Toll plaza operation is a critical component of roadway operations throughout the United States, as tolls provide both revenue for expansion and opportunity for demand management. Originally cash or physical currency based, tolling has morphed to meet the twentieth century demand in terms of throughput and efficiency in the form of electronic toll collection. Electronic tolling has introduced a new form of driver decision making at toll plazas due to the additional payment choice. Despite the user convenience these facilities provide to consumers, this form of collection has not come without safety and operational concerns. Confusion at the toll plaza, unsafe merging maneuvers, and the unexpected behavior has actually increased certain crash patterns at toll plazas in some electronic tolling facilities. Building upon existing research, further work was completed to quantify the related impacts of electronic toll collection on traffic operations through a microsimulation model, and static evaluation study. While in Massachusetts overall toll plaza crashes are a minimal portion of 200,000 crashes each year in the Commonwealth at less than 0.1 percent of all crashes some toll plazas have higher crash rates than the state wide urban interstate average. Interchange 14 in Weston, Massachusetts had the highest crash rate among state toll plazas. Rear-end and same direction sideswipe collisions accounted for the highest crash numbers between the years 2010 and 2012. Microsimulation of various lane configurations derived from static evaluation feedback on driver decision making created six alternate configurations. Current plaza configuration was verified by the validated VISSIM microsimulation model to be the highest performing in terms of efficiency. A lane configuration with grouped payment lanes provided the best overall performance for alternatives with less than 1 percent difference from the current West Springfield interchange configuration. Static evaluation and microsimulation results pointed to increased efficiency and safety benefits with combination lanes. Additionally, drivers tended to avoid following heavy vehicles through plaza lanes. Motorists were willing to make up to 3 lane changes to avoid queues and may avoid combination lanes as an electronic toll customer if they anticipate a greater delay than an adjacent dedicated electronic lane. Recommendations for future research include: 1) further microsimulation modeling to examine traffic flow and safety impacts at toll plazas under varying traffic conditions and demand with open road tolling lanes strategies; and 2) developing enhancements to VISSIM to address parameter limitations associated with discrete choice modeling at toll plazas.

A great deal of research has been conducted on Central Florida toll roads to better understand the characteristics of the tolling operation. In this thesis, the development and calibration of a toll plaza simulation models will be analyzed using two simulation programs varying mostly in their modeling theory. The two models utilized are, SHAKER, a deterministic queuing model for vehicles utilizing toll collection facilities, and VISSIM, a globally popular stochastic simulation software. The benefits of simulation models leads to the purpose of this thesis, which is to examine the effectiveness of two toll modeling programs that are similar in purpose but vary in approach and methodology. Both SHAKER and VISSIM toll plaza models have the potential to work as a tool that can estimate the maximum throughput and capacity of toll plazas. Major operational benefits resulting from developing these models are to simulate and evaluate how traffic conditions will change when demand increases, when and if queues increase when a lane is closed due to maintenance or construction, the impact of constructing additional lanes, or determining whether or not the best lane type configuration is currently implemented. To effectively calibrate any model available site data must be used to compare simulation results to for model validity. In an effort to correctly calibrate the SHAKER toll plaza tool and VISSIM model, an extensive field collection procedure was conducted at four Florida Turnpike operated toll facilities located in Central Florida. Each site differed from the others in terms of number of lanes, lane configuration, toll base fee, highway location, traffic demand, and vehicle percentage. The sites chosen for data collection were: the Lake Jesup Mainline Plaza along the Seminole Expressway (SR-417), the Beachline West Expressway Toll Plaza along the SR-528, the Daniel Webster Western Beltway Plaza along SR-429, and the Leesburg Toll Plaza along the Florida Turnpike Mainline SR-91. Upon completion of calibration of the two simulation models it is determined that each of the two software are successful in modeling toll plaza capacity and queuing. As expected, each simulation model does possess benefits over the other in terms of set up time, analysis reporting time, and practicality of results. The SHAKER model setup takes mere seconds in order to create a network and input vehicle, another few seconds to calibrate driving parameters, and roughly 10 additional seconds to report analysis. Conversely, setting up the VISSIM model, even for the most experienced user, can take several hours and the report analysis time can take several more hours as it is dependant on the number of required simulation runs and complexity of the network. VISSIM is most beneficial by the fact that its modeling allows for driver variability while SHAKER assumes equilibrium amongst lane choice and queuing. This creates a more realistic condition to observed traffic patterns. Even though differences are prevalent, it is important that in each simulation model the capacity is accurately simulated and each can be used to benefit operational situations related to toll plaza traffic conditions.

