Search

Connected and Autonomous Vehicles (CAVs) are becoming a reality and are progressively penetrating the markets level by level. CAVs are a promising solution for traffic safety. However, robust studies are needed to explore and assess the expected behavior. This study attempts to evaluate traffic safety resulting from a near-real introduction of CAVs with different levels of automation (from Level 1 to Level 4). The investigation consisted of modeling different CAV levels using Gipps’ model, followed by the simulation of nine mixed fleets at a motorway segment. Subsequently, the Surrogate Safety Assessment Model was used for safety analysis. According to the results: (1) the gradual penetration of CAV levels led to a progressive reduction in traffic conflicts, ranging from 18.9% when the penetration of high levels of automation (Level 3 and Level 4 vehicles) is 5%, to 94.1% when all the vehicles on the traffic flow are Level 4; (2) human-driven vehicles and vehicles with low levels of automation (Level 1 and Level 2 vehicles) are more frequently involved in conflicts (as follower vehicles) than vehicles with high automation levels. E.g. human-driven vehicles are involved in conflicts from 8% to 122% more, while vehicles with high automation levels are involved in conflicts from 80% to 18% less than their sharing percentages, respectively, depending on different mixed fleets. This study confirms the theory and conclusions from previous literature that indicate a safety gain due to CAV penetration. Moreover, it provides a broader perspective and support for the introduction of CAVs levels.

The paper proposes a general framework for the assessment of the impacts of the introduction of Connected and Automated Transport Systems (CATS) on traffic. The main objective is to address the question of scalability and transferability of the identified impacts of Autonomous Vehicles (AVs) in particular, focusing on network performance of urban areas. A combination of microscopic and macroscopic simulations as well as statistical methods are applied. Microscopic simulation is conducted to measure the changes in network capacities by utilizing the concept of the Macroscopic Fundamental Diagram (MFD), under different AV penetration rates. The resulting capacities are used to estimate the effects on the Passenger Car Units (PCU) under different AV penetration rates and derive functional relationships, which are further introduced to travel demand models to forecast the macroscopic impacts on network performance. The results indicate a positive impact in relation to capacity changes resulting from the presence of AVs which vary with penetration rate. Analysis of three different urban networks, Barcelona, Bilbao (Spain), and Athens (Greece), reveals consistent trends. However, notable differences are observed on the estimated PCUs for Athens, potentially because of the different mixed-traffic composition. Further exploration of the critical AV modeling specifications and network characteristics is therefore required for deriving transferable PCU functional relationships across networks. Nevertheless, the static assignment results verify the expected trends in network performance impacts in relation to the applied PCU relationships. Finally, the transferability of the proposed methodology across networks is demonstrated.

Congestion pricing has been proved to be an effective traffic management measure to reduce traffic congestion in metropolitan regions. Its aim is to limit the access of traffic towards heavily congested areas, thus to force road users to shift to other means of transportation by imposing toll fares to the vehicles willing to access these areas. Previous urban congestion pricing schemes have been efficiently discouraging individual vehicles from accessing congested areas by reducing traffic and mitigating environmental emissions. This study presents a new methodology for a dynamic zone-based congestion pricing system in networks with heterogeneous users, where the variable fares are calculated based on traffic congestion and environmental emission indicators. Congestion pricing methodology aims to determine optimal toll fares based on traffic congestion, vehicle type, and emissions produced based on the London Emission Model (LEM) to ensure reduction of emissions in the network-wide level. Further, methodology is developed within Dynamic Traffic Assignment (DTA) framework with a microscopic simulation model, that enables behavioral adaptation of the drivers during the pricing period as well as realistic congestion propagation. The performance of the proposed methodology is evaluated against a potential fixed pricing scheme in a Leicester city network, with 9652 sections, 3757 nodes where pricing policy area covers four zones with radius of 1.5 km from the centre. Simulation outputs have shown decrease in traffic delay time and NOxemissions within the cordon area where the congestion pricing policy is applied. However, further strategic traffic management improvements are required to control the mode shift phenomena in the boundary area produced by the congestion policy.

With the increasing availability of traffic data in large urban areas, there is an opportunity to infer more sophisticated traffic patterns and trends which till now has been difficult to obtain and understand. The inferred traffic patterns can then serve as input into short-term traffic prediction models or to predict traffic demand in the network for deployment of traffic management policies. However, the network-wide consistency, the interaction between variables and the difficulty during forecasting in choosing a pattern at early AM peak hour require robust, but flexible enough pattern extraction method. A wide range of researchers tackle some of these points, but it is difficult to find the one that fulfills all of them.

