Modern roundabouts are increasingly being considered as a preferred intersection design within urban street networks in North America due, in large part, to the increased safety provided by roundabout. In the last decade, much research has been conducted in North America to quantify the performance of roundabouts in terms of vehicle delay as a function of vehicle traffic volume and roundabout geometry. In most jurisdictions, vehicles entering and existing the roundabout are required by law to yield right-of-way to pedestrians crossing the roundabout approach, and consequently, the presence of substantial pedestrian volumes are expected to degrade operational performance of the roundabout for vehicles. However, very little research has been conducted to estimate the impact that pedestrian volumes have on average vehicle delay. The aim of this research is to study the effects of pedestrian volume, entry traffic volume, and conflicting or circulating volume, on the delays that vehicles experience when traversing the roundabout. An analytical model is proposed to estimate vehicle delays on the basis of traffic flow and queuing theory. The model is applicable to single lane roundabouts. The model was calibrated and validated using vehicle delays obtained from the micro-simulation software, VISSIM (version 5.2) for a range of different conditions. The research described in this thesis demonstrates that pedestrian flows cause delays to vehicles traversing the roundabout in four distinct ways. Existing analytical techniques included within most design manuals consider only one of these sources of delays and consequently, conventional models typically under-estimate the impact that pedestrian flows have in terms of increasing delays to vehicles traversing the roundabout.

Using microsimulation software, with a focus on VISSIM, the analysis of different applications of merge concepts through delay and safety is presented in this thesis. In order to appropriately draw conclusions and usage trends of different merge concepts from the microsimulation software, early merge, late merge, and signal merge were first explored in a thorough literature review. While focusing primarily on delay, queues, and safety, this thesis essentially provides an introduction to determining the ideal merge concept on freeway work zones for varying roadway configurations, roadway conditions, and user demands, among other factors. In addition to delay and queuing analysis completed using VISSIM, the Federal Highway Administration’s Surrogate Safety Assessment Model (SSAM) was used to address the effects of implementing signal merge on rear-end and lane-change conflicts. Compiling the VISSIM microsimulation outputs and SSAM signal merge safety outputs, general conclusions and decisions were provided. While this thesis provides determinations of ideal merge concepts for a variety of cases, it is important for the next researcher to assess some of the assumptions that were made, to ensure that they would not significantly affect the results and analysis. / text

Road safety and traffic congestion are two of the critical issues facing the transportation profession today. As a means to promote safety and efficiency at United States intersections modern roundabouts are becoming more and more common. Over the last ten years, roundabouts implementation methodologies have been developed using data collected at U.S. roundabouts. These methodologies were first published in National Cooperative Highway Report 572: Roundabouts in the United States and more recently in the second edition of the national roundabout guidelines. This work attempts to validate the use of these methodologies for roundabouts in the state of Texas and also enhance guidelines for evaluating roundabout operations by exploring the effects of exiting flow, origin-destination patterns, and mean speed on roundabout entry-lane capacity. Capacity results from VISSIM are compared to the Highway Capacity Manual entry-lane capacity curve and results from SIDRA. / text

Modern roundabouts are increasingly being considered as a preferred intersection design within urban street networks in North America due, in large part, to the increased safety provided by roundabout. In the last decade, much research has been conducted in North America to quantify the performance of roundabouts in terms of vehicle delay as a function of vehicle traffic volume and roundabout geometry. In most jurisdictions, vehicles entering and existing the roundabout are required by law to yield right-of-way to pedestrians crossing the roundabout approach, and consequently, the presence of substantial pedestrian volumes are expected to degrade operational performance of the roundabout for vehicles. However, very little research has been conducted to estimate the impact that pedestrian volumes have on average vehicle delay. The aim of this research is to study the effects of pedestrian volume, entry traffic volume, and conflicting or circulating volume, on the delays that vehicles experience when traversing the roundabout. An analytical model is proposed to estimate vehicle delays on the basis of traffic flow and queuing theory. The model is applicable to single lane roundabouts. The model was calibrated and validated using vehicle delays obtained from the micro-simulation software, VISSIM (version 5.2) for a range of different conditions. The research described in this thesis demonstrates that pedestrian flows cause delays to vehicles traversing the roundabout in four distinct ways. Existing analytical techniques included within most design manuals consider only one of these sources of delays and consequently, conventional models typically under-estimate the impact that pedestrian flows have in terms of increasing delays to vehicles traversing the roundabout.