Simulation modeling is the most cost-effective way of studying real life transportation problems, either existing or anticipated, without disturbing the balance of the transportation system. There is a vast suite of simulation models available in market, ready to choose from macroscopic, mesoscopic, or microscopic in nature, to study different transportation system elements like freeways, highways, signalized and un-signalized intersections. However, most of these network simulation models, like PARAMICS, VISSIM, CORSIM … etc, do not come readily available with built in toll plaza models. On the other hand, many researchers have independently developed toll plaza models, which can only model an isolated toll plaza without the road network. These toll plaza models, which are based on queuing theory (and some are macroscopic in nature), do not take into account headway, gap acceptance, or inter-vehicle interaction to follow a lead car or to perform lane changing maneuvers. Vehicles just upstream of the toll plaza are assigned to one of the toll lanes, solely based on the payment method (manual, automatic coin machine, or electronic toll collection) and queue lengths at the toll lanes. For instance, if a vehicle is traveling in the leftmost lane and the rightmost toll lane has the shortest queue length, then the queuing model will assign this vehicle to the rightmost lane, and the vehicle will do unrealistic maneuvering to reach to the assigned toll lane instantly. Microscopic network simulation models simulate the vehicular movements based on lane-changing and car-following rules. If such a model could be customized to serve the purpose of the toll plaza simulation, it will simulate the vehicular movements just upstream and downstream of the toll plaza more realistically. Being a network simulation model, it can also model the road network integrated with the plaza, which can be used to study the entire toll road corridor, unlike the isolated toll plaza models. In addition to being a microscopic network simulation model, PARAMICS has many simulation tools, which can be customized to develop a network model with enhanced toll plaza simulation capabilities. PARAMICS also provides the flexibility of using an aerial picture of the toll plaza and upstream/downstream sections of the road as overlay, to ensure that the toll plaza model operates under similar geometric conditions as the real plaza. Using an overlay, exact details of the transition area can be fed into the model. In real life, there is a smooth transition (in terms of the number of lanes and the width of the roadway) from the uniform free-flowing section of the roadway to the toll plaza. Detailed representation of the transition area, in terms of geometry and curb of the roadway along with the number of lanes, is essential for a realistic toll plaza simulation. This kind of detail is not available in a queuing model. As the roadway approaches the toll plaza, it contains more lanes compared to its upstream segments. However, in a simulation model vehicles have a tendency to maintain the same old lanes, and the newly added lanes remain unoccupied by the vehicles. Next-lane Allocation feature in PARAMICS can be used to map upstream lanes onto downstream lanes, preventing this unrealistic behavior from occurring in the simulation model. It tells the vehicles in a particular upstream lane to choose from one or more of the downstream lanes as per the settings. Next-lane allocation can be used in such a manner that all the downstream lanes are utilized. PARAMICS has several other tools such as Restrictions Manager, Vehicle Type Manager, Lane-choices Rules, HOV Lanes, and Vehicle Actuated (VA) Signals which can be used in combination to build a toll plaza model. A microscopic 'Holland East Plaza - SR408' network model has been developed using PARAMICS V5.1. This model contains the plaza and the downstream section of SR 408 Westbound till I-4 interchange in downtown Orlando. This model has been successfully calibrated and validated for the mainline toll plaza and ramp volumes for year 2004. Several hypothetical incident scenarios were simulated to study an entire corridor from the toll plaza to Interstate 4. It was found that the volumes on I-4 off-ramp and SR 408 mainline were affected the most under incident conditions. Volumes for other ramps were not affected in the same proportions. An incident on mainline toll road affected the throughput of the plaza significantly, but the same is not true for an incident on an off-ramp. Travel times to I-4 off-ramps and SR 408 thru lanes were the most sensitive in each of the incident scenarios. In case of the elimination of tolls during the hurricane evacuation, the throughput of the plaza increased significantly. Travel times for the vehicles coming through the plaza and going to different destinations decreased significantly, while it increased for vehicles using on-ramps, because of their inability to merge in the mainline traffic due to the increased toll road volume. The developed model in this thesis has the potential of transportation network wide applications with multiple toll plazas.

