Showing papers in "Transportation Research Part C-emerging Technologies in 1997"
TL;DR: In this paper, the authors present a tool for studying acceptance of new technological equipment that is presented here has a major advantage compared with many other studies in that esoteric knowledge of scaling techniques is not required.
Abstract: There is no standard way of measuring driver acceptance of new technology. A review of the literature shows that there are almost as many methods of assessment of acceptance as there are acceptance studies. The tool for studying acceptance of new technological equipment that is presented here has a major advantage compared with many other studies in that esoteric knowledge of scaling techniques is not required. The technique is simple and consists of nine 5-point rating-scale items. These items load on two scales, a scale denoting the usefulness of the system, and a scale designating satisfaction. The technique has been applied in six different studies in different test environments and analyses performed over these studies show that it is a reliable instrument for the assessment of acceptance of new technology. The technique was sensitive to differences in opinion to specific aspects of in-vehicle systems, as well as to differences in opinion between driver groups. In a concluding section explicit recommendations for use of the scale are given.
711 citations
TL;DR: The correlation among error components in repeated measurement data was addressed in this paper with individual-specific random error componentsIn a binary logit model with normal mixing distribution to underscore the significance of traffic information and the potential effect of advanced traveler information systems (ATIS) on route choice.
Abstract: This paper presents a statistical analysis of commuters' route choice including the effect of traffic information. The paper utilizes data collected from two stated preference survey techniques. Two route choice models were estimated. The first model used five hypothetical binary choice sets collected in a computer-aided telephone interview. The objective of the model was to determine how travel time variation affects route choice, and the potential interplay among travel time variation, traffic information acquisition and route choice. The second model used data collected in a mail survey from three binary route choice stated preference scenarios customized according to each respondent's actual commute route and travel time. The objective of the model was to investigate the potential effect of advanced traveler information systems on route choice. The correlation among error components in repeated measurement data was addressed in this paper with individual-specific random error components in a binary logit model with normal mixing distribution. The results underscored the significance of traffic information and the potential effect of advanced traveler information systems (ATIS) on route choice.
325 citations
TL;DR: It was found that not only is delay time more highly valued than normal travel time, which is to be expected, but that drivers become more sensitive to delay time as delay times increased across the range presented.
Abstract: This paper uses a Stated Preference approach to undertake a detailed assessment of the effect on drivers’ route choice of information provided by variable message signs (VMS). Although drivers’ response to VMS information will vary according to the availability of alternative routes and the extent to which they are close substitutes, our findings show that route choice can be strongly influenced by the provision of information about traffic conditions ahead. This has important implications for the use of VMS systems as part of comprehensive traffic management and control systems. The principal findings are that the impact of VMS information depends on: the content of the message, such as the cause of delay and its extent; local circumstances, such as relative journey times in normal conditions; and drivers’ characteristics, such as their age, sex and previous network knowledge. The impact of qualitative indicators, visible queues and delays were examined. It was found that not only is delay time more highly valued than normal travel time, which is to be expected, but that drivers become more sensitive to delay time as delay times increased across the range presented.
214 citations
TL;DR: In this paper, a parallel tabu search heuristic for solving the vehicle routing problem with time windows is developed and implemented on a network of workstations, and it is shown that parallelization of the original sequential algorithm does not reduce solution quality.
Abstract: The vehicle routing problem with time windows models many realistic applications in the context of distribution systems. In this paper, a parallel tabu search heuristic for solving this problem is developed and implemented on a network of workstations. Empirically, it is shown that parallelization of the original sequential algorithm does not reduce solution quality, for the same amount of computations, while providing substantial speed-ups in practice. Such speed-ups could be exploited to quickly produce high quality solutions when the time available for computing a solution is reduced, or to increase service quality by allowing the acceptance of new requests much later, as in transportation on demand systems.
197 citations
TL;DR: A day-to-day dynamic framework is developed to study network dynamics under real-time information and responsive signal control systems, primarily on commuter trips from home to work in a general network.
