Traffic information is increasingly regarded as a tool to achieve a more efficient use of the road network. As traffic information is often applied in the context of routine trips, the question arises how travellers integrate traffic information with the knowledge of travel conditions gained through daily experience. To describe this process, the paper proposes a model of perception updating of travel times in the context of departure time decisions. The model applies a CHAID-based classification algorithm to describe how travellers classify trips made under various conditions (departure time and presence of traffic information) into mental classes with comparable expectations in terms of travel time. Thus, it is assumed that the learning process depends on a set of conditions, one of which is the available travel time information. The model is tested through a series of numerical experiments. The results suggest that the model describes learning and adaptation behaviour in a plausible way. Through increased experience, perception of travel times is improved, and more departure time classes are distinguished. However, this does not seem to lead to shorter travel times or higher trip utilities. Also the presence of travel time information may be, depending on the history of trip outcomes, distinguished as a significant indicator of the expected travel time. We conclude that the model provides a good starting point for the further development of learning and adaptation models in the context of ITS.
This article presents an algorithm to generate optimal (real-time) signal timings that distribute queues over a number of signalized intersections and over a number of cycles on any signalized intersection. A discrete-time signal-coordination model is formulated as a dynamic optimization problem and solved using Genetic Algorithms (GA). Signal timings for all intersections in the network during congested periods are decision variables and are represented in the individual GA candidate solutions. The algorithm is applied to a one-way arterial network with 20 signalized intersections. Depending on the traffic demand's variation and the position of critical signals, the algorithm intelligently generates optimal signal timing (offsets) along individual arterials. If critical signals are located at the exit points, the algorithm sets the optimal signal timing that protects them from becoming excessively loaded. If critical signals are located at the entry points, the algorithm ensures that queues are reduced or cleared before released platoons arrive at a downstream signal system. In this article, the simple genetic algorithm (SGA) with multiple epochs is used to solve the signal coordination problem. When a serial SGA is applied to solve traffic control problems, its performance in terms of computation time diminishes as the size of signal networks increases, or the duration of congestion lengthens. Master-slave SGA executed in a parallel computing machine is then used to reduce the execution time. We found that with a master-slave parallelism, SGA can be efficiently executed with significant speed-up, allowing the opportunity to implement the algorithm on real-time signal control systems.
In this article, the relationships among technology ownership and availability, Advanced Traveler Information Systems (ATIS) awareness, and ATIS frequency of use are examined. The four main media considered are the Internet, television, radio, and telephone/other. Awareness and use of ATIS through the four media considered here are different for different users. They also change as household and person characteristics change over time. Since Information and (tele)Communication Technology (ICT) adoption is changing rapidly, this has a significant positive effect on Internet-based ATIS but not a clearly negative effect on all other media. The relationship between awareness and use appears to be nonlinear and dependent on the medium considered. Evidence of substitution between television and Internet and enhancement among other media were also found in this study.
In this article, a two-stage approach to the calculation of reserve capacity for a network is presented. The first stage uses a genetic algorithm to find signal timings that optimize network performance taking traffic reassignment into account. The genetic optimizer, referred to as Genetic Algorithm Transyt Path Flow Estimator (GATRANSPFE), combines the Traffic Network Study Tool (TRANSYT) model, used to estimate performance, with the Path Flow Estimator (PFE) logit assignment tool, used to predict traffic reassignment. In the second stage, the largest common multiplier that can be applied to the Origin-Destination (OD) matrix given the optimized signal timings of stage one is found, again taking reassignment into account. This study deals with the spare capacity available to accommodate the inevitable day-to-day fluctuations in demand. The application of the two-stage procedure is illustrated for a small network with two-signal controlled junctions taken from the literature.
Arterial progression schemes based on the bandwidth criterion are widely used for traffic signal optimization. The schemes provide robust plans for traffic control as well as a variety of design options that can be tailored to specific network and traffic conditions. In recent years, arterial progression optimization was also extended to grid networks. The programs use advanced mathematical programming models which are computationally demanding when applied to large-scale networks. This article describes procedures that dramatically improve the computability of such models and bring them into the realm of real-time application. The procedures are based on, first, selecting and optimizing an arterial priority network or a route priority network. Results are then used in a subsequent stage to determine an optimal plan for the entire network. The procedure is applicable to both uniform- and variable-bandwidth optimization and can accelerate computation by two orders of magnitude, ceteris paribus. This facilitates optimization of large-scale urban networks, provides a capability to analyze many design options and is also amenable for real-time implementation.
This article discusses the nature and consequences of uncertainty in transport systems. Drawing on work from a number of fields, it addresses travellers' abilities to predict variable phenomena, their perception of uncertainty, their attitude to risk and the various strategies they might adopt in response to uncertainty. It is argued that despite the increased interest in the representation of uncertainty in transport systems, most models treat uncertainty as a purely statistical issue and ignore the psychological aspects of response to uncertainty. The principal theories and models currently used to predict travellers' response to uncertainty are presented and a number of alternative modelling approaches are outlined. It is argued that the current generation of predictive models do not provide an adequate basis for forecasting response to changes in the degree of uncertainty or for predicting the likely effect of providing additional information. A number of alternative modelling approaches are identified to deal with travellers' acquisition of information, the definition of their choice set and their choice between the available options. The use of heuristic approaches is recommended as an alternative to more conventional probabilistic methods.
