Development and evaluation of a knowledge-based system for traffic congestion management and control

This paper describes a real-time knowledge-based system (KBS) for decision support to Traffic Operation Center personnel in the selection of integrated traffic control plans after the occurrence of non-recurring congestion, on freeway and arterial networks. The uniqueness of the system, called TCM, lies in its ability to cooperate with the operator, by handling different sources of input data and inferred knowledge, and providing an explanation of its reasoning process. A data fusion algorithm for the analysis of congestion allows to represent and interpret different types of data, with various levels of reliability and uncertainty, to provide a clear assessment of traffic conditions. An efficient algorithm for the selection of control plans determines alternative traffic control responses. These are proposed to an operator, along with an explanation of the reasoning process that led to their development and an estimation of their expected effect on traffic. The validation of the system, which is one of only few examples of validation of a KBS in transportation, demonstrates the validity of the approach. The evaluation results, in a simulated environment demonstrate the ability of TCM to reduce congestion, through the formulation of traffic diversion and control schemes.     

A multi-agent architecture for cooperative inter-jurisdictional traffic congestion management

This paper describes , an innovative multi-agent architecture for the provision of real-time decision support to Traffic Operations Center personnel for coordinated, inter-jurisdictional traffic congestion management on freeway and surface street (arterial) networks. is composed of two interacting knowledge-based systems that perform cooperative reasoning and resolve conflicts, for the analysis of non-recurring congestion and the on-line formulation of integrated control plans. The two agents support incident management operations for a freeway and an adjacent arterial subnetwork and interact with human operators, determining control recommendations in response to the occurrence of incidents. The multi-decision maker approach adopted by reflects the spatial and administrative organization of traffic management agencies in US cities, providing a cooperative solution that exploits the agencies’ willingness to cooperate and unify their problem-solving capabilities, yet preserves the different levels of authority and the inherent distribution of data and expertise. The interaction between the agents is based on the functionally accurate, cooperative paradigm, a distributed problem solving approach aimed at producing consistent solutions without requiring the agents to have shared access to all globally available information. The cornerstone of this approach is the assumption that effective solutions can be efficiently obtained even when complete and up-to-date information is not directly available to the agents, thus reducing the need for complex data communication networks and synchronization time delays. The simulation-based evaluation of the system performance validates this assumption. The paper focuses on the distributed architecture of the agents and on their communication and decision making characteristics.     

Evaluation of the effect of cooperative infrastructure-to-vehicle systems on driver behavior

In-vehicle technologies and co-operative services have potential to ease congestion problems and improve traffic safety. This paper investigates the impact of infrastructure-to-vehicle co-operative systems, case of CO-OPerative SystEms for Intelligent Road Safety (COOPERS), on driver behavior. Thirty-five test drivers drove an instrumented vehicle, twice, with and without the system. Data related to driving behavior, physiological measurements, and user acceptance was collected. A macro-level approach was used to evaluate the potential impact of such systems on driver behavior and traffic safety. The results in terms of speeds, following gaps, and physiological measurements indicate a positive impact. Furthermore, drivers’ opinions show that the system is in general acceptable and useful.     

A model for assessing the effects of dynamic travel time information via variable message signs

In view of the serious traffic congestion during peak hours in most metropolitan areas around the world and recent improvement of information technology, there is a growing aspiration to alleviate road congestion by applications of electronic information and communication technology. Providing drivers with dynamic travel time information such as estimated journey times on major routes should help drivers to select better routes and guide them to utilise existing expressway network. This can be regarded as one possible strategy for effective traffic management. This paper aims to investigate the effects and benefits of providing dynamic travel time information to drivers via variable message signs at the expressway network. In order to assess the effects of the dynamic driver information system with making use of the variable message signs, a time-dependent traffic assignment model is proposed. A numerical example is used to illustrate the effects of the dynamic travel time information via variable message signs. This revised version was published online in June 2006 with corrections to the Cover Date.     