Traffic flow prediction is a fundamental problem for efficient transportation control and management. However, most current data-driven traffic prediction work found in the literature have focused on predicting traffic from an individual task perspective, and have not fully leveraged the implicit knowledge present in a road-network through space and time correlations. Such correlations are now far easier to isolate due to the recent profusion of traffic data sources and more specifically their wide geographic spread.

Traffic management is being more important than ever, especially in overcrowded big cities with over-pollution problems and with new unprecedented mobility changes. In this scenario, road-traffic prediction plays a key role within Intelligent Transportation Systems, allowing traffic managers to be able to anticipate and take the proper decisions. This paper aims to analyze the situation in a commercial real-time prediction system with its current problems and limitations. We analyze issues related to the use of spatiotemporal information to reconstruct the traffic state. The analysis unveils the trade-off between simple parsimonious models and more complex models. Finally, we propose an enriched machine learning framework, Adarules,for the traffic state prediction in real-time facing the problem as continuously incoming data streams with all the commonly occurring problems in such volatile scenario, namely changes in the network infrastructure and demand, new detection stations or failure ones, among others. The framework is also able to infer automatically the most relevant features to our end-task, including the relationships within the road network, which we call as “structure learning”. Although the intention with the proposed framework is to evolve and grow with new incoming big data, however there is no limitation in starting to use it without any prior knowledge as it can starts learning the structure and parameters automatically from data.

Passing zones are provided to improve operational efficiency of two-lane highways. Minimum passing zone lengths of 120 m were established by FHWA and AASHTO. Some studies indicate that lengths may need to be increased, but no changes have been recommended, pending further research. The objective of this study was to develop design and marking criteria for minimum passing zone lengths, with traffic operational efficiency and safety taken into consideration. First, a traffic microsimulation was conducted with AIMSUN software. The calibration and validation included the observation of 1,750 passing maneuvers in Spain. Results indicated that passing zones shorter than 250 m added little to operational efficiency. Second, a reliability analysis was applied. The analysis quantified the probability that a passing maneuver was completed beyond the end of the passing zone (noncompliant passing maneuver). Then the number of noncompliant passing maneuvers was calculated. Traffic flow and passing zone length were contributing factors. Findings from the analysis indicated that the minimum passing zone length should be increased to a minimum of 275 m for high traffic volumes, 300 m for medium traffic volumes, and 350 m for low traffic volumes. With these lengths, noncompliant passing maneuvers decrease. The marginal increase in the minimum length of passing zones can potentially improve safety without significantly reducing operational efficiency. The results can be used by practitioners to establish minimum passing zone length on the basis of hourly volumes and level of risk.

Passing zones are provided to improve operational efficiency of two-lane highways. Minimum passing zone lengths of 120 m were established by FHWA and AASHTO. Some studies indicate that lengths may need to be increased, but no changes have been recommended, pending further research. The objective of this study was to develop design and marking criteria for minimum passing zone lengths, with traffic operational efficiency and safety taken into consideration. First, a traffic microsimulation was conducted with AIMSUN software. The calibration and validation included the observation of 1,750 passing maneuvers in Spain. Results indicated that passing zones shorter than 250 m added little to operational efficiency. Second, a reliability analysis was applied. The analysis quantified the probability that a passing maneuver was completed beyond the end of the passing zone (noncompliant passing maneuver). Then the number of noncompliant passing maneuvers was calculated. Traffic flow and passing zone length were contributing factors. Findings from the analysis indicated that the minimum passing zone length should be increased to a minimum of 275 m for high traffic volumes, 300 m for medium traffic volumes, and 350 m for low traffic volumes. With these lengths, noncompliant passing maneuvers decrease. The marginal increase in the minimum length of passing zones can potentially improve safety without significantly reducing operational efficiency. The results can be used by practitioners to establish minimum passing zone length on the basis of hourly volumes and level of risk.

One of the essential quantities in macro traffic simulation is the ‘who goes where’ OD matrix. This is usually estimated through an optimization problem of minimizing the squared differences between observed and simulated traffic counts. The choice of the mean square error loss is due to its good statistical properties and the fact that the related quadratic optimization problem is easy to deal with. However, in many applications following the recommendations of validation criteria suggested in many guidelines published by different road administrations, such as FHWA in USA, ARRB in Australia or Highways Agency in UK, the quality of an static OD adjustment is assessed with other quantities such as the relative error or the GEH index. In this paper we estimate OD matrices minimizing directly the mean geh or mean relative error.