This article identifies the prospective role of a range of intelligent transport systems technologies for the signal control of road traffic. We discuss signal control within the context of traffic management and control in urban road networks and then present a control-theoretic formulation for it that distinguishes the various roles of detector data, objectives of optimization, and control feedback. By reference to this, we discuss the importance of different kinds of variability in traffic flows and review the state of knowledge in respect of control in the presence of different combinations of them. In light of this formulation and review, we identify a range of important possibilities for contributions to traffic management and control through traffic measurement and detection technology, and contemporary flexible optimization techniques that use various kinds of automated learning.
Prospect theoretic hypotheses about the value function are considered in the context of morning commutes. After a review of the formulations of utility functions in the past analyses of commuter departure time choice, this study adopts a formulation where the desirability of an arrival time at work is determined based on decision frames defined in terms of reference points. Probit models are applied to the data collected from randomly selected commuters of a bedroom community in the Kyoto-Osaka-Kobe metropolitan area of Japan, in order to estimate value functions empirically. Two alternative decision frames are compared based on the results of estimation and the empirical validity of the basic properties of the value function is examined.
The article discusses a strategy, referred to as Categorized Arrivals-based Phase Reoptimization at Intersections (CAPRI), which integrates transit signal priority and rail/emergency preemption within a dynamic programming-based real-time traffic adaptive signal control system. The system takes as input sensor data, from detectors, automatic vehicle locators, transponders, etc., for real-time predictions of traffic flow, and “optimally” controls the flow through the network using signal phasing. The system utilizes a traffic adaptive signal control architecture that (1) decomposes the traffic control problem into several subproblems that are interconnected in a hierarchical fashion, (2) predicts traffic flows, at appropriate resolution levels (individual vehicles, platoons of vehicles, transit vehicles, emergency response units, and trains) to enable proactive control, (3) supports various optimization modules for solving the hierarchical subproblems, and (4) utilizes data structure and computer/communication approaches that allow for fast solution of the subproblems, so that each decision can be implemented in the field within an appropriate rolling time horizon of the corresponding subproblem. Simulation-based analyses illustrate the effectiveness of the CAPRI system.
The growing demand for real-time traffic information brought about various types of traffic collection mechanisms in the area of Intelligent Transport Systems (ITS). There are, however, two procedures in making various traffic data into information. First, a robust information-making process of utilizing data into the representative information for each traffic collection mechanism is required. Second, the integration process of fusing the “estimated” information into the “representative information” for each link out of each source is also required. That is, both data reduction and/or data-to-information process and a higher-level information fusion are required. This article focuses on the development of an information fusion algorithm based on a voting technique, fuzzy regression, and Bayesian pooling technique for estimating dynamic link travel time in congested urban road networks. The algorithm has been proposed and validated using field experimental data—GPS probes and detector data collected over various roadway segments. It has been found that the estimated link travel time from the proposed algorithm is more accurate than the mere arithmetic mean counterpart from each traffic source. The limitations of the algorithm and future research agenda have also been discussed.
Parallel traffic simulation is an application of parallel computing techniques that aims to decrease the computation time by engaging different processors of a multiprocessor system or different computers of a network. Very few traffic simulation models have this capability of parallel simulation. However, it is possible to upgrade a simulation program that is not capable of running parallel simulation by applying the method proposed in this article. The method involves dividing the network into regions and simulating each region under a separate instance of the program. Suitable interprocess communication techniques are employed to exchange data between different regions and synchronize time among the different regions. A significant increase in simulation speed is seen when the proposed method is applied to Paramics, a microscopic time-stepping simulation program.
The development of a system for automatically detecting and reporting traffic accidents at intersections was considered. A system with these properties would be beneficial in determining the cause of accidents and could also be useful in determining the features of the intersection that have an impact on safety. A complete system would automatically detect and record traffic conditions associated with accidents such as time of the accident, video of the accident, and the traffic light signal controller parameters. The basic research required to develop the system is considered. This involves developing methods for processing acoustic signals and recognizing accident events from the background traffic events. A database of vehicle crash sounds, car braking sounds, construction sounds, and traffic sounds was created. The mel-frequency cepstral coefficients were computed as a feature vector for input to the classification system. A neural network was used to classify these features into categories of crash and noncrash events. The classification testing results achieved 99 percent accuracy.
Neural networks have been increasingly applied to many problems in transport planning engineering and the feedforward network with the error backpropagation learning rule, usually called simply "Backpropagation," has been the most popular neural network. Backpropagation is easy to implement and has been shown to produce relatively good results in many applications. It is capable of approximating arbitrary nonlinear mappings. However, it is noted that one serious disadvantage in the standard Backpropagation is the slow rate of convergence, requiring very long training times.In order to overcome the long training time and susceptibility to trapping at local minima, several enhanced Backpropagation models have been proposed. In this research, the standard Backpropagation and three enhanced Backpropagation models, Backpropagation with Momentum, Quickprop, and Backpropagation with Momentum & Prime-offset (BPMP), have been studied to compare their performance in terms of computing cost and predictive accuracy.