As Intelligent Transportation Systems (ITS) become more prevalent, there is a need for a system capable of the rigorous evaluation of new ITS strategies for a wide variety of applications. Pre-deployment testing and fine-tuning of the system, performance evaluation, and alternatives analysis are all potential benefits that could be gained through the evaluation of ITS. Simulation, an increasingly popular tool for transportation analysis, would seem an ideal solution to this problem as it allows for the consideration of many scenarios that may be improbable or impossible to observe in the field. Also, simulation provides a framework that allows for the application of rigorous analysis techniques to the output data, providing an accurate and statistically significant conclusion. The difficulty is that many ITS strategies are difficult or impossible to implement in a simulated environment. The rapid nature of technology development and the complicated nature of many ITS solutions are difficult to emulate in simulation models. Furthermore, the emulation of a particular ITS solution is not guaranteed to provide the same result that the physical system would, were it subject to the same inputs. This study seeks to establish a framework for the analysis of advanced ITS applications through the use of Hardware-in-the-Loop Simulation (HILS), which provides a procedure for interfacing simulation models with real-world hardware to conduct analysis. This solution provides the benefits of both advanced ITS evaluation and simulation for powerful and accurate analysis. A framework is established that includes all the steps of the modeling process including construction, validation, calibration, and output analysis. This ensures that the process surrounding the HILS implementation is valid so that the results of the evaluation are accurate and defendable. Finally, a case study of the application of the developed framework to the evaluation, a real-world implementation of an advanced ITS application (SCATS in this case) is considered. The effectiveness of the framework in creating and evaluating a corridor using a simulation model wed to real-world hardware is shown. The results of the analysis show the power of this method when correctly applied and demonstrate where further analysis could expand upon the proposed procedure.