Abstract: A day-to-day dynamic framework, in which the DYnamic Network Assignment Simulation Model for Advanced Road Telematics (DYNASMART) simulation-assignment model is applied to evaluate the performance of traffic networks, is developed to study network dynamics under real-time information and responsive signal control systems. The focus in this paper is primarily on commuter trips from home to work in a general network. Two levels of tripmaker decision-making processes are incorporated: (1) day-to-day dynamics and (2) real-time dynamics. Day-to-day dynamics consider the choices of departure time and route according to indifference bands of tolerable ‘schedule delay’, defined as the difference between the user's actual and preferred arrival times, and are thus governed by tripmakers' daily learning processes. Real-time dynamics consider en-route switching decisions in response to real-time information on prevailing traffic conditions. The resulting flows could be used in updating the supplied real-time information as well as the traffic control parameters. Two types of traffic control responsiveness are evaluated: (1) daily adjustment of signal timing parameters to reflect the preceding day's traffic patterns; and (2) real-time traffic-responsive signal control driven by prevailing flow patterns. The framework is illustrated through numerical experiments to investigate the day-to-day evolution of network flows under real-time information and responsive signal control, and assess the effectiveness of such information in a proper dynamic perspective.
177 citations
TL;DR: A cooperation based neural networks traffic flow model is proposed, which aims at being integrated into a real time adaptive urban traffic control system.
Abstract: Over the past few years, artificial intelligence techniques have played important roles in the design of sophisticated traffic management systems. In this paper, we propose a cooperation based neural networks traffic flow model, which aims at being integrated into a real time adaptive urban traffic control system. The modelling is separated into two steps. Firstly, the traffic flow is modelled on a single signalized link by a local neural network. Secondly, based on communications between local neural networks, the traffic flow is modelled over a wide network of junctions. Based on simulated data, the paper concludes on the potentials of neural networks applied to traffic flow modelling. One minute ahead predictions of the queue lengths and the output flows have been obtained with fairly good accuracy. Nevertheless, it emphasizes the real need to further investigate these techniques on experimental data.
157 citations
TL;DR: A path flow estimator suitable for use in conjunction with urban traffic monitoring, control and guidance is set out, in particular the incorporation of prior information on the relative magnitudes of origin-destination movements.
Abstract: The paper sets out a path flow estimator suitable for use in conjunction with urban traffic monitoring, control and guidance. Travel time for each link in the network is partitioned into undelayed travel time and delay. The links are assumed to be of two types. For the first type of link, an external estimate of flow and travel time over the estimation interval is provided. The second type of link is characterised by a finite capacity, and delay is incurred where demand would otherwise be in excess of capacity. Demand is determined by a logit route choice model. An equivalent convex programming problem is formulated and an iterative solution procedure is set out. The estimation of the dispersion parameter in the logit model is discussed, and a column generation method to avoid path enumeration is proposed. Diagnostic procedures and a number of other practical enhancements to the procedure, in particular the incorporation of prior information on the relative magnitudes of origin-destination movements, are considered.
151 citations
TL;DR: The results presented in this paper confirm that neural network models can provide fast and reliable incident detection on freeways and demonstrate how improvements in model performance can be achieved using variable decision thresholds.
Abstract: This paper discusses a multi-layer feedforward (MLF) neural network incident detection model that was developed and evaluated using field data. In contrast to published neural network incident detection models which relied on simulated or limited field data for model development and testing, the model described in this paper was trained and tested on a real-world data set of 100 incidents. The model uses speed, flow and occupancy data measured at dual stations, averaged across all lanes and only from time interval t. The off-line performance of the model is reported under both incident and non-incident conditions. The incident detection performance of the model is reported based on a validation-test data set of 40 incidents that were independent of the 60 incidents used for training. The false alarm rates of the model are evaluated based on non-incident data that were collected from a freeway section which was video-taped for a period of 33 days. A comparative evaluation between the neural network model and the incident detection model in operation on Melbourne's freeways is also presented. The results of the comparative performance evaluation clearly demonstrate the substantial improvement in incident detection performance obtained by the neural network model. The paper also presents additional results that demonstrate how improvements in model performance can be achieved using variable decision thresholds. Finally, the model's fault-tolerance under conditions of corrupt or missing data is investigated and the impact of loop detector failure/malfunction on the performance of the trained model is evaluated and discussed. The results presented in this paper provide a comprehensive evaluation of the developed model and confirm that neural network models can provide fast and reliable incident detection on freeways.
144 citations
TL;DR: In this paper, a nonparametric approach based on multi-layer perceptron neural networks (MLP) is proposed to measure performance of public transport services based on the concept of productive efficiency.