The concept of automated highway systems (AHS) has been primarily motivated by the rapidly worsening traffic congestion on metropolitan highways and the potential of AHS for drastically increasing vehicle throughput in the existing right-of-way. The overwhelming majority of the research has been focused on automobile-AHS. This paper focuses on the automation of inter-city trucking for the purpose of increasing trucking productivity, of which vehicle throughput is only one of many factors. In the AHS literature, various operating concepts have been developed for an "end-state AHS," but little attention has been paid to the critical issue of how to realize such ultimate systems through a planned sequence of deployment steps. We believe that truck-AHS operating concepts and their deployment sequences must be developed with a needs-driven and technology-steered approach. We first identify the needs of the long-haul trucking industry and the major concerns of key stakeholders. A major need of the long-haul trucking industry is to increase the productivity of its drivers and trucks. A major desire of transportation agencies is to help that industry increase its productivity in a safe and efficient manner but without sacrificing either safety or the infrastructure. A major desire of the driving public is safety. Based on customer needs, stakeholder concerns, and available or promising truck-automation technologies, we then develop design options for several key aspects of truck-AHS operations. After comparing the merits of these options, we develop system operating concepts and deployment sequences to satisfy the customer needs.
Automatic Passenger Counters (APCs) were deployed as part of a Field Operational Test (FOT) of a Fare Transaction and Vehicle Management/Monitoring System (Faretrans VMS). The test took place across several transit operators in Ventura County, California between May 1995 and June 1997. The automatic passenger counters relied on laser sensors installed on buses to count passenger boardings and alightings. The APC data was intended to help the operators meet Federal Section 15 reporting requirements. There were several major technical problems with the APCs. Some of these problems, such as incompatibility with other hardware and software components of the system, were overcome during the course of the FOT. However, data processing problems were not surmounted, and the test revealed significant differences between APC passenger counts and manual passenger counts. The authors conclude that, while APC technology may eventually be a useful means of gathering ridership data, the technology did not perform successfully in this test.
The economics of roadways, and their variability in demand, favor construction of multi-layered and inter-connected networks. Different network layers are designed to different standards and to perform somewhat different functions, though all provide the common function of mobility for a reasonably homogeneous class of vehicles. This paper investigates the geometric orientation of a two-layered roadway system, along with the placement of the interfaces that connect the roadway systems. The path for accessing a highway, and whether the highway is used at all, is derived as a function of the angular orientation and the spacing between highways and interfaces. The research utilizes analytical models, applied to an idealized rectangular grid of city streets. The objective is to use this idealization to investigate how strategic issues in roadway design affect system performance.
The ADVANCE Project, conducted from 1990 through 1997, was one of the first attempts to undertake a large-scale field test of an emerging technology in the ITS field, specifically in the area of in-vehicle dynamic route guidance. It was also an early example of a public-private partnership involving a large corporation, two university research centers, state and federal governments, and other participants in more specific or minor roles. This memoir, initially written at the close of my involvement in the Project in early 1996, seeks to describe the many institutional difficulties of this undertaking, as well as some of the lessons I learned.
The objective of this article is to present a methodology for the estimation of the impact of Advanced Driver Assistance Systems (ADAS) on network efficiency and the environment. The methodology proposes a set of traffic and emissions simulation models, both microscopic and macroscopic. Issues relevant to the models are discussed, and the inputs to the methodology are presented, along with its expected outputs. An indicative application of the methodology is performed, where the traffic impact of the Adaptive Cruise Control (ACC) system is tested in an urban network. The simulation results indicate that the impact of the introduction of Adaptive Cruise Control can be significant in the improvement of the traffic parameters for high system penetration levels and peak period traffic conditions.
This paper presents a new short-term traffic flow prediction system based on an advanced Time Delay Neural Network (TDNN) model, the structure of which is synthesized using a Genetic Algorithm (GA). The model predicts flow and occupancy values at a given freeway section based on contributions from their recent temporal profile (over a few minutes) as well the spatial profile (including inputs from neighboring upstream and downstream sections). An in-depth investigation of the variables pertinent to traffic flow prediction was conducted examining the extent of the "look-back" in time interval, the extent of prediction in the future, the extent of spatial contribution, the resolution of the input data, and their effects on prediction accuracy The model's performance is validated using both simulated and real traffic flow data obtained from the ATMS Testbed in Orange County, California. Both temporal and spatial effects were found to be essential for proper prediction. Results obtained indicate that the prediction errors vary inversely with the extent of the spatial contribution, and that the inclusion of three loop stations in both directions of the subject station is sufficient for practical purposes. Also, the longer the extent of prediction, the more the predicted values tend toward the mean of the actual, for which case the optimal look-back interval also shortens. The results also indicate that the level of data aggregation/resolution should be comparable to the prediction horizon for best accuracy. The model performed acceptably using both simulated and real data. The model also showed potential to be superior to such other well-known neural network models as the Multi layer Feed-forward (MLF) when applied to the same problem.
