Natural or human-made disasters often force individuals to evacuate to safer places. Depending on the nature of the disaster, evacuations may require people to travel short distances or long distances, and may affect a city, multiple states, or at times an entire nation. From a transportation perspective, resource constraints such as limited roadway capacity and vehicle availability restrict the ability to plan a safe and efficient evacuation. The chapter discusses the difficulties that surround a large-scale evacuation and presents a summary of studies that use mathematical models to address these problems. It highlights some of the challenges in post-disaster management. The chapter describes modeling approaches to forecast the decisions, and how simulation tools built on the principles can support advance planning for evacuations. Vehicle availability also plays a major role in evacuation, even in regions with a high per-capita automobile ownership rate, since households without automobiles often have fewer resources in general.

This paper formulates the reliable routing of electric vehicles in stochastic networks as a multicriteria shortest path problem with travel time and charging cost components. The reliability term is defined as the probability of finishing the trip without running out of charge. The arc travel times are represented as stochastic variables, and arc energy consumption is modeled as a linear function of arc length and arc travel time. The traveler aims to minimize the generalized cost, formulated as a linear function of travel time and charging cost, subject to a minimum reliability threshold, representing the level of risk a traveler is willing to take in favor of routes with lower cost. We propose a solution algorithm based on generalized dynamic programming and show that the optimal solution may include cycles that visit at least one charging station. The properties of the proposed multicriteria shortest path problem are mathematically proved. The simulation results on randomly-generated networks show that cyclic paths are very rare, and that the generalized cost of travel is a monotone increasing function of minimum reliability threshold.

This paper describes a spatial parallelization scheme for the static traffic assignment problem. In this scheme, which we term a decomposition approach to the static traffic assignment problem (DSTAP), the network is divided into smaller networks, and the algorithm alternates between equilibrating these networks as subproblems, and master iterations using a simplified version of the full network. The simplified network used for the master iterations is based on linearizations to the equilibrium solution for each subnetwork obtained using sensitivity analysis techniques. We prove that the DSTAP method converges to the equilibrium solution on the full network, and demonstrate computational savings of 35-70% on the Austin network. Natural applications of this method are statewide or national assignment problems, or cities with rivers or other geographic features where subnetworks can be easily defined. (C) 2017 Elsevier Ltd. All rights reserved.

This paper formulates the problem of online charging and routing of a single electric vehicle in a network with stochastic and time-varying travel times. Public charging stations, with nonidentical electricity prices and charging rates, exist through the network. Upon arrival at each node, the traveler learns the travel time on all downstream arcs and the waiting time at the charging station, if one is available. The traveler aims to minimize the expected generalized cost-formulated as a weighted sum of travel time and charging cost-by considering the current state of the vehicle and availability of information in the future. The paper also discusses an offline algorithm by which all routing and charging decisions are made a priori. The numerical results demonstrate that cost savings of the online policy, compared with that for the offline algorithm, is more significant in larger networks and that the number of charging stations and vehicle efficiency rate have a significant impact on those savings.

Calculating equilibrium sensitivity on a bush can be done very efficiently, and serve as the basis for a network contraction procedure. The contracted network (a simplified network with a few nodes and links) approximates the behavior of the full network but with less complexity. The network contraction method can be advantageous in network design applications where many equilibrium problems must be solved for different design scenarios. The network contraction procedure can also be used to increase the accuracy of subnetwork analysis. This method requires calculating travel time derivatives between two nodes, with respect to the demand between them, assuming that the flow distributes in a way that equilibrium is maintained. Previous research describes two methods for calculating these derivatives. This paper presents a third method, which is simpler, faster, and just as accurate. The method presented in this paper reformulates the linear system of equations defining these sensitivities as the solution to a convex programming problem, which can be solved by making minor modifications to static user equilibrium algorithms. In addition, the model is extended to capture the interactions between the path travel times and network flows, and a heuristic is proposed to compute these interactions. The accuracy and complexity of the proposed methodology are evaluated using the network of Barcelona, Spain. Further, numerical experiments on the Austin, Texas regional network validate its performance for subnetwork analysis applications. (C) 2015 Elsevier Ltd. All rights reserved.

Accessibility is an important factor in determining public transit use. Accessibility models have been studied in the literature for planning purposes for a long time. However, as transportation models are being improved in spatial and temporal resolution, the effect of accessibility on transit assignment needs to be thoroughly examined. In this study, the authors investigated possible improvements for modeling transit access in an assignment model. Two models are used to determine the transit accessible area in each traffic zone and to estimate the equivalent walking distance from each zone to each transit stop. The models, taking advantage of the network distance and distance decay factor, are alternatives to replace the traditional buffer or centroid connector methods in assignment models. The accessibility models are then integrated with a schedule-based transit assignment model to determine the impact on route- and network- level transit usage. The integrated model is compared with the current state of the model, and the results are evaluated by comparing with observed data from a transit on-board survey. The evaluation shows significant improvement in the effectiveness of the assignment model in estimating route ridership, after incorporating the more accurate accessibility model.