The effectiveness of modifying a conventional toll plaza for implementation of an open road tolling concept with express ETC lanes was evaluated in this thesis. Speed controlled dedicated ETC lanes were replaced with express ETC lanes at the Orlando-Orange County Expressway Authority (OOCEA) University Mainline Toll Plaza. This evaluation was accomplished by utilizing collected field data and simulated scenarios using Toll Plaza SIMulation (TPSIM) software developed by the University of Central Florida. The speed controlled dedicated ETC lanes were located within toll lanes (contained within a toll plaza canopy) with widths ranging between 10 to 14 ft. These types of lanes required all vehicles to reduce their speed from the highway speed to 35 mph. Express ETC lanes (sometimes referenced as open road tolling or non-stop tolling) allow vehicles to pass through the plaza at high speeds. Open road tolling is a concept that employs high speed toll lanes. A before and after study of the University toll plaza was conducted. Benefits in the form of reduced delays and increased capacities were observed when making the comparison between the before and after studies. Since we expect the capacity of an express ETC lane to be greater than the dedicated ETC lanes (due to an increase in free-flow speed), further analysis using equations and car-following theory proved that if the ETC speed was increased, then the capacity would increase as well. Using equations derived from the Highway Capacity Manual (HCM) and car-following theory, the capacity was increased from 2016 to 2314 vph when the ETC speed increased from 31 mph to 65 mph. This indicated an increase in capacity of 14.8 percent (based on the conversion from dedicated to express ETC lanes). The field data was also used as input for TPSIM (a computer simulation model) in order to perform a sensitivity analysis of the express ETC lanes by varying the type of ETC lane, number of approach lanes, and plaza configurations (the addition of an ACM lane) between scenarios. Results that were observed during the after study were verified using the TPSIM scenarios. Reductions in delays for the entire plaza were observed using the TPSIM model when making similar improvements to the plaza as in the after study. The changes made to the University Mainline Toll Plaza after construction was completed resulted in benefits by reducing delays and increasing the capacity of the toll plaza (by converting dedicated ETC lanes to express ETC lanes and adding an additional A/ETC lane per direction). These benefits were measured using field data and confirmed when performing the TPSIM scenarios. A customer's travel time along the toll facility will be reduced by using the express ETC lanes (since they are not required to decelerate at the toll plaza). In addition, weaving maneuvers downstream of the plaza are no longer required by customers using the express ETC lanes due to the location of the downstream travel lanes in relation to the express ETC lanes. These benefits may have led to changes in the number and percentage of ETC users in each of the toll lanes. Changes in ETC usage in the conventional mixed-use lanes directly impacted the throughput and delays for each of these lanes, since ETC equipped vehicles have a service time of zero seconds. In addition to the operational benefits, other possible benefits for express ETC lanes were identified and recommended for further evaluation and research. The re-distribution of customers at the plaza due to the implementation of open road tolling, in the form of express ETC lanes, was a great benefit to the overall traffic operations for the University Mainline Toll Plaza in Orlando, Florida.

In spite of the most up-to-date investigation of the relevant techniques to analyze the traffic characteristics and traffic operations at a toll plaza, there has not been any note worthy explorations evaluating delay from toll transaction data and using Artificial Neural Networks (ANN) at a toll plaza. This thesis lays an emphasis on the application of ANN techniques to estimate the total vehicular delay according to the lane type at a toll plaza. This is done to avoid the laborious task of extracting data from the video recordings at a toll plaza. Based on the lane type a general methodology was developed to estimate the total vehicular delay at a toll plaza using ANN. Since there is zero delay in an Electronic Toll Collection (ETC) lane, ANN models were developed for estimating the total vehicular delay in a manual lane and automatic coin machine lane. Therefore, there are two ANN models developed in this thesis. These two ANN models were trained with three hours of data and validated with one hour of data from AM and PM peak data. The two ANN models were built with the dependent and independent variables. The dependent variables in the two models were the total vehicular delay for both the manual and automatic coin machine lane. The independent variables are those, which influence delay. A correlation analysis was performed to see if there exists any strong relationship between the dependent (outputs) and independent variables (inputs). These inputs and outputs are fed into the ANN models. The MATLABTB code was written to run the two ANN models. ANN predictions were good at estimating delay in manual lane, and delay in automatic coin machine lane.

