spatial prediction. This study uses an anonymized bike trip dataset from 34 stations in downtown San Francisco to compare the traditionally-known QoS measure and the proposed Optimal Occupancy measure. Results reveal that the Optimal Occupancy is beneficial, outperforms the traditionally-known QoS measure, and produces a better prediction of the QoS at nearby locations. In addition, the Optimal Occupancy can be used to predict candidate locations for the introduction of new stations in an existing BSS.

The paper presents a novel de-centralized traffic signal controller, achieved using a Nash bargaining game-theoretic framework, that operates a flexible phasing sequence to adapt to dynamic changes in traffic demand levels. The Nash bargaining algorithm is used to optimize the traffic signal timings at each signalized intersection by modeling each phase as a player in a game, where players cooperate to reach a mutually agreeable outcome. The algorithm was implemented in the INTEGRATION microscopic traffic assignment and simulation software and tested on two sample networks. The proposed control approach was compared to the operation of an optimum fixed-time coordinated plan, an actuated controller, a centralized adaptive phase split controller, a decentralized phase split and cycle length controller, and a fully coordinated adaptive phase split, cycle length, and offset optimization controller to evaluate its performance. Testing was initially conducted on an isolated intersection, showing a 77% reduction in queue length, a a 64% reduction in vehicle delay, and a 17% reduction in vehicle emission levels. In addition, the algorithm was tested on an arterial network producing statistically significant reductions in total delay ranging between 36% and 67% and vehicle emission reductions ranging between 6% and 13%.

This paper presents a novel method for estimating the number of vehicles traveling along signalized approaches using probe vehicle data only. The proposed method uses the Kalman Filtering technique to produce reliable vehicle count estimates using real-time probe vehicle estimates of the expected travel times. The proposed method introduces a novel variable estimation interval that allows for higher estimation precision, as the updating time interval always contains a fixed number of probe vehicles. The proposed method is evaluated using empirical and simulated data, the former of which were collected along a signalized roadway in downtown Blacksburg, VA. Results indicate that vehicle count estimates produced by the proposed method are accurate. The paper also examines the model's accuracy when installing a single stationary sensor (e.g., loop detector), producing slight improvements especially when the probe vehicle market penetration rate is low. Finally, the paper investigates the sensitivity of the estimation model to traffic demand levels, showing that the model works better at higher demand levels given that more probe vehicles exist for the same market penetration rate.

Urban traffic congestion is a chronic problem faced by many cities in the US and worldwide. It results in inefficient infrastructure use as well as increased vehicle fuel consumption and emission levels. Congestion is intertwined with delay, as road users waste precious hours on the road, which in turn reduces productivity. Researchers have developed, and continue to develop, tools and systems to alleviate this problem. Network perimeter control is one such tool that has been studied extensively. It attempts to control the flow of vehicles entering a protected area to ensure that the congested regime predetermined by the Network Fundamental Diagram (NFD) is not reached. In this paper, an approach derived from sliding mode control theory is presented. Its main advantages over proportional-integral controllers include (1) minimal tuning, (2) no linearization of the governing equations, (3) no assumptions with regard to the shape of the NFD, and (4) ability to handle various demand profiles without the need to retune the controller. A sliding mode controller was implemented and tested on a congested grid network. The results show that the proposed controller produces network-wide delay savings and disperses congestion effectively.

Consecutive traffic signalized intersections can increase vehicle stops, producing vehicle accelerations on arterial roads and potentially increasing vehicle fuel consumption levels. Eco-driving systems are one method to improve vehicle energy efficiency with the help of vehicle connectivity. In this paper, an eco-driving system is developed that computes a fuel-optimized vehicle trajectory while traversing more than one signalized intersection. The system is designed in a modular and scalable fashion allowing it to be implemented in large networks without significantly increasing the computational complexity. The proposed system utilizes signal phasing and timing (SPaT) data that are communicated to connected vehicles (CVs) together with real-time vehicle dynamics to compute fuel-optimum trajectories. The proposed algorithm is incorporated in the INTEGRATION microscopic traffic assignment and simulation software to conduct a comprehensive sensitivity analysis of various variables, including: system market penetration rates (MPRs), demand levels, phase splits, offsets and traffic signal spacings on the system performance. The analysis shows that at 100% MPR, fuel consumption can be reduced by as high as 13.8%. Moreover, higher MPRs and shorter phase lengths result in larger fuel savings. Optimum demand levels and traffic signal spacings exist that maximize the effectiveness of the algorithm. Furthermore, the study demonstrates that the algorithm works less effective when the traffic signal offset is closer to its optimal value. Finally, the study highlights the need for further work to enhance the algorithm to deal with over-saturated traffic conditions.