As fuel prices increase, drivers may make travel choices to minimize not only travel time, but also fuel consumption. Consideration of fuel consumption would affect route choice and influence trip frequency and mode choice. For instance, travelers may elect to live closer to their workplace, or use public transit to avoid fuel consumption and the associated costs. To incorporate network characteristics into predictions of the effects of fuel prices, we develop a multi-class combined elastic demand, mode choice, and user equilibrium model using a generalized cost function of travel time and fuel consumption with a combined solution algorithm. The algorithm is implemented in a custom software package, and a case study application on the Austin, Texas network is presented. We evaluate the fuel-price sensitivity of key variables such as drive-alone and transit class proportions, person-miles traveled, link-level traffic flow and per capita fuel consumption and emissions. These effects are examined across a heterogeneous demand set, with multiple user-classes categorized based on their value of travel time. The highest relative transit elasticities against fuel price are observed among low value of time classes, as expected. Although total personal vehicle travel decreases, congestion increases on some roads due to the generalized cost function. Reductions in system-wide fuel consumption and greenhouse gas emissions are observed as well. The study uncovers the combined interactions among fuel prices, multi-modal choice behavior, travel performance, and resultant environmental impacts, all of which dictate the urban travel market. It also equips agencies with motivation to tailor emissions reduction and transit-ridership stimulus policies around the most responsive user classes.

Understanding evacuation practices and outcomes helps crisis and disaster personnel plan, manage, and rebuild during disasters. Yet the recent expansion in the number of information and communication technologies (ICTs) available to individuals and organizations has changed the speed and reach of evacuation-related messages. This study explores ICTs' influences on evacuation decision-making and traffic congestion. Drawing from both social science and transportation science, we develop a model representative of individual decision making outcomes based on the amount of ICT use, evacuation sources, and the degree of evacuation urgency. We compare the evacuation responses when individuals receive both advance notice of evacuation (ANE) and urgent evacuation (UE) messages under conditions of no ICTs and prolific ICT use. Our findings from the scenarios when there is widespread ICT use reveal a shift in the evacuation time-scale, resulting in traffic congestion early in the evacuation cycle. The effects of this congestion in urgent situations are significantly worse than traffic congestion in the advance notice condition. Even under conditions where face-to-face communication is the only option, evacuations still occur, but at a slower rate, and there are virtually no traffic congestion issues. Our discussion elaborates on the theoretical contributions and focuses on how ICTs have changed evacuation behavior. Future research is needed to explore how to compensate for the rush to the road.

Advanced traffic assignment models, such as simulation-based dynamic traffic assignment, typically incorporate more detailed network representations than do traditional planning models. In this context, the placement of centroid connectors may have a significant effect on model performance, and attention must be paid to their number and location to avoid unrealistic congestion or low utilization of minor roadways by local traffic. Given that the manual inspection of centroid connector placement may be too time-consuming in large regional networks, this paper proposes two simple automatic centroid connector placement strategies for dynamic traffic assignment applications. The first approach radially distributes the connectors to the nearest nodes and is intended to exemplify some limitations of the most common techniques in practice. The second strategy involves dividing the centroid and subsequent demand into two parts, distributing the demand across one sub-centroid linked to nearby nodes and one linked to the periphery, and thus effectively establishing a bilevel distribution. A modification of this strategy involves eliminating nodes at signalized intersections as viable candidates for connection. As part of the evaluation of the methods, a new metric, the locality factor, has been introduced to describe the use of minor streets by local traffic. The numerical experiments, conducted on two real-world networks, exemplify the effects of the incorporation of local streets and the placement of centroid connectors on model results. Sensitivity testing and limited field data comparisons suggest that the bilevel centroid connector placement strategy achieves more realistic results.

Traffic surveillance using the information embedded in aerial imagery has been widely used recently and has many advantages over current ground based data collection approaches which have limited performance in the case of disasters or crucial circumstances. Accurate detection and tracking of moving vehicles is of interest to many researchers, and the keypoint for these tasks is image registration. Image registration geometrically aligns different images taken from the same scene at different times or by different sensors. Two main tasks are performed and described in this paper: georeferencing and local registration with car detection. Georeferencing focuses on global registration of the helicopter frame (target image) to a satellite image (reference image) to estimate latitude and longitude of each pixel according to a base image. Due to complexities of this work, the georeferencing was broken into two steps. The first step finds a small area from the satellite image covering the helicopter image. The target image is registered to this extracted sub-image at the second step. The next task is local registration and car detection. In this case the target and reference images are both from helicopter imagery which makes it simpler than georeferencing. Then, these registered images are subtracted to form the difference image which can be used for car detection. However, the presence of noise encouraged us to do registration in a smaller scale. The images were partitioned to smaller sub-images and a transformation model was designed for each sub-image. Finally these small registered areas were matched to build the target image registered to the reference image. The resultant difference image has lower noise.