Este trabalho propõe um método para análise de níveis de serviço em praças de pedágio. Para atingir o objetivo proposto foi realizada revisão bibliográfica, e levantados (i) fatores que afetam as capacidades de rodovias e (ii) praças de pedágio, (iii) principais trabalhos que analisaram capacidades e níveis de serviço em praças, e suas limitações e (iv) indicadores de desempenho para praças de pedágio utilizados em trabalhos da literatura e na prática brasileira. Níveis de serviço estão diretamente relacionados à percepção de desconforto e inconveniência de usuários quando trafegando em distintas situações de tráfego. Neste sentido, foram elaborados cenários micro-simulados contendo alguns fatores intervenientes na percepção de qualidade de tráfego em praças. Os cenários foram utilizados em pesquisa qualitativa junto a usuários, com vistas a captar as percepções de qualidade de tráfego. O indicador de desempenho selecionado para representar a qualidade de tráfego nas praças foi o comprimento médio das filas nas cabines. De posse das percepções de qualidade do tráfego declaradas, realizou-se modelagem e foi obtida equação que relaciona os fatores (variáveis explicativas) com as percepções de qualidade de tráfego (variável dependente). Comprimentos de filas estão diretamente relacionados com o número de veículos em fila e a composição do fluxo de tráfego, assim foram propostos fatores de equivalência veicular. Com a finalidade de realizar aplicação do modelo, foram elaboradas tabelas, que relacionaram a percepção de qualidade do tráfego e o indicador de filas médias equivalentes nas cabines. Por fim, foram realizadas análises operacionais de processamentos de veículos, e obtidas capacidades de cabines para fluxos exclusivos. A análise proporcionou a obtenção de fatores de equivalência operacionais para processamento de veículos nas cabines e tempos de espera em fila, assim como identificação de que volumes de tráfego próximos as capacidades são julgados pelo método proposto como nível de serviço "D", diferentemente de critério verificado, por exemplo, no HCM-2000. / This work proposes a method for the analysis of the level of service at toll plazas. To achieve the proposed objective, a literature review was performed to identify: (i) factors that affect the capacity of roads, (ii) factors that affect the capacity of toll plazas, (iii) studies that examined capacity and level of service for toll plazas, and its limitations and, (iv) measures of efficiency for toll plazas. The levels of service are directly related to driver's perception of discomfort and inconvenience when traveling in different traffic conditions. To evaluate the relationship between driver's perception and level of service, scenarios were designed taking into account factors that involve the perception of quality of traffic at toll plazas. The scenarios were used in a qualitative study where drivers were asked concerning their perceptions about traffic quality. The control variable used to represent the traffic quality at toll plazas was the average queue length at booths. The outcome of the study was a model that relates the factors (explanatory variables) with the perceptions of traffic quality (dependent variable). Since queue length is directly proportional to the number of vehicles in queue and to traffic flow composition, equivalence factors were proposed for heavy vehicles in the queue. Results obtained from the model were placed into tables. Finally, operational analyses for processing vehicles were held and the capacity of microsimulated tollbooths was obtained. Through the analyses it was possible to define operational equivalence factors for processing vehicles at booths and corresponding waiting time in queue. In addition, unlike other methods, i.e. HCM2000, the levels of traffic flow near toll plaza capacities were evaluated by the proposed method as service level "D".

