This article presents a Takagi–Sugeno–Kang Fuzzy Neural Network (TSKFNN) approach to predict freeway corridor travel time with an online computing algorithm. TSKFNN, a combination of a Takagi–Sugeno–Kang (TSK) type fuzzy logic system and a neural network, produces strong prediction performance because of its high accuracy and quick convergence. Real world data collected from US-290 in Houston, Texas are used to train and validate the network. The prediction performance of the TSKFNN is investigated with different combinations of traffic count, occupancy, and speed as input options. The comparison between online TSKFNN, offline TSKFNN, the back propagation neural network (BPNN) and the time series model (ARIMA) is made to evaluate the performance of TSKFNN. The results show that using count, speed, and occupancy together as input produces the best TSKFNN predictions. The online TSKFNN outperforms other commonly used models and is a promising tool for reliable travel time prediction on a freeway corridor.

Activity-based travel scenario analysis and network design using a household activity pattern problem (HAPP) can face significant computational cost and inefficiency. One solution approach, called reoptimization, makes use of an optimal solution of a prior problem instance to find a new solution faster and more accurately. Although the method is generally NP-hard as well, the approximation bound has been shown in the literature to be tighter than a full optimization for several traveling salesman problem variations. To date, however, there have not been any computational studies conducted with the method for scenario analysis with generalized vehicle routing problems, nor has there been any metaheuristics designed with reoptimization in mind. A generalized, selective household activity routing problem (G-SHARP) is presented as an extension of the HAPP model to include both destination and schedule choice for the purpose of testing reoptimization. Two reoptimization algorithms are proposed: a simple swap heuristic and a new class of evolutionary algorithms designed for reoptimization, dubbed a Genetic Algorithm with Mitochondrial Eve (GAME). The two algorithms are tested against a standard genetic algorithm in a computational experiment involving 100 zones that include 400 potential activities (resulting in a total of 802 nodes per single-traveler household). Five hundred households are synthesized and computationally tested with a base scenario, a scenario where an office land use in one zone is dezoned, and a scenario where a freeway is added onto the physical network. The results demonstrate the effectiveness of reoptimization heuristics, particularly GAME, and the capability of G-SHARP to capture reallocations of activities and schedules with respect to spatiotemporal changes.

Users of road facilities generally express a judgment on road quality based on their psychophysical conditions, in relation to the environment they refer to. This judgment is made considering many aspects, for example, the presence of traffic lights, the frequency of interchanges, of lay-bys and gas stations, route conformation, environmental conditions, quality of road signs, etc. In this article, we propose a new index called Global Satisfaction Index, which uses vehicular traffic quality and quality of road pavement, to summarize these aspects, and to express the users’ global judgment about the ride comfort on rural roads. Since in this kind of judgment a subjective perception process is involved, we have used fuzzy theory to handle uncertainty embedded in the process. The attributes of the aspects considered have been expressed through fuzzy numbers, and the global judgment has been obtained through a fuzzy inference system. In this way the proposed index overcomes the limits of other existing indices, since it incorporates uncertainties and/or imprecision inherent in the drivers’ perception of the ride comfort. Moreover, it can be used for evaluation and comparison of different types of road sections. Finally, a numerical example is presented to assist in understanding the practical aspects of the proposed index.

In a passenger railroad system, the stopping pattern optimization problem determines the train stopping strategy, taking into consideration multiple train classes, station types, and customer origin-destination (OD) demand, to maximize the profit made by a rail company. The stopping pattern is traditionally decided by rule of thumb, an approach that leaves much room for improvement. In this article, we propose an integer program for this problem and provide a systematic approach to determining the optimal train stopping pattern for a rail company. Commonly used commercial optimization packages cannot solve this complex problem efficiently, especially when problems of realistic size need to be solved. Therefore, we develop two genetic algorithms, namely binary-coded genetic algorithm (BGA) and integer-coded genetic algorithm (IGA). In many of the past evolutionary programming studies, the chromosome was coded using the binary alphabet as BGA. The encoding and genetic operators of BGA are straightforward and relatively easy to implement. However, we show that it is difficult for the BGA to converge to feasible solutions for the stopping pattern optimization problem due to the complex solution space. Therefore, we propose an IGA with new encoding mechanism and genetic operators. Numerical results show that the proposed IGA can solve real-world problems that are beyond the reach of commonly used optimization packages.

This article presents a stochastic computational framework to investigate the chloride-induced corrosion of reinforced concrete superstructures. For this purpose, three-dimensional finite-element models are developed to determine the extent of chloride penetration into the superstructure components. One of the unique capabilities of this framework is to simultaneously consider all the major factors that affect the corrosion process. Furthermore, the developed framework integrates various sources of uncertainty into the performance predictions. This will be achieved by modifying the element properties and boundary conditions of the finite-element models at each time step. For a reliable durability assessment of deteriorating structural components, the proposed framework incorporates the temporal and spatial uncertainties of the influential parameters into the finite-element analysis. Considering that most of the parameters involved in the corrosion process follow non-normal distributions, a series of non-Gaussian stochastic fields are generated following a computationally efficient procedure. The results calculated from extensive stochastic simulations are expressed in terms of the likelihood and extent of corrosion initiation. The outcome of this study indicates that the computational investigation of the corrosion process can be significantly improved if a reliable probabilistic framework is employed instead of conventional deterministic approaches.