A logit transit route choice model has been estimated using on-board transit survey data. The model is supposed to capture more detailed path attributes and user characteristics such as value of time and trip purpose. The model is suitable for application in a recently-developed schedule-based transit assignment model, compatible with the advanced transportation models. Statement of Financial Interest There would be no financial gain for any of the authors from publishing this brief. Modeling results are analyzed from a methodological standpoint, and are not directly related to the characteristics of the selected software tools. Statement of Innovation Modeling transit user behavior in a dynamic (schedule-based) network can be a useful tool for planning and decision making. Unlike the current state-of-the-practice transit network models, the proposed transit assignment model provides more detailed information about supply-demand interaction. However, the requirement is that the user behavior is properly studied and used in the assignment. There might be many parameters affecting the transit users’ decision making process, leading them to use the paths other than the “shortest path”. In this study, we aimed to investigate these parameters, and incorporate them in the dynamic assignment model, so to turn a state-of-the-art model into a practice-ready model. Moreover, the application of the transit data, collected by transit agencies, is investigated for calibration and validation of the assignment model. In general, the model is compatible with the activity-based demand models and dynamic traffic assignment models. Therefore, it can contribute in the next generation of the travel demand models.

This paper proposes a methodology to model the network-level impact of eco-routing policies using a Dynamic Traffic Assignment (DTA) platform. The DTA model is used to estimate recurrent traffic conditions based on Dynamic User Equilibrium (DUE) principles, and provides the inputs required to find feasible eco-routes for all origin-destination (OD) pairs in the network. A number of parametric tests are conducted on two networks (Austin, TX, and Nicosia, Cyprus) assuming that different fractions of drivers are re-routed into the corresponding eco-path. The simulation engine in the DTA package is used to assess the impact of the proposed policies and to compute fuel consumption. Future modeling efforts can easily incorporate CO2 emissions and other air-quality measures. Eco-routing schemes have recently been considered (and applied) due to advancements in In-vehicle navigation systems and real-time estimates of fuel consumption and greenhouse gas emissions. However, the network-level impacts of such strategies are not simple to predict. The use of DTA in the context of eco-routing allows studying very large areas which could not be modeled using other approaches. It also provides the means to compute a number of interesting performance metrics, including changes in distance traveled, average speed, and system travel time. These may lead to a more comprehensive understanding of the considered problem. Further, our modeling efforts suggest that in order to achieve system-level energy savings, fairly complex eco-routing strategies may be needed, particularly when long-term driver’s behavior is considered. DTA models can be a powerful tool to develop and test such approaches.

In this paper a comparison for non-adaptive and adaptive traffic state estimators based on the nonlinear macroscopic traffic flow model is presented. In non-adaptive estimator, a Least Square based method was used to estimate model parameters with off-line data. Beside this non-adaptive estimator, three adaptive estimators were designed and tested. In these adaptive estimators, online estimation of the model parameters were done based on Least Square, joint filtering and dual filtering methods. In all mentioned estimators, extended Kalman filtering method was used to estimate traffic variables. Finally, Real data testing results of these estimators for Interstate 494 in metro freeway, Minnesota, USA are presented.

To develop mathematical models and estimate their parameters are an essential issue for studying dynamic behaviors of traffic systems. METANET model is one of the most applicable models in traffic modeling which parameters have plenty of effects on model behavior. In this paper, we describe the effects of the model parameters on the model behavior and the estimation quality of system states in the case of undetermined parameters. The preliminary results show that EKF can accurately estimate the parameters and predict states in nonlinear state-space equations for modeling dynamic traffic networks for preparing proper signal in traffic control.

This paper presents real-data testing results of a real-time freeway traffic state estimator. The general approach to real-time adaptive of freeway traffic state estimation is based on nonlinear macroscopic traffic flow modeling and extended Kalman fliter algorithm. Macroscopic traffic flow model contains three important and unknown parameters (free speed, critical density and exponenet), which should be estimated with off-line or on-line methods. One innovative aspects of the estimator is the real-time joint estimation of traffic flow variables (traffic flow, mean speed and traffic density) and model parameters, that leads to some significant features such as: avoidance of prior model calibration, automatic adaption to changing external conditions (e.g. weather conditions, traffic composition,...). The purpose of the reported real-data testing is, first, to demonstrate some obstacles in previous methods, second, to propose two methods based on dual filtering and fosion of the model parameter estimates for improving the previous methods.