Adaptive traffic signal control with actor-critic methods in a real-world traffic network with different traffic disruption events

The transportation demand is rapidly growing in metropolises, resulting in chronic traffic congestions in dense downtown areas. Adaptive traffic signal control as the principle part of intelligent transportation systems has a primary role to effectively reduce traffic congestion by making a real-time adaptation in response to the changing traffic network dynamics. Reinforcement learning (RL) is an effective approach in machine learning that has been applied for designing adaptive traffic signal controllers. One of the most efficient and robust type of RL algorithms are continuous state actor-critic algorithms that have the advantage of fast learning and the ability to generalize to new and unseen traffic conditions. These algorithms are utilized in this paper to design adaptive traffic signal controllers called actor-critic adaptive traffic signal controllers (A-CATs controllers).The contribution of the present work rests on the integration of three threads: (a) showing performance comparisons of both discrete and continuous A-CATs controllers in a traffic network with recurring congestion (24-h traffic demand) in the upper downtown core of Tehran city, (b) analyzing the effects of different traffic disruptions including opportunistic pedestrians crossing, parking lane, non-recurring congestion, and different levels of sensor noise on the performance of A-CATS controllers, and (c) comparing the performance of different function approximators (tile coding and radial basis function) on the learning of A-CATs controllers. To this end, first an agent-based traffic simulation of the study area is carried out. Then six different scenarios are conducted to find the best A-CATs controller that is robust enough against different traffic disruptions. We observe that the A-CATs controller based on radial basis function networks (RBF (5)) outperforms others. This controller is benchmarked against controllers of discrete state Q-learning, Bayesian Q-learning, fixed time and actuated controllers; and the results reveal that it consistently outperforms them.     

An impact analysis of traffic image information system on driver travel choice

A driver is one of the main components in a transportation system that influences the effectiveness of any active demand management (ADM) strategies. As such, the understanding on driver behavior and their travel choice is crucial to ensure the successful implementation of ADM strategies in alleviating traffic congestion, especially in city centres. This study aims to investigate the impact of traffic information dissemination via traffic images on driver travel choice and decision. A relationship of driver travel choice with respect to their perceived congestion level is developed by an integrated framework of genetic algorithm–fuzzy logic, being a new attempt in driver behavior modeling. Results show that drivers consider changing their travel choice when the perceived congestion level is medium, in which changing departure time and diverting to alternative roads are two popular choices. If traffic congestion escalates further, drivers are likely to cancel their trip. Shifting to public transport system is the least likely choice for drivers in an auto-dependent city. These findings are important and useful to engineers as they are required to fully understand driver (user) sensitivity to traffic conditions so that relevant active travel demand management strategies could be implemented successfully. In addition, engineers could use the relationships established in this study to predict drivers’ response under various traffic conditions when carrying out modeling and impact studies.     

The Gothenburg congestion charge. Effects,design and politics

This paper summarizes the traffic effects of the Gothenburg congestion charges introduced in 2013. The system is similar to the system introduced in Stockholm in 2006; both are designed as time-of-day dependent cordon pricing systems. We find that many effects and adaptation strategies are similar to those found in Stockholm, indicating a high transferability between smaller and larger cities with substantial differences in public transport use. However, there are also important differences regarding some of the effects, the accuracy of the model forecasts and public support arising from different topologies, public transport use, congestion levels and main objectives communicated to the public. Finally, the Gothenburg case suggests that whether congestion charges are introduced or not depends on the support among the political parties, and that this is determined primarily by the prevailing institutional setting and power over revenues, and to a lower extent by the public support, and benefits from congestion reduction.     

Pricing scheme design of ridesharing program in morning commute problem

This paper examines the dynamic user equilibrium of the morning commute problem in the presence of ridesharing program. Commuters simultaneously choose departure time from home and commute mode among three roles: solo driver, ridesharing driver, and ridesharing rider. Considering the congestion evolution over time, we propose a time-varying compensation scheme to maintain a positive ridesharing ridership at user equilibrium. To match the demand and the supply of ridesharing service over time, the compensation scheme should be set according to the inconvenience cost functions and the out-of-pocket cost functions. When the price charged per time unit is higher than the inconvenience cost per time unit perceived by the ridesharing drivers, the ridesharing participants will travel at the center of peak hours and solo drivers will commute at the two tails. Within the feasible region with positive ridership, the ridesharing program can reduce the congestion and all the commuters will be better off. To support system optimum (SO), we derive a time-varying toll combined with a flat ridesharing price from eliminating queuing delay. Under SO toll, the ridesharing program can attract more participants and have an enlarged feasible region. This reveals that the commuters are more tolerant to the inconvenience caused by sharing a ride at SO because of the lower travel time. Compared with no-toll equilibrium, both overall congestion and individual travel cost are further reduced at SO.     