Abstract: This paper is concerned with measuring performance of public transport services based on the concept of productive efficiency. A new nonparametric approach is proposed based on multi-layer perceptron neural networks (MLPs). The advantages and limitations of this approach are discussed and compared with those of mathematical programming and econometric techniques. The MLP is used, along with data envelopment analysis (DEA) and corrected least squares (COLS), to set out comparative annual efficiency measures for the London Underground, for the period 1970 to 1994. It is argued that the MLP approach is superior to traditionally applied techniques since it is both nonparametric and stochastic and offers greater flexibility. Finally, it is demonstrated that the proposed MLP efficiency analysis has important practical implications for decision making.
134 citations
TL;DR: This paper presents a fundamentally different approach for optimal signal timing that eliminates the need for extremely complex models of the traffic dynamics as a component of the control strategy.
Abstract: A long-standing problem in traffic engineering is to optimize the flow of vehicles through a given road network. Improving the timing of the traffic signals at intersections in the network is generally the most powerful and cost-effective means of achieving this goal. However, because of the many complex aspects of a traffic system—human behavioral considerations, vehicle flow interactions within the network, weather effects, traffic accidents, long-term (e.g. seasonal) variation, etc.—it has been notoriously difficult to determine the optimal signal timing. This is especially the case on a system-wide (multiple intersection) basis. Much of this difficulty has stemmed from the need to build extremely complex models of the traffic dynamics as a component of the control strategy. This paper presents a fundamentally different approach for optimal signal timing that eliminates the need for such complex models. The approach is based on a neural network (or other function approximator) serving as the basis for the control law, with the weight estimation occurring in closed-loop mode via the simultaneous perturbation stochastic approximation (SPSA) algorithm. The neural network function uses current traffic information to solve the current (instantaneous) traffic problem on a system-wide basis through an optimal signal timing strategy. The approach is illustrated by a realistic simulation of a nine-intersection network within the central business district of Manhattan, New York.
118 citations
TL;DR: Initial simulation results show that such an approach is promising, and a nonlinear approach for designing local traffic-responsive ramp controls using artificial neural networks is proposed.
Abstract: This paper proposes a nonlinear approach for designing local traffic-responsive ramp controls using artificial neural networks. The problem is formulated as a nonlinear feedback control problem, where the system model is the well known hydrodynamic model developed by Lighthill and Whitham (1955), and Richards (1956), the model's flow-density relationship is nonlinear, and the feedback nonlinear controllers are composed of one or a number of feed-forward neural networks. These neural network controllers are of integral (I) or proportional-plus-integral (PI) type, and can be tuned on-line to achieve prescribed performance. Initial simulation results show that such an approach is promising.
TL;DR: The question of when do unfamiliar drivers become familiar is addressed and several criteria for assessing familiarity are suggested, and an approximate-reasoning based model and a random utility model are implemented and compared.
Abstract: Route choice behavior of familiar and unfamiliar drivers who use the same network is explored and compared. The data were collected using a driver simulator. The results obtained indicate larger homogeneity among the unfamiliar drivers in terms of their switching and diverting behavior, while familiar drivers demonstrate larger taste and preferences variations. Two choice models are implemented and compared: an approximate-reasoning based model and a random utility model. The two models produce comparable results and provide interesting insights into choice behavior of familiar and unfamiliar drivers. The question of when do unfamiliar drivers become familiar is addressed and several criteria for assessing familiarity are suggested.
TL;DR: In this article, the authors describe the validation of a route choice simulator known as VLADIMIR (Variable Legend Assessment Device for Interactive Measurement of Individual Route choice), which is an interactive computer-based tool designed to study drivers' route choice behavior.
