No description available.

One of the main difficulties in the development of urban traffic microsimulation models is the collection of traffic data for calibration and validation. However, the city of Fortaleza has an important mesosimulation tool that, in addition to controlling urban traffic in real time, estimates traffic variables: the well-known SCOOT system. This system, implemented in cities around the world, controls and estimates traffic in the densest urban area of Fortaleza, based on the continuous detection of vehicle occupation on its more than 900 detectors spread throughout the city. However, because these data are simulated, they require validation before being used. The main aim of this work was to develop and implement a methodology to validate the mesoscopic simulation model of SCOOT, so its data can be used in the development of traffic microsimulation models, having as a case-study the system operating in Fortaleza. Based on experiments, the effects of two factors in the estimation error were investigated: the calibration of the parameter SATO, and the average travel time between the loop detector and the stop-bar. The results show that these two factors affect the quality of the prediction of volume, delay and number of vehicle-stops. These results contribute with a validation methodology that allows a better use of the data provided by SCOOT. / Um das maiores dificuldades na construÃÃo de redes viÃrias urbanas microssimuladas reside na coleta dos dados de trÃfego para as fases de calibraÃÃo e validaÃÃo. PorÃm, a cidade de Fortaleza dispÃe de uma importante ferramenta de mesossimulaÃÃo que, alÃm de controlar o trÃfego urbano em tempo real, estima indicadores de trÃfego: sistema SCOOT â Split Cycle Offset Optmisation Technique. Este sistema, implantado em vÃrias cidades do mundo, controla e modela o trÃfego na regiÃo mais adensada da Ãrea urbana de Fortaleza, baseando-se na coleta contÃnua de ocupaÃÃo veicular sobre os seus mais de 900 laÃos detectores espalhados pela cidade. No entanto, por se tratar de valores simulados, carecem de verificaÃÃo antes de serem utilizados. O objetivo geral deste trabalho à desenvolver e implementar uma metodologia para validaÃÃo do modelo de simulaÃÃo mesoscÃpica do SCOOT, tendo em vista o uso de seus dados no desenvolvimento de modelos de microssimulaÃÃo do trÃfego, tendo como estudo de caso o sistema em operaÃÃo em Fortaleza. Por meio de experimentos, foram investigados os efeitos de dois fatores no erro de estimaÃÃo: a calibraÃÃo do parÃmetro SATO e o tempo de percurso mÃdio entre o laÃo detector e a faixa de retenÃÃo. Os resultados mostram que estes dois fatores afetam a qualidade da modelagem das variÃveis volume, atraso veicular e nÃmero de paradas. Os resultados desta pesquisa contribuem no sentido de oferecer uma metodologia de validaÃÃo que permita um melhor uso dos dados fornecidos pelo SCOOT.