Clarifying traffic flow phases is a primary requisite for applying length-based vehicle classifications with dual-loop data under various traffic conditions. One challenge lies in identifying traffic phases using variables that could be directly calculated from the dual-loop data. This article presents an innovative approach and associated algorithm for identifying traffic phases through a hybrid method that incorporates level of service method and K-means clustering method. The “phase representative variables” are identified to represent traffic characteristics in the traffic flow phase identification algorithm. The traffic factors influencing the vehicle classification accuracy under non-free traffic conditions are successfully identified using video-based vehicular trajectory data, and the innovative length-based vehicle classification models are then developed. The result of the concept-of-evidence test with use of sample data indicates that compared with the existing model, the accuracy of the estimated vehicle lengths is increased from 42% to 92% under synchronized and stop-and-go conditions. The results also foster a better understanding of the traffic stream characteristics and associated theories to lay out a good foundation for further development of relevant microscopic simulation models with other sensing traffic data sources.

To progress the development of deployable, lightweight infrastructure for relief and recovery efforts in the aftermath of natural and man-made disasters, this article develops a proof of concept for a smart mast that leverages the versatility of pantograph systems and advances in sensor, actuator, and informatics technologies. More specifically the article addresses key design criteria of transportability, deployability, global stability, and site responsiveness through the development of analytical expressions and control framework, and reduced-scale physical model testing. The case study, a three-tiered tetrahedral mast is composed of three connected sets of planar pantograph systems and deployed by single actuator located between two of the three mast supports. The article discusses the optimum configurations for the individual design criteria and trade-offs to be made between compactness, overturning, and operational power. The design, construction, and experimentation with a 73-cm tall fully deployed physical model reinforce the feasibility of the presented smart mast concept.

This article introduces an adaptation of the Network GEV model for modelling joint choices, named Joint Network GEV (JNG), and its application to the route choice context, named Link-Based JNG (LB-JNG), assuming the choice of a route as the joint choice of all links belonging to that route. The LB-JNG model aims at reproducing the effects of routes overlapping with a theoretical robust framework (since it belongs to the Network GEV, to date the most flexible closed-form model in reproducing covariances), allowing at the same time for easy application to real networks through a closed-form probability statement, a proper definition of its parameters and the availability of an implicit route enumeration algorithm for network loading. The article carries out first an overview of the theoretical properties of the JNG model. Then, the LB-JNG adaptation to route choice is described, and its capability to reproduce the effects of routes overlapping is investigated using some test networks, wherein the performances of the proposed model are also compared with those of other route choice models available in the literature. Finally, an implicit route enumeration algorithm for macroscopic static stochastic network loading, based on a double-step generalization of Dial's STOCH algorithm, is proposed and tested on real size networks.

A Frequency Response Functions (FRFs)-based two-step algorithm to identify stiffness, mass, and viscous damping matrices is developed in this work. The proposed technique uses the difference between the experimentally recorded FRF and their analytical counterparts by minimizing the resultant error function at selected frequency points. In the first step, only mass and stiffness matrices are updated while keeping the uncalibrated viscous damping matrix constant. In the second step, the damping matrix is updated via changes on the selected unknown modal damping ratios. By using a stacking procedure of the presented error function that combines multiple data sets, adverse effects of noise on the estimated modal damping ratios are decreased by averaging the FRF amplitudes at resonant peaks. The application of this methodology is presented utilizing experimentally obtained data. The presented algorithm can perform an accurate structural identification via model updating, with a viscous damping matrix that captures the variation of the modal damping ratios with natural frequencies as opposed to other conventional proportional damping matrix formulations.

This article describes an electric vehicle equipped with a laser scanner and a highly accurate absolute positioning system aimed at surveying the geometry of roads. The main advantages of the proposed system with respect to conventional topographic procedures are the possibility of achieving a much higher density of surveyed points and its efficiency while keeping almost the same accuracy—a standard deviation of 12 mm of absolute error. The data acquisition process is managed by an on-board computer which, in a synchronized way, deals with laser scanning and readings from three real-time-kinematics-enabled millimeter GPS receivers. The three-dimensional position and orientation of the vehicle (6 degrees of freedom) all along its trajectory is calculated off-line by a custom software and, with that information, the system also obtains the absolute coordinates of the road scanned points. This article also presents a rigorous description regarding the theory behind 3D reconstruction and the calibration process.