Abstract: This paper describes the validation of a route choice simulator known as VLADIMIR (Variable Legend Assessment Device for Interactive Measurement of Individual Route choice). VLADIMIR is an interactive computer-based tool designed to study drivers’ route choice behaviour. It has been extensively used to obtain data on route choice in the presence of information sources such as Variable Message Signs or In-Car Navigation devices. The simulator uses a sequence of digitized photographs to portray a real network with junctions, links, landmarks and road signs. Subject drivers are invited to make journeys between specified origins and destinations under a range of travel scenarios, during which the simulator automatically records their route choices. This paper describes validation experiments carried out during the period Summer 1994 to Autumn 1995 and reports on the results obtained. Each experiment involved a comparison of routes selected in real life with those driven under simulated conditions in VLADIMIR. The analysis included investigation of the subjects’ own assessment of the realism of the VLADIMIR routes they had chosen, a comparison of models based on the real life routes with models based on VLADIMIR routes, and a statistical comparison of the two sets of routes. After an extensive series of data collection exercises and analyses, we have concluded that a well designed simulator is able to replicate real life route choices with a very high degree of detail and accuracy. Not only was VLADIMIR able to precisely replicate the route choices of drivers who were familiar with the network but it also appears capable of representing the kind of errors made and route choice strategies adopted by less familiar drivers. Furthermore, evidence is presented to suggest that it can accurately replicate route choice responses to roadside VMS information.
TL;DR: This research investigates the feasibility of developing a traffic monitoring detector for the purpose of reliable on-line vehicle classification to aid traffic management systems and provides an overview of the LPC pre-processing and feature extraction technique as applied to audio monitoring of road traffic.
Abstract: The aim of this research is to investigate the feasibility of developing a traffic monitoring detector for the purpose of reliable on-line vehicle classification to aid traffic management systems. The detector used was a directional microphone connected to a DAT (Digital Audio Tape) recorder. The digital signal was pre-processed by LPC (Linear Predictive Coding) parameter conversion based on autocorrelation analysis. A Time Delay Neural Network (TDNN) was chosen to classify individual travelling vehicles based on their speed-independent acoustic signature. Locations for data acquisition included roadside recordings at a number of two-way urban road sites in the city of Leeds with no control over the environmental parameters such as background noise, interference from other travelling vehicles or the speed of the recorded vehicles. The results and performance analysis of TDNN vehicle classification, the convergence for training patterns and accuracy of test patterns are fully illustrated. The paper also provides a description of the TDNN architecture and training algorithm, and an overview of the LPC pre-processing and feature extraction technique as applied to audio monitoring of road traffic. In the final phase of the experiment, the four broad categorisations of vehicles for training the network consisted of: buses or lorries; small or large saloons; various types of motorcycles; and light goods vehicles or vans. A TDNN network was successfully trained with 94% accuracy for the training patterns and 82.4% accuracy for the test patterns.
TL;DR: Two shared-memory and one message-passing algorithms for time-dependent shortest-path algorithms that can be used in real-time ITS applications are designed, implemented, coded and computationally tested on actual and random networks.
Abstract: The development of Intelligent Transportation Systems (ITS) and the resulting need for real-time traffic management and route guidance models require fast shortest-path algorithms that can account for the dynamics of traffic networks. The objective of this paper is to introduce parallel designs for time-dependent shortest-path algorithms that can be used in real-time ITS applications. In this paper, two shared-memory and one message-passing algorithms are designed, implemented, coded and computationally tested on actual and random networks. The reported tests are performed on CRAY supercomputers, but the algorithms can be readily ported to lower-end multiprocessor machines.
TL;DR: The performance of parallel computing in group-based optimisation of signal timings improves as the number of independent paths in the task schedule increases.
Abstract: This paper presents the use of parallel computing in the group-based optimisation of signal timings for area traffic control. The scope of parallel computing to speed up the optimisation heuristics, consisting of a mixture of network-wide steps, in which control variables for signal settings at all junctions are changed simultaneously, and junction-based steps, in which the variables at each junction are changed in turn, is examined. Parallel algorithms and a dynamic load balancing scheme are developed for these optimisation heuristics. A trial network from Leicestershire, England is used to demonstrate the performance of parallel computing. A useful saving in computing time is obtained for the example of the standard optimisation procedure, and for the examples of random offsets calculation and sequencing analysis substantial reductions of computing times are achieved. In general, the performance of parallel computing in group-based optimisation of signal timings improves as the number of independent paths in the task schedule increases.
TL;DR: In this paper, a decision task model that summarizes the need for information technology support during transportation improvement site selection is presented. But, the task model is based on a decision support system requirement specification that outlines integrated information capabilities provided by geographic information system and group support system (GSS) technologies.