This paper proposes a novel supervised clustering algorithm to analyze large datasets. The proposed clustering algorithm models the problem as a matching problem between two disjoint sets of agents, namely, centroids and data points. This novel view of the clustering problem allows the proposed algorithm to be multi-objective, where each agent may have its own objective function. The proposed algorithm is used to maximize the purity and similarity in each cluster simultaneously. Our algorithm shows promising performance when tested using two different transportation datasets. The first dataset includes speed measurements along a section of Interstate 64 in the state of Virginia, while the second dataset includes the bike station status of a bike sharing system (BSS) in the San Francisco Bay Area. We clustered each dataset separately to examine how traffic and bike patterns change within clusters and then determined when and where the system would be congested or imbalanced, respectively. Using a spatial analysis of these congestion states or imbalance points, we propose potential solutions for decision makers and agencies to improve the operations of I-64 and the BSS. We demonstrate that the proposed algorithm produces better results than classical K -means clustering algorithms when applied to our datasets with respect to a time event. The contributions of our paper are: 1) we developed a multi-objective clustering algorithm; 2) the algorithm is scalable (polynomial order), fast, and simple; and 3) the algorithm simultaneously identifies a stable number of clusters and clusters the data.

Our research aims at developing intelligent systems to reduce the transportation-related energy expenditure of a large city by influencing individual behavior. We introduce Copter - an intelligent travel assistant that evaluates multi-modal travel alternatives to find a plan that is acceptable to a person given their context and preferences. We propose a formulation for acceptable planning that brings together ideas from AI, machine learning, and economics. This formulation has been incorporated in Copter producing acceptable plans in real-time. We adopt a novel empirical evaluation framework that combines human decision data with high-fidelity simulation to demonstrate a 4% energy reduction and 20% delay reduction in a realistic deployment scenario in Los Angeles, California, USA.

Non-recurrent events (e.g. football events or evacuation in case of natural disasters) pose great challenges to vehicle routing and traffic management in Intelligent Transportation Systems (ITSs). The high traffic demand during such events, combined with the road network resource constraints, bring forward the need for efficient and smart management techniques that better utilize the network facilities while maintaining a certain performance level. Vehicular Ad-hoc Networks (VANETs) and the advancement in information technology bring new opportunities to address such problems. This paper utilizes VANETs to build a vehicular crowd management system that utilizes system-optimum stochastic routing. The objective is to clear the network in a shorter time by better utilizing the available network resources. To build this system, vehicles are used as sensors that communicate the network state information to the Traffic Management Center (TMC) in real time. A linear programming model is developed to minimize the network-wide travel time constrained by the road link capacities based on the collected information. The results show that the proposed system decreases the network-wide travel time and is successful in clearing the network earlier by up to 38% compared to deterministic user-equilibrium traffic assignment.

Our research aims at developing intelligent systems to reduce the transportation-related energy expenditure of a large city by influencing individual behavior. We introduce Copter - an intelligent travel assistant that evaluates multi-modal travel alternatives to find a plan that is acceptable to a person given their context and preferences. We propose a formulation for acceptable planning that brings together ideas from AI, machine learning, and economics. This formulation has been incorporated in Copter that produces acceptable plans in real-time. We adopt a novel empirical evaluation framework that combines human decision data with a high fidelity multi-modal transportation simulation to demonstrate a 4% energy reduction and 20% delay reduction in a realistic deployment scenario in Los Angeles, California, USA. This article is part of the special track on AI and Society.

This paper develops a novel two-layer hierarchical classifier that increases the accuracy of traditional transportation mode classification algorithms. This paper also enhances classification accuracy by extracting new frequency domain features. Many researchers have obtained these features from global positioning system data; however, this data was excluded in this paper, as the system use might deplete the smartphone's battery and signals may be lost in some areas. Our proposed two-layer framework differs from previous classification attempts in three distinct ways: 1) the outputs of the two layers are combined using Bayes' rule to choose the transportation mode with the largest posterior probability; 2) the proposed framework combines the new extracted features with traditionally used time domain features to create a pool of features; and 3) a different subset of extracted features is used in each layer based on the classified modes. Several machine learning techniques were used, including k-nearest neighbor, classification and regression tree, support vector machine, random forest, and a heterogeneous framework of random forest and support vector machine. Results show that the classification accuracy of the proposed framework outperforms traditional approaches. Transforming the time domain features to the frequency domain also adds new features in a new space and provides more control on the loss of information. Consequently, combining the time domain and the frequency domain features in a large pool and then choosing the best subset results in higher accuracy than using either domain alone. The proposed two-layer classifier obtained a maximum classification accuracy of 97.02%.