Este trabalho propõe um método para análise de níveis de serviço em praças de pedágio. Para atingir o objetivo proposto foi realizada revisão bibliográfica, e levantados (i) fatores que afetam as capacidades de rodovias e (ii) praças de pedágio, (iii) principais trabalhos que analisaram capacidades e níveis de serviço em praças, e suas limitações e (iv) indicadores de desempenho para praças de pedágio utilizados em trabalhos da literatura e na prática brasileira. Níveis de serviço estão diretamente relacionados à percepção de desconforto e inconveniência de usuários quando trafegando em distintas situações de tráfego. Neste sentido, foram elaborados cenários micro-simulados contendo alguns fatores intervenientes na percepção de qualidade de tráfego em praças. Os cenários foram utilizados em pesquisa qualitativa junto a usuários, com vistas a captar as percepções de qualidade de tráfego. O indicador de desempenho selecionado para representar a qualidade de tráfego nas praças foi o comprimento médio das filas nas cabines. De posse das percepções de qualidade do tráfego declaradas, realizou-se modelagem e foi obtida equação que relaciona os fatores (variáveis explicativas) com as percepções de qualidade de tráfego (variável dependente). Comprimentos de filas estão diretamente relacionados com o número de veículos em fila e a composição do fluxo de tráfego, assim foram propostos fatores de equivalência veicular. Com a finalidade de realizar aplicação do modelo, foram elaboradas tabelas, que relacionaram a percepção de qualidade do tráfego e o indicador de filas médias equivalentes nas cabines. Por fim, foram realizadas análises operacionais de processamentos de veículos, e obtidas capacidades de cabines para fluxos exclusivos. A análise proporcionou a obtenção de fatores de equivalência operacionais para processamento de veículos nas cabines e tempos de espera em fila, assim como identificação de que volumes de tráfego próximos as capacidades são julgados pelo método proposto como nível de serviço "D", diferentemente de critério verificado, por exemplo, no HCM-2000. / This work proposes a method for the analysis of the level of service at toll plazas. To achieve the proposed objective, a literature review was performed to identify: (i) factors that affect the capacity of roads, (ii) factors that affect the capacity of toll plazas, (iii) studies that examined capacity and level of service for toll plazas, and its limitations and, (iv) measures of efficiency for toll plazas. The levels of service are directly related to driver's perception of discomfort and inconvenience when traveling in different traffic conditions. To evaluate the relationship between driver's perception and level of service, scenarios were designed taking into account factors that involve the perception of quality of traffic at toll plazas. The scenarios were used in a qualitative study where drivers were asked concerning their perceptions about traffic quality. The control variable used to represent the traffic quality at toll plazas was the average queue length at booths. The outcome of the study was a model that relates the factors (explanatory variables) with the perceptions of traffic quality (dependent variable). Since queue length is directly proportional to the number of vehicles in queue and to traffic flow composition, equivalence factors were proposed for heavy vehicles in the queue. Results obtained from the model were placed into tables. Finally, operational analyses for processing vehicles were held and the capacity of microsimulated tollbooths was obtained. Through the analyses it was possible to define operational equivalence factors for processing vehicles at booths and corresponding waiting time in queue. In addition, unlike other methods, i.e. HCM2000, the levels of traffic flow near toll plaza capacities were evaluated by the proposed method as service level "D".