This study adopted a metamodel-based technique for model sensitivity analysis and applied it to the AIMSUN mesoscopic model. The application of sensitivity analysis is crucial for the true comprehension and correct use of the traffic simulation model, although the main obstacle to an extensive use of the most sophisticated techniques is the high number of model runs such techniques usually require. For this reason, the possibility of performing a sensitivity analysis was tested not on a model but on its metamodel approximation. Important issues concerning metamodel estimation were investigated and commented on in the specific application to the AIMSUN model. Among these issues are the importance of selecting a proper sampling strategy based on low-discrepancy random number sequences and the importance of selecting a class of metamodels able to reproduce the inputs-outputs relationship in a robust and reliable way. Sobol sequences and Gaussian process metamodels were recognized as the appropriate choices. The proposed methodology was assessed by comparing the results of the application of variance-based sensitivity analysis techniques with the simulation model and with a metamodel estimated with 512 model runs for a variety of traffic scenarios and model outputs. Results confirmed the power of the proposed methodology and also made a more extensive application of sensitivity analysis techniques available for complex traffic simulation models.

This study adopted a metamodel-based technique for model sensitivity analysis and applied it to the AIMSUN mesoscopic model. The application of sensitivity analysis is crucial for the true comprehension and correct use of the traffic simulation model, although the main obstacle to an extensive use of the most sophisticated techniques is the high number of model runs such techniques usually require. For this reason, the possibility of performing a sensitivity analysis was tested not on a model but on its metamodel approximation. Important issues concerning metamodel estimation were investigated and commented on in the specific application to the AIMSUN model. Among these issues are the importance of selecting a proper sampling strategy based on low-discrepancy random number sequences and the importance of selecting a class of metamodels able to reproduce the inputs–outputs relationship in a robust and reliable way. Sobol sequences and Gaussian process metamodels were recognized as the appropriate choices. The proposed methodology was assessed by comparing the results of the application of variance-based sensitivity analysis techniques with the simulation model and with a metamodel estimated with 512 model runs for a variety of traffic scenarios and model outputs. Results confirmed the power of the proposed methodology and also made a more extensive application of sensitivity analysis techniques available for complex traffic simulation models.

The evaluation of ATMS requires a network-wide assessment of their impact as opposed to an isolated analysis of key intersections. Microscopic traffic simulation models emulate the dynamics of individual vehicles in a detailed network representation, based on car-following, lane changing, etc; include route choice models to implement dynamic traffic assignment. Microscopic models are usually appropriate for operational analysis due the detail of information provided by the simulator. However, they are data intensive and have a significant computational cost. Mesoscopic models combine simplified flow dynamics with explicit treatment of interrupted flows at intersections and allow modeling of large networks with high computational efficiency. However, the loss of realism implied by a mesoscopic model makes it necessary to emulate detailed outputs (detector measurements, instantaneous emissions). The above give rise to the need to combine meso and micro approaches into new hybrid traffic simulators where very large-scale networks are modeled mesoscopically and areas of complex interactions benefit from the finer detail of microscopic simulation.This paper presents the new system architecture implemented in Aimsun of the hybrid model with special emphasis on the consistency of the transition between the mesoscopic and the microscopic models, and vice versa and aims to demonstrate how consistency is complemented using computational results in a large-size network.

Driving simulation has been used for many years now in research, industry and training to simulate virtual driving conditions. This is largely because simulation saves engineering time and costs, and virtual situations provide an easier and safer way to carry out road and traffic safety studies. Scenarios or scenes that realistically reproduce traffic behaviour according to the traffic flow theory has been developed since the 1980s and traffic simulators are a widely used technology in the research, planning, evaluation and development of new traffic systems (for instance Intelligent Traffic Systems). The aim of this paper is to explain how providing a simulation framework based on the integration of a driving simulation engine and a traffic simulation can respond to the issues addressed by offering multi-level and multi-domain simulation models. This paper describes the integration of the driving simulator SCANER and the traffic simulator Aimsun, where the driving simulator (including visual, audio and kinaesthetic restitution) manages the simulation in the immediate driver environment and the traffic simulator manages the whole road network situation, creating a large-scale, realistic and minutely detailed virtual world.

The deployment of intelligent transportation systems must be assisted by suitable tools to conduct the feasibility studies required to test the designs and evaluate the expected impacts. Microscopic traffic simulation has proved to be the suitable methodological approach to achieve these goals. One of the most critical aspects of the dynamic simulation of road networks based on a microscopic approach–-namely, how to perform a heuristic dynamic assignment, the implied route choice models, and whether under certain criteria the approach can achieve stochastic user equilibrium–-is discussed.

The deployment of intelligent transportation systems must be assisted by suitable tools to conduct the feasibility studies required to test the designs and evaluate the expected impacts. Microscopic traffic simulation has proved to be the suitable methodological approach to achieve these goals. One of the most critical aspects of the dynamic simulation of road networks based on a microscopic approach—namely, how to perform a heuristic dynamic assignment, the implied route choice models, and whether under certain criteria the approach can achieve stochastic user equilibrium—is discussed.