Abstract: Transportation improvement site selection exemplifies transportation decision making that is collaborative in nature and geographically based. Such decision-making is part of a broad societal trend toward shared and participatory discussions about public investment. Perspectives from three different transportation decision contexts in the Puget Sound Region of Washington State, a regional council, a county government and a public–private Coalition group, are combined with a literature review to develop a decision task model that summarizes the need for information technology support during transportation improvement site selection. The task model guides the development of a decision support system requirement specification that outlines integrated information capabilities provided by geographic information system (GIS) and group support system (GSS) technologies. Together, GIS and GSS capabilities contribute to evolving group-based GIS. The kinds of capabilities a group-based GIS could offer in addressing the needs are identified. A report on the use of a prototype, group-based GIS called Spatial Group Choice highlights the possibilities in an inter-organizational coalition decision context. The conclusions discuss needs for future technology developments and social–behavioral science studies on these developments.
TL;DR: This conceptual paper provides an overview of a parallel transportation/land use modelling environment and suggests a parallel distributed processing structure composed of processors and links between processors forming a neural network for spatial analysis and modelling.
Abstract: We provide in this conceptual paper an overview of a parallel transportation/land use modelling environment. We argue that sequential urban modelling does not well represent complex urban dynamics. Instead, we suggest a parallel distributed processing structure composed of processors and links between processors. Each processor is a set of neurons and weights between neurons forming a neural network. For spatial systems neural networks have two main paradigms which are processes simulation and pattern association. Parallel distributed processing offers a new methodology to represent the relational structure between elements of a transportation/land use system and thus helping to model those systems. We also provide a set of advantages, drawbacks and some research directions about the usage of neural networks for spatial analysis and modelling.
TL;DR: In this paper, the authors examined the positive effects on highway capacity implementing automatic vehicle control and proposed general rules for traffic control in order to make current highways more efficient during the transition stage.
Abstract: This work examines the positive effects on highway capacity implementing automatic vehicle control. Mixed traffic streams of a mimic freeway interchange simulate an on-ramp section with autopilot-equipped vehicles and normal manual vehicles. The simulation illustrates characteristics of speed, volume and concentration of mixed flows. The capacity trend is presented with mixed ratios of equipped cars and its market occupation rate. In order to make current highways more efficient during the transition stage, general rules for traffic control are proposed.
TL;DR: In this paper, the link layer is modeled using vehicle conservation flow models and traffic conditions on the highway are given by a pair of density and velocity profiles which are assumed to be consistent with the demand on, and the capability of, the highway system.
Abstract: Controls for the link layer in the Automated Highway System (AHS) hierarchy proposed in the Partners for Advanced Transit and Highways (PATH) are developed. The link layer is modeled using vehicle conservation flow models. Desired traffic conditions on the highway are given by a pair of density and velocity profiles which are assumed to be consistent with the demand on, and the capability of, the highway system. The link layer control laws presented in this paper then stabilize the actual traffic condition to the desired values. Control laws are derived for three highway topologies: a single lane highway, a highway with multiple discrete lanes and a two-dimensional highway with an arbitrary flow pattern. The control laws obtained for each of the topologies is distributed and are suited for implementation in the lower levels of the AHS control hierarchy. Simulation results are also presented.
TL;DR: In this paper, four neural network representations for detecting incidents on signalized arterials using multiple data sources were considered, including inductive loop detectors and travel times collected from vehicle probes travelling through the street network.
Abstract: This research considers four neural network representations for detecting incidents on signalized arterials using multiple data sources. Two incident detection algorithms process unique data sources separately: inductive loop detectors, and travel times collected from vehicle probes travelling through the street network. The networks then combine the algorithm inferences about traffic conditions to identify highway links on which incidents are occurring. The four networks consider the following input and structure representations, added incrementally: (1) the two algorithm output values alone; (2) a weighted geometric sum of previous network output values; (3) algorithm scores from links immediately upstream and downstream of the subject link; and (4) weighted geometric sums of previous input values. The four representations were trained as feed-forward networks using error back propagation. Time series inputs were represented with extra processing units and fixed weight connections. The networks were trained with data generated by traffic simulation permitting deliberate control of traffic demand, operation and incident conditions. Each network was trained until performance began to degrade on a reserved data set not used for training. Adding the output time series permitted two of the 24 incidents to be detected sooner than with the network that did not include this input. Similarly, using information from adjacent links in time series permitted all of the incidents to be detected by at least the third time period.