Optimisation of parking guidance and information systems display configurations

Operators of parking guidance and information (PGI) systems often have difficulty in determining the best car park availability information to present to drivers in periods of high demand. This paper describes a behavioural model of parking choice incorporating drivers perceptions of waiting times at car parks based on PGI signs. This model was used to predict the influence of PGI signs on the overall performance of the traffic system.Relationships were developed for estimating the arrival rates at car parks based on trip patterns, driver characteristics, car park attributes as well as the car park availability information displayed on PGI signs. Drivers' perceptions of waiting times at car parks were assumed to be influenced by the PGI signs for observers of the signs and actual car park utilisation levels for non-observers. The model assumes that the choice of car park does not change after entering the city centre, even if conditions observed are different from those initially perceived.A mathematical programme was formulated to determine the optimal display PGI sign configuration to minimise queue lengths and vehicle kilometres of travel (VKT). The model was limited to off-street parking choices and illegal parking was not incorporated. A simple genetic algorithm was used to identify solutions that significantly reduced queue lengths and VKT compared with existing practices.These procedures were applied to an existing PGI system operating in Tama New Town near Tokyo. Significant reductions in queue lengths and VKT were predicted using the optimisation model. This would reduce traffic congestion and lead to various environmental benefits.     

Evaluating the benefits of a combined route guidance and road pricing system in a traffic network with recurrent congestion

When drivers do not have complete information on road travel time and thus choose their routes in a stochastic manner or based on their previous experience, separate implementations of either route guidance or road pricing cannot drive a stochastic network flow pattern towards a system optimum in a Wardropian sense. It is thus of interest to consider a combined route guidance and road pricing system. A road guidance system could reduce drivers' uncertainty of travel time through provision of traffic information. A driver who is equipped with a guidance system could be assumed to receive complete information, and hence be able to find the minimum travel time routes in a user-optimal manner, while marginal-cost road pricing could drive a user-optimal flow pattern toward a system optimum. Therefore, a joint implementation of route guidance and road pricing in a network with recurrent congestion could drive a stochastic network flow pattern towards a system optimum, and thus achieve a higher reduction in system travel time. In this paper the interaction between route guidance and road pricing is modeled and the potential benefit of their joint implementation is evaluated based on a mixed equilibrium traffic assignment model. The private and system benefits under marginal-cost pricing and varied levels of market penetration of the information systems are investigated with a small and a large example. It is concluded that the two technologies complement each other and that their joint implementation can reduce travel time more efficiently in a network with recurrent congestion.     

Effects of alternative road pricing systems on network performance

Four road pricing systems, with charges based on cordons crossed, distance travelled, time spent travelling and time spent in congestion, have been tested using the congested assignment network model SATURN and its elastic assignment demand response routine, SATEASY. All tests have been based on a SATURN application of the city of Cambridge, with charges imposed inside an appropriate ring of bypasses. While initial results showed that congestion pricing achieved the greatest increase in average speed in the charged area, later analysis cast doubt on its superiority. Congestion pricing is able to distinguish more effectively the extent to which different types of journey contribute to congestion and achieves given reductions in travel at lower levels of charge. However, it is much less effective in reducing distance travelled and, by encouraging use of minor roads, may achieve far smaller environmental benefits. Time-based pricing performs better than the other systems on most indicators. Generally, the results suggest that when rerouting effects are included in the predictive modelling process the benefits of road pricing may be significantly smaller than previously expected.     