Traffic signal control systems have been tremendously improved since the first colored traffic signal light was installed in London in December 1868. There are many different types of traffic signal control systems that can be categorized into three major control types: fixed-time, actuated, and adaptive. Choosing a proper traffic signal system is very important since there exists no perfect signal control strategy that fits every traffic network. One example is traffic signal coordination, which is the most widely used traffic signal control system. It is believed that performance measures, such as travel times, vehicle delay, and number of stops, can be enhanced by synchronizing traffic signals over a corridor. However, it is not always true that the coordination will have the same benefits for all the traffic in the network. Most of the research on coordination has focused only on strengthening the major movement along the coordinated routes without considering system-wide impacts on other traffic. Therefore, before implementing a signal control system to a specific traffic network, a thorough investigation should be conducted to see how the control strategy may impact the entire network in terms of the objectives of each type of traffic control system. This dissertation first considers two different kinds of systematic performance analyses for traffic signal control systems. Then, it presents two types of signal control strategies that account for current issues in coordination and priority control systems, respectively. First, quantitative analysis of smooth progression for traffic flow is investigated using connected vehicle technology. Many studies have been conducted to measure the quality of progression, but none has directly considered smooth progression as the significant factor of coordination, despite the fact that the definition of coordination states that the goal is to have smooth traffic flow. None of the existing studies concentrated on measuring a continuous smooth driving pattern for each vehicle in terms of speed. In order to quantify the smoothness, this dissertation conducts an analysis of the speed variation of vehicles traveling along a corridor. A new measure is introduced and evaluated for different kinds of traffic control systems. The measure can be used to evaluate how smoothly vehicles flow along a corridor based on the frequency content of vehicle speed. To better understand the impact of vehicle mode, a multi-modal analysis is conducted using the new measure. Second, a multi-modal system-wide evaluation of traffic signal systems is conducted. This analysis is performed for traffic signal coordination, which is compared with fully actuated control in terms of a systematic assessment. Many optimization models for coordination focus mainly on the objective of the coordinated route and do not account for the impacts on side street movements or other system-wide impacts. In addition, multi-modality is not considered in most optimized coordination plans. Thus, a systematic investigation of traffic signal coordination is conducted to analyze the benefits and impacts on the entire system. The vehicle time spent in the system is measured as the basis of the analysis. The first analysis evaluates the effect of coordination on each route based on a single vehicle mode (regular passenger vehicles). The second analysis reveals that how multi-modality affects the performance of the entire system. Third, in order to address traffic demand fluctuation and traffic pattern changes during coordination periods, this dissertation presents an adaptive optimization algorithm that integrates coordination with adaptive signal control using data from connected vehicles. Through the algorithm, the coordination plan can be updated to accommodate the traffic demand variation and remain optimal over the coordination period. The optimization framework consists of two levels: intersection and corridor. The intersection level handles phase allocation in real time based on connected vehicle trajectory data, while the corridor level deals with the offsets optimization. The corridor level optimization focuses on the performance of the vehicle movement along the coordinated phase, while at the intersection level, all movements are considered to create the optimal signal plan. The two levels of optimizations apply different objective functions and modeling methodologies. The objective function at the intersection level is to minimize individual vehicle delay for both coordinated and non-coordinated phases using dynamic programming (DP). At the corridor level, a mixed integer linear programming (MILP) is formulated to minimize platoon delay for the coordinated phase. Lastly, a peer priority control strategy, which is a methodology that enhances the multi modal intelligent traffic signal system (MMITSS) priority control model, is presented based on peer-to-peer (P2P) and dedicated short range communication (DSRC) in a connected vehicle environment. The peer priority control strategy makes it possible for a signal controller to have a flexible long-term plan for prioritized vehicles. They can benefit from the long-term plan within a secured flexible region and it can prevent the near-term priority actions from having a negative impact on other traffic by providing more flexibility for phase actuation. The strategy can be applied to all different modes of vehicles such as transit, freight, and emergency vehicles. Consideration for far side bus stops is included for transit vehicles. The research that is presented in this dissertation is constructed based on Standard DSRC messages from connected vehicles such as Basic Safety Messages (BSMs), Signal Phasing and Timing Messages (SPaTs), Signal Request Messages (SRMs), and MAP Messages, defined by Society of Automotive Engineers (SAE) (SAE International 2016).

Cities worldwide are faced with the challenge of improving transit service in urban areas using lower cost means. Transit signal priority is considered to be one of the most effective ways to improve the service of transit vehicles. Transit signal priority has become a very popular topic in transportation in the past 20 to 30 years and it has been implemented in many places around the world. In this thesis, transit signal priority strategies are categorized and an extensive literature review on past research on transit signal priority is conducted. Then a case study on Columbia Pike in Arlington (including 21 signalized intersections) is conducted to assess the impacts of integrating transit signal priority and SCOOT adaptive signal control. At the end of this thesis, an isolated intersection is designed to analyze the sensitivity of major parameters on performance of the network and transit vehicles. The results of this study indicate that the prioritized vehicles usually benefit from any priority scheme considered. During the peak period, the simulations clearly indicate that these benefits are typically obtained at the expense of the general traffic. While buses experience reductions in delay, stops, fuel consumption, and emissions, the opposite typically occurs for the general traffic. Furthermore, since usually there are significantly more cars than buses, the negative impacts experienced by the general traffic during this period outweigh in most cases the benefits to the transit vehicles, thus yielding overall negative impacts for the various priority schemes considered. For the off-peak period, there are no apparent negative impacts, as there is more spare capacity to accommodate approaching transit vehicles at signalized intersections without significantly disrupting traffic operations. It is also shown in this study that it is generally difficult to improve the system-wide performance by using transit priority when the signal is already optimized according to generally accepted traffic flow criteria. In this study it is also observed that the system-wide performance decreases rapidly when transit dwell time gets longer. / Master of Science