TL;DR: The main intention of the study was not the design of new control strategies but the investigation of the qualitative impact of various control measures such as ramp metering, route diversion via variable message signs and signal control, and of their integration under realistic conditions.
Abstract: The paper presents the work performed within the European DRIVE II Project EUROCOR (European Urban Corridor Control) in relation to modelling and integrated control of the M8 Eastbound Corridor in Glasgow. The main intention of the study was not the design of new control strategies but the investigation of the qualitative impact of various control measures such as ramp metering, route diversion via variable message signs and signal control, and of their integration under realistic conditions. To this end, several scenarios of included control measures were investigated using the macroscopic modelling tool METACOR (Modele d’Ecoulement du Trafic sur Coridor), demonstrating the issues and potential benefits arising from different levels of integration.
TL;DR: The dynamic traffic assignment problem is formulated in the space of splitting rates rather than link and route flows and a distributed algorithm for computation of dynamic user-equilibria is specified.
Abstract: The dynamic traffic assignment problem is formulated in the space of splitting rates rather than link and route flows. A distributed algorithm for computation of dynamic user-equilibria is specified. The algorithm has been implemented on a Meiko Computing Surface with 32 T800 processors and some numerical results are given. We do not yet have a general proof of convergence for the algorithm but we have been able to demonstrate convergence with all test networks used.
TL;DR: A parallel video-based image analysis system which is capable of extracting movement information, including direction and speed, of road vehicular traffic over any part of a road surface and has been tested using data for a signal-controlled junction aiming to capture an opposed turning traffic movement with promising results.
Abstract: Road traffic movement is a very important source of information in traffic management. Although systems exist which can detect the presence of a vehicle and its speed under certain conditions, there is generally a lack of effective means to measure both the speed and direction of traffic movement. This is particularly true for road junctions, where conflicting traffic shares the same space and where some control strategy could be more effectively applied with the help of speed and direction estimates. The increasing use of closed circuit television (CCTV) systems has provided the opportunity to apply image processing techniques to extract such information. However, such techniques are computationally intensive in general, and the application of parallel processing methods is one of the best choices which could bring the desired acquisition of movement information into practical reality. This paper describes a parallel video-based image analysis system which is capable of extracting movement information, including direction and speed, of road vehicular traffic over any part of a road surface. The prototype has been implemented on an array of 36 transputers and an image grabber with a SUN SPARC IPC as the host machine. The software mainly consists of median filtering, feature extraction, spatio-temporal analysis, matching of image features in successive images by neural networks and aggregation of matched results. This algorithm has been tested using data for a signal-controlled junction aiming to capture an opposed turning traffic movement with promising results. It has also been shown that a real-time system based on the described algorithm is feasible.
TL;DR: A lane routing algorithm for vehicles entering the highway to exit successfully at their destinations while balancing the lane capacity usage of the highway (even in the face of lane-blocking incident) is proposed.
Abstract: This paper considers a lane routing problem in an Automated Highway System (AHS). Lane routing is important because the problem of managing the traffic flow on a multilane AHS plays a major role in an urban environment. On the other hand, because analytical approaches are impossible for various, almost unpredictable traffic situations, traffic simulations are indispensible for the design and the evaluation of various control strategies of the AHS. The discrete event modeling technique can be employed to describe both macroscopic and microscopic models according to the level of abstraction. The DEVS (Discrete Event System Specification) formalism has been employed for traffic modeling and simulation of the AHS. In this paper, we propose a lane routing algorithm for vehicles entering the highway to exit successfully at their destinations while balancing the lane capacity usage of the highway (even in the face of lane-blocking incident). After that, we present a methodology for the performance evaluation of the proposed algorithm using discrete event modeling and simulation. The simulation results show that the proposed algorithm gives good performance for traffic flow control in the AHS.
Journal Article•
TL;DR: This model is built upon a cooperation of local neural networks, with each neural net being in charge of modeling the traffic flow on one single signalized link.
Abstract: This paper presents a cooperation-based neural network traffic flow model. The model is built upon a cooperation of local neural networks, with each neural net being in charge of modeling the traffic flow on one single signalized link. Exchanges of information are established between each local neural net to yield traffic flow modeling on a junction. The model meets the requirements for being integrated into a real time adaptive urban traffic control system.