In recent years, automated vehicles (AVs) are emerging as a realistic and viable option. Consequently, substantial research efforts are being dedicated to the development of systems and technologies for the field implementation of AVs. One of the challenges researchers need to address is how to optimize the movement of AVs through roadway intersections. In this paper, an attempt to address this need is introduced. The developed model is an optimization problem subjected to dynamical constraints (i.e., ordinary differential equations governing the motion of a vehicle) and static constraints (i.e., maximum achievable velocities). By virtue of Pontryagin's minimum principle, the solution that minimizes the trip time is obtained. Given the problem parameters, the solution to the formulated problem is expected to be the true optimum and delivers the lowest possible delay that satisfies the previously mentioned constraints. This logic is simulated and compared with the operation of a roundabout, a stop sign, and a traffic signal-controlled intersection. The results demonstrate that an 80% reduction in delay is achievable compared with the best of these three intersection control strategies, on average. An interesting byproduct of this new logic is a reduction in vehicular fuel consumption and CO2emissions by 42.5% and 40%, respectively.

Real-time, accurate travel time prediction algorithms are needed for individual travelers, business sectors, and government agencies. They help commuters make better travel decisions, avert traffic congestion, help the environment by reducing carbon emissions, and improve traffic efficiency. Travel time prediction has begun to attract more attention with the rapid development of intelligent transportation systems (ITSs), and is considered one of the more important elements required for successful ITS subsystems deployment. However, the stochastic nature of travel time makes accurate prediction a difficult task. This paper proposes travel time modeling using a mixture of linear regressions. The proposed model consists of two normal components. The first component models the congested regime while the other models the free-flow regime. The means of the two components are modeled by two linear regression equations. The predictors used in the linear regression equation are selected out of the spatiotemporal speed matrix using a random forest machine-learning algorithm. The proposed model is tested using archived data from a 74.4-mile freeway stretch of I-66 eastbound connecting I-81 and Washington, D.C. The experimental results show the ability of the model to capture the stochastic nature of travel time and to predict travel time accurately.

Traffic signals typically introduce vehicle stops along a trip and thus increase vehicle fuel consumption levels. In attempt to enhance vehicle fuel efficiency, eco-cooperative adaptive cruise control (Eco-CACC) systems are being developed that receive signal phasing and timing data from downstream signalized intersections via vehicle-to-infrastructure communication. In this paper, an Eco-CACC algorithm is developed that computes the fuel-optimum vehicle trajectory through a signalized intersection by ensuring that the vehicle arrives at the intersection stop bar just as the last queued vehicle is discharged. A simulation analysis demonstrates that the proposed Eco-CACC system produces vehicle fuel savings up to 40%, when the market penetration rate (MPR) is 100%. The results also demonstrate that for single-lane approaches, the algorithm reduces the overall fuel consumption levels for all vehicles, and that higher MPRs result in larger savings. Alternatively, on multi-lane approaches, lower MPRs produce negative impacts on the overall intersection fuel efficiency, and only when MPRs are greater than 30% does the algorithm produce overall intersection fuel consumption savings.

Global warming, environmental pollution, and fuel shortage are currently major worldwide challenges. Ecorouting is one of several tools that attempt to address this challenge by minimizing network-wide vehicle fuel consumption and emission levels. Eco-routing systems select the most environmentally friendly route. The subpopulation feedback eco-routing (SPF-ECO) algorithm that is implemented in the INTEGRATION software can produce a reduction in fuel consumption levels by approximately 17%. However, in some cases, due to delayed updates or the lack for updates, its performance degrades. In this paper, we propose the ant colony based eco-routing technique (ACO-ECO), which is a novel feedback eco-routing and cost updating algorithm to overcome these shortcomings. In the ACO-ECO algorithm, real-time performance measures on various roadway links are shared. Vehicles build their minimum path routes using the latest real-time information to minimize their fuel consumption and emission levels. ACO-ECO is also able to capture randomness in route selection, pheromone updating, and pheromone evaporation. The results show that the ACO-ECO algorithm and SPF-ECO have similar performances in normal cases. However, in the case of link blocking, the ACO-ECO algorithm reduces the network-wide fuel consumption and CO2 emission levels in the range of 2.3% to 6.0%. It also reduces the average trip time by approximately 3.6% to 14.0%.