Uncertainty involved in the experiment data prohibits the wide applications of the finite element (FE) model updating technique into engineering practices. In this article, the Markov Chain Monte Carlo approach with a Delayed Rejection Adaptive Metropolis algorithm is investigated to perform the Bayesian framework for FE updating under uncertainty. A major advantage of this algorithm is that it adopts global and local adaptive strategies, which makes the FE model updating be robust to uncertainty. Another merit of the studied method is that it not only quantitatively predicts structural responses, but also calculates their statistical parameters such as the confidence interval. Impact test data of a grid structure are investigated to demonstrate the effectiveness of the presented FE model updating technique, in which the uncertainty parameters include the vertical and longitudinal spring stiffness that simulate the boundary conditions, the end-fixity factor for modeling semi-rigid connections, and the elastic modulus for simulating the uncertainty associated with material property.

Because car-following (CF) models are fundamental to replicating traffic flow they have received considerable attention over the last 50 years. They are in a continuous state of improvement due to their significant role in traffic microsimulations, intelligent transportation systems, and safety engineering models. This article uses the local linear model tree (LOLIMOT) approach to model driver's CF behavior to incorporate human perceptual imperfections into a CF model. This model defines some localities in the input space. These localities are fuzzy and have overlaps with each other. Specific models for each of the localities are then defined and combined in a fuzzy manner to predict the final output. The model was developed using real world dynamic data sets. Three different data sets were used for training, testing, and validating the model. The performance of the model was compared to a number of existing CF models. The results showed very close agreement between the real data and the LOLIMOT outputs.

An approach for predicting incident durations that are susceptible to severe congestion, the occurrence of secondary incidents, and their joint effect is proposed. First, a fuzzy entropy feature selection methodology is applied to determine redundant factors and rank factor importance with respect to their contribution on the predictability of incident duration. Second, neural network models for incident duration prediction with single and competing uncertainties are developed. The results indicate that alignment, collision type, and downstream geometry may be considered as redundant when modeling incident duration. Rainfall intensity is a highly contributing feature, while lane volume, number of blocked lanes, as well as number of vehicles involved in the incident are among the top ranking factors for determining the extent of duration. Finally, the joint consideration of severe congestion and secondary incident occurrence may improve the generalization power of the prediction models.

User costs of different maintenance actions need to be assessed in road maintenance as well as the maintenance costs. The vehicle operating cost (VOC) and the travel delay cost are two major components of the user costs associated with road maintenance actions. This article simplifies the general calculation models of these two user cost components and develops a multiobjective Markov-based model to minimize both maintenance cost and user cost subject to a number of constraints including the average annual budget limit and the performance requirement. The road deterioration process is modeled as a discrete-time Markov process, the states of road performance are defined in terms of the road roughness, and the state transition probabilities are estimated considering the effects of deterioration and maintenance actions. An example is provided to illustrate the use of the proposed road maintenance optimization model. The results show that the optimal road maintenance plan obtained from the model is practical to implement and is cost-effective compared with the periodical road maintenance plan. The results also indicate that the maintenance cost and the user cost are competitive. When maintenance works are carried out more frequently, the life-cycle maintenance costs will increase while the life-cycle user costs will decrease. This is because the VOC contributes the most amount of the user cost and its change has a contrary trend to the change of the maintenance cost over time.

Household members tend to evacuate as a unit. However, most engineering-based evacuation models treat evacuees as independent and separate entities, and overlook the interactions among household members during an evacuation (i.e., gathering children/spouses or uniting with other family members at home). The omission of these behaviors leads to imprecise modeling of evacuation situations. Additionally, transportation mode choice in a no-notice evacuation has been seldom investigated. We present a framework to incorporate both household-gathering behavior and mode choice in an emergency into an evacuation model to examine the effects of these two issues on evacuation efficiency and network performance. The framework was tested in the Chicago metropolitan region for two hypothetical incidents with evacuation radii 5 and 25 miles. Models that omit gathering behavior yield dangerously optimistic evacuation times and network congestion levels compared to models that include family interactions. These optimistic estimates are significant for a large-scale evacuation—the reduction in the number of evacuees who can reach safe zones in a certain time threshold is nearly 50% between the gathering and no-gathering models. Gathering behavior could also cause distinct effects on network performance for inner and outer areas, the break point of which may be where severe bottlenecks are located. In this study, average travel speed increases on the overall network within 15 miles of the incident location (where downtown Chicago is located), but decreases outside the 15-mile radius. These results would be completely overlooked without incorporating the gathering behavior yet are critical for evacuation management.

A novel AR system is presented, that is most appropriate for urban applications. It is based on monocular vision, is markerless, and does not rely on beacon-based localization technologies (like GPS) or inertial sensors. Additionally, the system automatically calculates occlusions of the built environment on the augmenting virtual objects.

Three datasets from real environments presenting different levels of complexity (geometrical complexity, textures, occlusions) are used to demonstrate the performance of the proposed system. Videos augmented with our system are shown to provide realistic and valuable visualizations of proposed changes of the urban environment. Limitations are also discussed with suggestions for future work.