A malha rodoviária brasileira é constituída predominantemente de rodovias de pista simples e boa parte delas encontra-se em regiões montanhosas. Rodovias de pista simples em regiões montanhosas possuem dificuldades de modelagem, porque as rampas elevadas e os raios de curvatura pequenos possuem impacto significativo no desempenho do tráfego. Adicionalmente, existem poucos estudos de simulação de tráfego de veículos em rodovias de pista simples, sobretudo em regiões montanhosas. Sendo assim, a combinação de rodovias de pista simples com região montanhosa é ainda mais restrita para codificação dessas redes em microssimuladores de tráfego. Dados de geometria de rodovias são comumente obtidos por meio de plataformas como o Google Earth, contudo sua precisão pode ser insuficiente para uma boa representação em regiões montanhosas, devido à influência significativa dos perfis planialtimétricos no desempenho do tráfego em rodovias de pistas simples. O presente trabalho desenvolveu uma metodologia para a codificação de rodovias de pista simples em região montanhosa. Esta metodologia foi aplicada em um estudo de simulação de tráfego através do microssimulador de tráfego VISSIM. O trecho de estudo de caso corresponde à 16 km de extensão da ERS-115 entre as cidades de Três Coroas e Gramado na Serra Gaúcha. A partir de dois bancos de dados (Google Earth e levantamento veicular a Laser), foram desenvolvidos três modelos: (i) modelo baseado nos dados originais do Google (Google original), (ii) modelo com perfil altimétrico suavizado (LOESS) e (iii) modelo com perfil planialtimétrico baseado em levantamento de campo a Laser (Laser). A comparação do perfil altimétrico dos três modelos revelou diferenças significativas entre o modelo do Google original em relação aos demais. O perfil altimétrico decorrente dos dados originais do Google Earth apresentou grandes declividades, na prática, incompatíveis com a classe da rodovia. O método proposto de suavização (LOESS) resultou em um modelo bem mais próximo do perfil real da rodovia, representado pelo modelo construído a partir do levantamento a laser. Com o objetivo de avaliar o desempenho através da simulação de tráfego nos 3 modelos foram executados 11 cenários de simulação. Os cenários envolveram 2 níveis de demanda: demanda baixa, compatível com condições de fluxo livre e uma demanda representativa das condições típicas de tráfego verificadas no trecho. Em relação à demanda em fluxo livre, veículos nos modelos Laser e LOESS apresentaram velocidades similares ao longo do estaqueamento, enquanto que o modelo Google original revelou diferenças significativas de velocidades. A análise do desempenho do tráfego nos cenários de demanda típica foi baseada nos tempos de viagem. Cenários nesta análise envolveram a segmentação da rodovia em trechos homogêneos e variações na proporção de veículos por sentido (split direcional). A comparação entre os tempos de viagem dos modelos LOESS e Laser resultou em R²=0,99, enquanto que R² obtidos entre os modelos Google original e Laser variaram no intervalo de 0,04 a 0,99. Os resultados obtidos apontam a importância da metodologia desenvolvida pelo presente trabalho na modelagem de rodovias de pista simples, sobretudo em região montanhosa. / The Brazilian road network consists predominantly of two-lane highways and great part of them are in mountainous regions. Two-lane highways in mountainous regions have modeling difficulties because high slopes and small bend radii have a significant impact on traffic performance. Additionally, there are few simulation studies of vehicle traffic on two-lane highways, especially in mountainous regions. Thus, the combination of two-lane highways and mountainous region is even more restricted for coding these networks into traffic microsimulators. Road geometry data are commonly obtained through platforms such as Google Earth, but their accuracy may be insufficient for good representation in mountainous regions due to significant influence of planialtimetric profiles on traffic performance on two-lane highways. The present work developed a methodology for the codification of two-lane highways in mountainous region. This methodology was applied in a simulation study of traffic through the VISSIM traffic microsymulator. The case study section corresponds to the 16 km extension of the ERS-115 between the cities of Três Coroas and Gramado in the Serra Gaúcha. From two databases (Google Earth and Laser Vehicle Survey), three models were developed: (i) model based on original Google data (Google original), (ii) model with smoothed altimetric profile (LOESS) and (iii) model with planialtimetric profile based on laser field survey (Laser). The comparison of altimetric profiles of the three models revealed significant differences between the Google original model in relation to the others. The altimetric profile derived from the original data of Google Earth presented great slopes, in practice, incompatible with the highway class. The proposed smoothing method (LOESS) resulted in a model that is closer to the actual road profile, represented by the model constructed from the laser survey. In order to evaluate the performance through the simulation of traffic in the three models, 11 simulation scenarios were executed. The scenarios involved two levels of demand: low demand, compatible with free flow conditions and a representative demand of the typical traffic conditions verified in the road segment. Regarding the free-flow demand, vehicles in Laser and LOESS models presented similar speeds along the road, while the Google original model revealed significant differences in speeds. The analysis of traffic performance in typical demand scenarios was based on travel times. Scenarios in this analysis involved the segmentation of the highway in homogeneous segments and variations in the proportion of vehicles by direction (directional split). The comparison of travel times of the LOESS and Laser models resulted in R² = 0.99, while R² obtained between the Google original and Laser models varied in the range of 0.04 to 0.99. The results obtained point out the importance of the methodology developed by the present work in the modeling of two-lane highways, especially in the mountainous region.

En el presente trabajo de investigación, analizamos la calibración simultánea de vehículos y peatones que interactúan en determinados puntos de conflicto del sistema vial. Para esto, trabajamos en un caso de una avenida regional, donde hay una gran demanda del parque automotor tanto de vehículos ligeros, de transporte público y provincial, así como el de pesados, en la que los peatones prefieren el paso a nivel en relación a un puente peatonal existente en la zona. La información de campo requerida se recopiló para el análisis, como aforo de vehículos, cantidad de peatones que hacen uso o no del puente, tiempos de viaje y distancias. Posteriormente, la información se procesó en el software Vissim 9, en el que se modeló el estudio de caso y se realizó la respectiva calibración simultánea del modelo para su respectiva validación. Para esto, fue necesario iterar los parámetros como el Wiedemann 74 (vehículos), Tau y Lambda (peatones) hasta obtener la representación más cercana a los datos tomados en el campo. Finalmente, los resultados se presentan mediante la validación de la simulación en términos de número de simulaciones, número de carreras, tiempo de viaje y tasa de uso de la carretera, todo ello para que pueda servir de base a futuras simulaciones que reúnan las características de la zona en estudio. / In this research, we analyze the simultaneous calibration of vehicles and pedestrians that interact at certain points of conflict in the road system. For this, we work in a case of a regional avenue, where there is a great demand for the automotive fleet of both light vehicles, public and provincial transport, as well as heavy vehicles, in which pedestrians prefer the level crossing in relation to an existing pedestrian bridge in the area. The required field information was collected for the analysis, such as capacity of vehicles, number of pedestrians who use the bridge or not, travel times and distances. Subsequently, the information was processed in the Vissim 9 software, in which the case study was modeled and the respective simultaneous calibration of the model was performed for its respective validation. For this, it was necessary to iterate the parameters such as the Wiedemann 74 (vehicles), Tau and Lambda (pedestrians) until obtaining the closest representation to the data taken in the field. Finally, the results are presented through the validation of the simulation in terms of number of simulations, number of races, travel time and road use rate, all so that it can serve as a basis for future simulations that meet the characteristics of the area under study. / Trabajo de investigación