Real-time automatic congestion identification is one of the important routines of intelligent transportation systems (ITS). Previous efforts usually use traffic state measurements (speed, flow, occupancy) to develop congestion identification algorithms. However, the impacts of weather conditions to identify congestion have not been investigated in the existing studies. In this paper, we proposed an algorithm that uses the speed probe data and the corresponding weather and visibility to build a transferable model. This model can be used on any road stretch. Our algorithm assumes traffic states can be classified into three regimes: congestion, speed at capacity and free-flow. Moreover, the speed distribution follows a mixture of three components whose means are functions in weather and visibility. The mean of each component is defined using a linear regression using different weather conditions and visibility levels as predictors. We used three data sets from VA, CA and TX to estimate the model parameters. The fitted model is used to calculate the speed cut-off between congestion and speed at capacity which minimize either the Bayesian classification error or the false positive (congestion) rate. The test results demonstrate the proposed method produces promising congestion identification output by considering weather condition and visibility.

The paper addresses the problem of stretch wide short-term prediction of traffic stream state. The problem is a multivariate problem where the responses are the speeds or flows on different road segments at different time horizons. Recognizing that short-term traffic state prediction is a multivariate problem, there is a need to maintain the spatiotemporal traffic state correlations. Two cutting-edge machine learning algorithms are used to predict the stretch-wide traffic stream traffic state up to 120 minutes in the future. Furthermore, the divide and conquer approach was used to divide the large prediction problem into a set of smaller overlapping problems. These smaller problems are solved using a medium configuration PC in a reasonable time (less than a minute), which makes the proposed technique suitable for practical applications.

In this paper, we present a new carpooling recommendation system whose main objective is to find the best carpool matchings and recommend individuals to join their friends during trips or travels. The proposed recommendation system utilizes user's mobility history and user social network information to find carpool matchings. The proposed system employs a probabilistic model based on continuous time Markov chain to model user's mobility and to predict user future movements. Moreover, it uses two similarity measures (interest based similarity and friendship based similarity) to find the similarities between users. The interest based similarity uses a weighted bipartite graph between users and places, where the edges are weighted by the term frequency-inverse document frequency. The friendship based similarity uses the common friends as a similarity measure. The proposed system is evaluated using the number of carpool matchings that can be found, this number gives an indication of the reduction that can be made in vehicular traffic congestion, pollutant emissions and energy consumption.

Vehicular ad hoc networks (VANETs) are a special case of mobile ad hoc networks (MANETs). The distinctive characteristics of VANETs include high-speed vehicular nodes and significant variability in node density. Collecting data from VANETs is important in monitoring, controlling, and managing road traffic. However, efficient collection of the needed data is challenging because vehicles are continuously moving and generating a significant amount of events and data. The focus of this paper is on vehicle data collection using 3G/LTE. Initially, a comparison of proactive and reactive data collection schemes is conducted using simulation. The results show that proactive schemes produce the lowest delay and bandwidth usage but the highest loss ratio. To efficiently use the available bandwidth, an adaptive data collection scheme is developed and described. This adaptive data collection scheme is based on a proactive scheme using variable polling periods depending on the vehicle positions in the network and travel time to provide accurate traffic and travel time information to the Traffic Management Center (TMC). Simulation results, using taxi traces in Qatar, show that the proposed algorithm consumes an acceptable amount of megabytes (≈31 MB) per month when the basic polling interval is set to 10 s. Furthermore, traffic simulation results demonstrate that the proposed scheme has a minimum impact on delay and travel time estimates (relative error less than 2.5%), but can produce significant degradations in fuel consumption and emission estimates if computations are made at the TMC (relative error greater than 28% and 65%, respectively). These errors can be eliminated if computations are done on the vehicle.