Este trabalho propõe um método para análise de níveis de serviço em praças de pedágio. Para atingir o objetivo proposto foi realizada revisão bibliográfica, e levantados (i) fatores que afetam as capacidades de rodovias e (ii) praças de pedágio, (iii) principais trabalhos que analisaram capacidades e níveis de serviço em praças, e suas limitações e (iv) indicadores de desempenho para praças de pedágio utilizados em trabalhos da literatura e na prática brasileira. Níveis de serviço estão diretamente relacionados à percepção de desconforto e inconveniência de usuários quando trafegando em distintas situações de tráfego. Neste sentido, foram elaborados cenários micro-simulados contendo alguns fatores intervenientes na percepção de qualidade de tráfego em praças. Os cenários foram utilizados em pesquisa qualitativa junto a usuários, com vistas a captar as percepções de qualidade de tráfego. O indicador de desempenho selecionado para representar a qualidade de tráfego nas praças foi o comprimento médio das filas nas cabines. De posse das percepções de qualidade do tráfego declaradas, realizou-se modelagem e foi obtida equação que relaciona os fatores (variáveis explicativas) com as percepções de qualidade de tráfego (variável dependente). Comprimentos de filas estão diretamente relacionados com o número de veículos em fila e a composição do fluxo de tráfego, assim foram propostos fatores de equivalência veicular. Com a finalidade de realizar aplicação do modelo, foram elaboradas tabelas, que relacionaram a percepção de qualidade do tráfego e o indicador de filas médias equivalentes nas cabines. Por fim, foram realizadas análises operacionais de processamentos de veículos, e obtidas capacidades de cabines para fluxos exclusivos. A análise proporcionou a obtenção de fatores de equivalência operacionais para processamento de veículos nas cabines e tempos de espera em fila, assim como identificação de que volumes de tráfego próximos as capacidades são julgados pelo método proposto como nível de serviço "D", diferentemente de critério verificado, por exemplo, no HCM-2000. / This work proposes a method for the analysis of the level of service at toll plazas. To achieve the proposed objective, a literature review was performed to identify: (i) factors that affect the capacity of roads, (ii) factors that affect the capacity of toll plazas, (iii) studies that examined capacity and level of service for toll plazas, and its limitations and, (iv) measures of efficiency for toll plazas. The levels of service are directly related to driver's perception of discomfort and inconvenience when traveling in different traffic conditions. To evaluate the relationship between driver's perception and level of service, scenarios were designed taking into account factors that involve the perception of quality of traffic at toll plazas. The scenarios were used in a qualitative study where drivers were asked concerning their perceptions about traffic quality. The control variable used to represent the traffic quality at toll plazas was the average queue length at booths. The outcome of the study was a model that relates the factors (explanatory variables) with the perceptions of traffic quality (dependent variable). Since queue length is directly proportional to the number of vehicles in queue and to traffic flow composition, equivalence factors were proposed for heavy vehicles in the queue. Results obtained from the model were placed into tables. Finally, operational analyses for processing vehicles were held and the capacity of microsimulated tollbooths was obtained. Through the analyses it was possible to define operational equivalence factors for processing vehicles at booths and corresponding waiting time in queue. In addition, unlike other methods, i.e. HCM2000, the levels of traffic flow near toll plaza capacities were evaluated by the proposed method as service level "D".

No-lane closure workzone is typical during the construction of open road tolling lanes of a toll plaza. The influence of no-lane closure on toll plazas' performance is unknown because very few studies have been conducted to address this topic. The open road tolling (ORT) has become the new trend of operating an efficient toll plaza. So, the upgrading of a toll plaza from gated E-pass to open road E-pass has become common. The better the toll plaza authority knows about the influence of this construction and congestion effects, the better it can serve the costumers. This project mainly deals with the effects of no-lane closure workzone on the toll plaza performance, and with the collected data, a model was developed predicting 15 minutes throughput and queue length. To better study the workzone impact on toll plaza performance, three sites with different characteristics were selected. They are Lake Jesup Mainline Plaza along the Seminole Expressway (SR-417), the Beachline West Expressway Toll Plaza along the SR-528 and Conway toll plaza along the Holland East-West Expressway (SR-408) in Orlando area of Central Florida. Data preparation includes demand, throughput, processing rates, and queue lengths of different toll categories. Data was collected during peak period for before and during the no-lane closure construction (phase 1) at SR-528 and Lake Jesup toll plaza at SR-417, and middle lane construction (phase 2) and after opening ORT lanes (phase 3) at Conway toll plaza at SR-408.Comparisons were conducted between non-construction stage and construction stage for non-lane closure workzone effects study using data from 417 and 528, and comparisons between middle-lane-construction and complete of construction stage for ORT impact study using data from 408. Analysis results showed that when the toll plaza is operating at or close to its capacity, the no-lane closure workzone can have a negative impact on its performance. But when the toll plaza's demand is lower than the capacity, the no-lane closure workzone has no impact at the toll plaza's performance. And the ORT lanes have a positive influence on the capacity and throughput of the toll plaza. After the impact of no-lane closure workzone on toll plaza has been analyzed, all the data from three toll plazas are put together and a model is built using the variables of Demand/Capacity ratio, percentage of each category of vehicles, E-pass, Automatic or Manual, number of Manual lanes, workzone or no-workzone. Throughput and Queue length can be predicted by this model.