Machine learning for multi-jurisdictional optimal traffic corridor control

Urban traffic corridors are often controlled by more than one agency. Typically in North America, a state of provincial transportation department controls freeways while another agency at the municipal or city level controls the nearby arterials. While the different segments of the corridor fall under different jurisdictions, traffic and users know no boundaries and expect seamless service. Common lack of coordination amongst those authorities due to lack of means for information exchange and/or possible bureaucratic ‘institutional grid-lock’ could hinder the full potential of technically-possible integrated control. Such institutional gridlock and related lack of timely coordination amongst the different agencies involved can have a direct impact on traffic gridlock. One potential solution to this problem is through integrated automatic control under intelligent transportation systems (ITS). Advancements in ITS and communication technology have the potential to considerably reduce delay and congestion through an array of network-wide traffic control and management strategies that can seamlessly cross-jurisdictional boundaries. Perhaps two of the most promising such control tools for freeway corridors are traffic-responsive ramp metering and/or dynamic traffic diversion possibly using variable message signs (VMS). Technically, the use of these control methods separately might limit their potential usefulness. Therefore, integrated corridor control using ramp metering and VMS diversion simultaneously might be synergetic and beneficial. Motivated by the above problem and potential solution approach, the aim of the research presented in this paper is to develop a self-learning adaptive integrated freeway-arterial corridor control for both recurring and non-recurring congestion. The paper introduces the use of reinforcement learning, an Artificial Intelligence method for machine learning, to provide optimal control using ramp metering and VMS routing in an integrated agent for a freeway-arterial corridor. Reinforcement learning is an approach whereby the control agent directly learns optimal strategies via feedback reward signals from its environment. A simple but powerful reinforcement learning method known as Q-learning is used. Results from an elaborate simulation study on a key corridor in Toronto are very encouraging and discussed in the paper.     

Stated preferences for investigating commuters' diversion propensity

A reasonable response to increasing traffic congestion may come from the rapidly developing traveler information systems. Such systems may be successful if they effectively influence drivers' enroute decisions; in this regard, a critical factor may be commuters' willingness to divert from their regular route in response to information about traffic congestion. This study evaluates the effects of real-time traffic information along with driver attributes, roadway characteristics and situational factors on drivers' willingness to divert.The empirical portion of this study is based on a survey of downtown Chicago automobile commuters. The stated preference approach was used to study commuters' diversion propensity. Drivers expressed a higher willingness to divert if expected delays on their usual route increased, if the congestion was incident-induced as opposed to recurring, if delay information was received from radio traffic reports compared with observing congestion, and if trip direction was home-to-work rather than work-to-home. Respondents were less willing to divert if their alternate route was unfamiliar, unsafe or had several traffic stops. Socioeconomic characteristics were also significant in predicting willingness to divert.     

User-centric interdependent urban systems: Using time-of-day electricity usage data to predict morning roadway congestion

Urban systems are interdependent as individuals’ daily activities engage using those urban systems at certain time of day and locations. There may exist clear spatial and temporal correlations among usage patterns across all urban systems. This paper explores such a correlation among energy usage and roadway congestion. We propose a general framework to predict congestion starting time and congestion duration in the morning using the time-of-day electricity use data from anonymous households with no personally identifiable information. We show that using time-of-day electricity data from midnight to early morning from 322 households in the City of Austin, can make reliable prediction of congestion starting time of several highway segments, at the time as early as 2 am. This predictor significantly outperforms a time-series predictor that uses only real-time travel time data up to 6 am. We found that 8 out of the 10 typical electricity use patterns have statistically significant affects on morning congestion on highways in Austin. Some patterns have negative effects, represented by an early spike of electricity use followed by a drastic drop that could imply early departure from home. Others have positive effects, represented by a late night spike of electricity use possible implying late night activities that can lead to late morning departure from home.     

A dynamic stochastic model for evaluating congestion and crowding effects in transit systems

One of the most common motivations for public transport investments is to reduce congestion and increase capacity. Public transport congestion leads to crowding discomfort, denied boardings and lower service reliability. However, transit assignment models and appraisal methodologies usually do not account for the dynamics of public transport congestion and crowding and thus potentially underestimate the related benefits.This study develops a method to capture the benefits of increased capacity by using a dynamic and stochastic transit assignment model. Using an agent-based public transport simulation model, we dynamically model the evolution of network reliability and on-board crowding. The model is embedded in a comprehensive framework for project appraisal.A case study of a metro extension that partially replaces an overloaded bus network in Stockholm demonstrates that congestion effects may account for a substantial share of the expected benefits. A cost-benefit analysis based on a conventional static model will miss more than a third of the benefits. This suggests that failure to represent dynamic congestion effects may substantially underestimate the benefits of projects, especially if they are primarily intended to increase capacity rather than to reduce travel times.     