The paper introduces a Vehicular Network Integrated Simulator (VNetIntSim) that integrates transportation modelling with Vehicular Ad Hoc Network (VANET) modelling. Specifically, VNetIntSim integrates the OPNET software, a communication network simulator, and the INTEGRATION software, a microscopic traffic simulation software. The INTEGRATION software simulates the movement of travellers and vehicles, while the OPNET software models the data exchange through the communication system. Information is exchanged between the two simulators as needed. As a proof of concept, the VNetIntSim is used to quantify the impact of mobility parameters (traffic stream speed and density) on the communication system performance, and more specifically on the data routing (packet drops and route discovery time).

This paper adopts different supervised learning methods from the field of machine learning to develop multiclass classifiers that identify the transportation mode, including driving a car, riding a bicycle, riding a bus, walking, and running. Methods that were considered include K-nearest neighbor, support vector machines (SVMs), and tree-based models that comprise a single decision tree, bagging, and random forest (RF) methods. For training and validating purposes, data were obtained from smartphone sensors, including accelerometer, gyroscope, and rotation vector sensors. K-fold cross-validation as well as out-of-bag error was used for model selection and validation purposes. Several features were created from which a subset was identified through the minimum redundancy maximum relevance method. Data obtained from the smartphone sensors were found to provide important information to distinguish between transportation modes. The performance of different methods was evaluated and compared. The RF and SVM methods were found to produce the best performance. Furthermore, an effort was made to develop a new additional feature that entails creating a combination of other features by adopting a simulated annealing algorithm and a random forest method.

This research develops a simple vehicle powertrain model that can be incorporated within microscopic traffic simulation software for the modeling of intelligent vehicle applications. This simple model can be calibrated using vehicle parameters that are publically available without the need for field data collection. The model uses the driver throttle level input to compute the engine speed and, subsequently, the engine torque and power to finally compute the vehicle acceleration, speed, and position. The model is tested using field measurements and is demonstrated to produce vehicle power, fuel consumption, acceleration, speed, and position estimates that are consistent with field observations.

Within the context of transportation modeling, driver route choice is typically captured using mathematical programming approaches, which assume that drivers, in attempting to minimize some objective function, have full knowledge of the transportation network state. Typically, drivers are assumed to either minimize their travel time (user equilibrium) or minimize the total system travel time (system optimum). Given the dynamic and stochastic nature of the transportation system, the assumption of a driver's perfect knowledge is at best questionable. While it is well documented in psychological sciences that humans tend to minimize their cognitive efforts and follow simple heuristics to reach their decisions, especially under uncertainty and time constraints, current models assume that drivers have perfect or close to perfect knowledge of their choice set, as well as the travel characteristics associated with each of the choice elements. Only a few of the many route choice models that are described in the literature are based on observed human behavior. With this in mind the research presented in this paper monitors and analyzes actual human route choice behavior. It compares actual drivers experiences, perceptions and choices, and demonstrates that (a) drivers perceptions are significantly different from their actual experiences, and that drivers' choices are better explained by their perceptions than their experiences; (b) drivers perceive travel speeds better than travel times (c) perceived travel speeds seem to influence route choice more than perceived travel times, and (d) drivers' route choice behavior differs across different driver groups.

This special issue contains revised versions of selected papers originally presented at the 9th IEEE International Conference on Intelligent Transportation Systems (ITSC 2006) held in Toronto, Canada, on September 17-20, 2006.

This paper involves a field test on 60 test participants to characterize driver behavior (perception-reaction time (PRT) and stopping/running decisions) at the onset of a yellow phase. Driver behavior is analyzed for five trigger distances that are measured from the vehicle position at the start of the yellow indication to the stop bar. This paper demonstrates that the 1.0-s 85th-percentile PRT that is recommended in traffic-signal-design procedures is valid and consistent with the field observations. Furthermore, this paper clearly shows that brake PRTs are impacted by the vehicle's time to intersection (TTI) at the onset of a yellow-indication introduction. This paper also demonstrates that either a lognormal or a beta distribution is sufficient to model the stochastic nature of the brake PRT. In terms of stopping decisions, this paper demonstrates that the probability of stopping varies from 100% at a TTI of 5.5 s to 9% at a TTI of 1.6 s. This paper also indicates a decrease in the probability of stopping for male drivers when compared with female drivers. Furthermore, this study suggests that drivers 65 years of age and older are significantly less likely to clear the intersection at short yellow-indication trigger distances when compared with other age groups. The dilemma zone for the less than 40 year old group is found to range from 3.9 to 1.85 s, whereas the dilemma zone for the greater than 70 year old group is found to range from 3.2 to 1.5 s.