El texto completo de este trabajo no está disponible en el Repositorio Académico UPC por restricciones de la casa editorial donde ha sido publicado. / The fully actuated signal control detects the pedestrian density using sensors and, according to that, it prioritizes pedestrians crossing. One major problem, worldwide, is using fixed time traffic light as a traffic regulator at intersections with high pedestrian and vehicular volume. Lima is no exception, continuing to use this kind of traffic lights completely harms pedestrian accessibility, it increases their waiting and crossing times, it also affects road safety and service levels at the structures. The proposal on this article is to design a fully actuated signal control using logical controls that are able to perceive the pedestrian density on the refuge islands, making everything more accessible. In order to do this, a study to identify the pedestrian and vehicle volume was conducted on the Lima Panamerican highway. There was a total of 7506 pedestrians during rush hour, proving there is a large amount of people at the intersection at that time. Thereby, by using the VisVap module of the Vissim, the study managed to simulate and validate the priority control required. All in all, the results showed a remarkable improvement, the pedestrian crossing time was reduced by 6.84% and the service level of the intersection went from E to D.

The evacuation planning has become an important issue addressed by many research studies and publications aiming to improve the security of the daily life for our public inside the United States of America. The main objective of this research was to address the growing need for evacuation planning using traffic simulation. With increased interests and awareness in emergency evacuation and first responder access to emergencies in public locations (airports, transit stations, ports or stadiums), the traffic simulation can be helpful in orchestrating the traffic flow during emergencies. Related to this issue, Federal Transit Administration has issued a large number of publications and guidelines concerning emergency preparedness and incident management. These guidelines are used to develop a simulation-based activity to evaluate the current plan and alternative plans for the deployment of transit during an emergency situation. A major task for this project is to study the effect of evacuation on the surrounding traffic network and help the local transit company (LYNX) to evaluate their evacuation plan and consider different possibilities without the risk and cost of actual evacuation drills. A set of different scenarios and alternatives for each scenario were simulated and studied to reach the best possible evacuation strategy. The main findings were evacuation as pedestrians have less impact on traffic network and rerouting decreases the congestion resulting from the evacuation process.

The microscopic traffic simulation program VISSIM is a powerful tool that has been used by transportation engineers and urban planners around the world. A VISSIM simulation is meant to depict the performance of the physical road network through the use of modeling tools and behavioral parameters. The process which gets the model to the point of matching real world conditions is called calibration and requires a means of relating the real world to the simulated world. The topic of this thesis discusses a new means of calibration using the two-fluid model. The two-fluid model is a macroscopic modeling technique which provides quantitative characteristics of the performance of traffic flow on an urban road network. The model does this by generating a relationship between the travel time, stopped time, and running time per mile. The two-fluid model has been used to evaluate the performance of road networks for decades but now it is possible to use it to calibrate a VISSIM model. For this thesis, the two-fluid model to be used for calibration was generated from data collected on the Orlando, Florida, downtown network in February, 2008, during three traffic peaks for three typical weekdays. The network was then modeled in VISSIM which required a large amount of data regarding network geometry, signal timings, signal coordination schemes, and turning movement volumes. A similar data collection exercise was conducted during November, 2008, to capture the effects of changes that took place in the network during the ten month period. Another VISSIM network was also made to match the conditions of the November network. The February field data was used to successfully calibrate the VISSIM model and the November data was used to validate the calibrated network. The validation proved that the two-fluid models from the November field data and VISSIM data are statistically similar. With the network calibrated and validated, it could be used to perform scenario tests to see how the network performance would be affected by changes to the network. The two-fluid model has often been used to compare two different physical networks or explore how the performance of a single physical network has changed over time. A similar comparison can be done with the two-fluid models from a calibrated, simulated network. By using the original calibrated models as base cases, scenarios ranging from lane closures due to traffic incidents to the addition of a whole new signalized corridor on the network can be modeled in VISSIM and compared with the corresponding base case. This would allow a governing agency to preview the effects of proposed changes.