Network-wide impacts of eco-routing strategies: A large-scale case study

This paper quantifies the system-wide impacts of implementing a dynamic eco-routing system, considering various levels of market penetration and levels of congestion in downtown Cleveland and Columbus, Ohio, USA. The study concludes that eco-routing systems can reduce network-wide fuel consumption and emission levels in most cases; the fuel savings over the networks range between 3.3% and 9.3% when compared to typical travel time minimization routing strategies. We demonstrate that the fuel savings achieved through eco-routing systems are sensitive to the network configuration and level of market penetration of the eco-routing system. The results also demonstrate that an eco-routing system typically reduces vehicle travel distance but not necessarily travel time. We also demonstrate that the configuration of the transportation network is a significant factor in defining the benefits of eco-routing systems. Specifically, eco-routing systems appear to produce larger fuel savings on grid networks compared to freeway corridor networks. The study also demonstrates that different vehicle types produce similar trends with regard to eco-routing strategies. Finally, the system-wide benefits of eco-routing generally increase with an increase in the level of the market penetration of the system.     

Ridesharing: The state-of-the-art and future directions

Although ridesharing can provide a wealth of benefits, such as reduced travel costs, congestion, and consequently less pollution, there are a number of challenges that have restricted its widespread adoption. In fact, even at a time when improving communication systems provide real-time detailed information that could be used to facilitate ridesharing, the share of work trips that use ridesharing has decreased by almost 10% in the past 30 years.In this paper we present a classification to understand the key aspects of existing ridesharing systems. The objective is to present a framework that can help identify key challenges in the widespread use of ridesharing and thus foster the development of effective formal ridesharing mechanisms that would overcome these challenges and promote massification.     

A day-to-day dynamical model for the evolution of path flows under disequilibrium of traffic networks with fixed demand

Transportation networks are often subjected to perturbed conditions leading to traffic disequilibrium. Under such conditions, the traffic evolution is typically modeled as a dynamical system that captures the aggregated effect of paths-shifts by drivers over time. This paper proposes a day-to-day (DTD) dynamical model that bridges two important gaps in the literature. First, existing DTD models generally consider current path flows and costs, but do not factor the sensitivity of path costs to flow. The proposed DTD model simultaneously captures all three factors in modeling the flow shift by drivers. As a driver can potentially perceive the sensitivity of path costs with the congestion level based on past experience, incorporating this factor can enhance real-world consistency. In addition, it smoothens the time trajectory of path flows, a desirable property for practice where the iterative solution procedure is typically terminated at an arbitrary point due to computational time constraints. Second, the study provides a criterion to classify paths for an origin–destination pair into two subsets under traffic disequilibrium: expensive paths and attractive paths. This facilitates flow shifts from the set of expensive paths to the set of attractive paths, enabling a higher degree of freedom in modeling flow shift compared to that of shifting flows only to the shortest path, which is behaviorally restrictive. In addition, consistent with the real-world driver behavior, it also helps to preclude flow shifts among expensive paths. Improved behavioral consistency can lead to more meaningful path/link time-dependent flow profiles for developing effective dynamic traffic management strategies for practice. The proposed DTD model is formulated as the dynamical system by drawing insights from micro-economic theory. The stability of the model and existence of its stationary point are theoretically proven. Results from computational experiments validate its modeling properties and illustrate its benefits relative to existing DTD dynamical models.     

Do new sources of traffic data make the application of Chaos Theory to traffic management a realistic possibility?

Current signal systems for managing road traffic in many urban areas around the world lack a coordinated approach to detecting the spatial and temporal evolution of congestion across control regions within city networks. This severely inhibits these systems’ ability to detect reliably, on a strategic level, the onset of congestion and implement effective preventative action. As traffic is a time-dependent and non-linear system, Chaos Theory is a prime candidate for application to Urban Traffic Control (UTC) to improve congestion and pollution management. Previous applications have been restricted to relatively uncomplicated motorway and inter-urban networks, arguably where the associated problems of congestion and vehicle emissions are less severe, due to a general unavailability of high-resolution temporal and spatial data that preserve the variability in short-term traffic patterns required for Chaos Theory to work to its full potential. This paper argues that this restriction can now be overcome due to the emergence of new sources of high-resolution data and large data storage capabilities. Consequently, this opens up the real possibility for a new generation of UTC systems that are better able to detect the dynamic states of traffic and therefore more effectively prevent the onset of traffic congestion in urban areas worldwide.