In this dissertation, it is aimed to develop an approach for the trip distribution element which is one of the important phases of four-step travel demand modelling. The trip distribution problem using back-propagation artificial neural networks has been researched in a limited number of studies and, in a critically evaluated study it has been concluded that the artificial neural networks underperform when compared to the traditional models. The underperformance of back-propagation artificial neural networks appears to be due to the thresholding the linearly combined inputs from the input layer in the hidden layer as well as thresholding the linearly combined outputs from the hidden layer in the output layer. In the proposed neural trip distribution model, it is attempted not to threshold the linearly combined outputs from the hidden layer in the output layer. Thus, in this approach, linearly combined iv inputs are activated in the hidden layer as in most neural networks and the neuron in the output layer is used as a summation unit in contrast to other neural networks. When this developed neural trip distribution model is compared with various approaches as modular, gravity and back-propagation neural models, it has been found that reliable trip distribution predictions are obtained.

Regional growth caused the emergence of traffic congestion and pollution in the past few decades, which have started to affect small urban areas. These problems are not only related to transportation system design but also to land use planning. There has been growing recognition that the relationship between land use and transportation needs to be understood and analyzed in a consistent and systematic way. Integrated urban models have recently been introduced and implemented in several metropolitan areas to systematically examine the relationship between land use and transportation. The general consensus in the field of integrated urban models is that each model has its own limitations and assumptions because they are each designed for different application purposes. This dissertation proposes a new type of methodology to integrate the regression-based land use model and the combined trip distribution-assignment transportation model that can be applied to both metropolitan areas and small urban areas. The proposed integrated land use and transportation model framework has three components: the regression-based land use model, the combined trip distributionassignment transportation model, and the interaction between these two models. The combined trip distribution-assignment model framework provides the platform to simultaneously integrate the transportation model with the land use model. The land use model is developed using an easy-to-implement method in terms of correlation and regression analysis. The interaction between the land use model and the transportation model is examined by two model frameworks: feedback model framework and simultaneous model framework. The feedback model framework solves the land use model and the transportation model iteratively. The simultaneous model framework brings the land use model and the transportation models into one optimization program after introducing the used path set. Both the feedback model and the simultaneous model can be solved to estimate link flow, origin-destination (OD) trips, and household distribution with the results satisfying network equilibrium conditions. The proposed integrated model framework has an “affordable and easy-toimplement” land use model; it can be performed in small urban areas with limited resources. The model applications show that using the proposed integrated model framework can help decision-makers and planners in preparing for the future of their communities.

O objetivo deste trabalho é testar um modelo de distribuição de viagens do tipo gravitacional para a calibração de matrizes origem/destino (O/D) em linhas de transporte público por ônibus a partir de dados de pesquisa de contagem de embarque/desembarque (E/D). A metodologia proposta possibilita a obtenção de matrizes O/D de forma rápida e barata, pois combina um método de pesquisa simples e de baixo custo (pesquisa de contagem de usuários) com um modelo de distribuição de viagens. O modelo associa a cada ponto de origem um valor de parâmetro. A utilização de um valor de parâmetro associado a cada origem busca neste caso reproduzir o custo médio de distribuição de viagens de uma origem em relação aos seus diferentes destinos. O modelo incorpora ainda como restrição a probabilidade de um passageiro desembarcar no ponto seguinte ao seu ponto de embarque. Os dados de pesquisa foram cedidos pelo Departamento de Transportes da Universidade Federal do Paraná. O teste de desempenho do modelo foi realizado através da comparação entre matrizes O/D observadas e simuladas para as cidades de Curitiba e Paranaguá. / The aim of this work is to test a gravity model for trip distribution designed to estimate bus routes origin/destination (O/D) matrices based on boarding and alighting data. The proposed method combines a simple and low-cost survey method (on/off passenger counting) with a mathematical model for trip distribution, which enables the estimation of an O/D matrix in a fast and inexpensive manner. The model assumption that each origin point is associated to a parameter value tries to reproduce the average costs of the actual trip distribution from each origin to every single destination along the bus route. The model brings also as a built-in restriction an expected (usually low) probability of passengers getting off the vehicle in the bus stop following the boarding point. The survey data used in this work have been collected by researchers of the Transportation Department at the Federal University of Paraná. The model performance has been tested by the comparison of observed and simulated O/D matrices in the cities of Curitiba and Paranaguá. The results found in most of the simulations showed that for an estimated trip frequencies did not statistically differ from the actual values for a required level of significance.

Dans la pratique de la planification urbaine, la prévision de la demande de transport fait en général appel au modèle à quatre étapes (génération, distribution, répartition modale et affectation), malgré des avancées théoriques considérables dans le domaine. Cette persistance s’explique par une facilité relative de mise en oeuvre, liée notamment à la forme des données disponibles et susceptibles d’alimenter les modèles. Cependant, la nature statique de l’approche pose des interrogations quant à sa pertinence pour faire des prévisions de moyen-long terme. Cette thèse étudie, la validité de l’hypothèse de stabilité temporelle des trois premières étapes du modèle de prévision. Pour ce faire, en prenant l’agglomération lyonnaise comme terrain d’étude, nous avons codifié des réseaux routiers et de transports en commun à différentes dates (1985, 1995 et 2006). Cette donne, généralement indisponible, combinée aux enquêtes ménages déplacements correspondantes, nous permet de calibrer les trois premières étapes du modèle traditionnel et de tester leur capacité prédictive. Pour les modèles de génération, on note des prévisions acceptables à un horizon de 10 ans. À 20 ans, certaines évolutions dans les styles de vie se sont traduites par une baisse du nombre moyen de sorties pour le motif travail, que les modèles traditionnels ne permettent pas de prévoir complètement. Au niveau de la distribution, l’allongement des distances entre lieux de réalisation de certaines activités et le lieu de domicile peut être relativement bien reproduit par des modèles gravitaires avec des paramètres stables dans le temps. Au niveau de la répartition modale, les paramètres ne sont pas stables et les modèles estimés n’auraient pas permis de prévoir le regain de parts de marché des transports en commun observé ces dernières années. / In the practice of urban planning, travel demand forecasts are generally obtained by using the four-step model (generation, distribution, modal split and assignment), despite considerable theoretical advances in the field. This persistence can be explained by the relative ease of implementation of the four-step modelling sequence, which is related, in particular, to the kind of data available that could be used as an input in a model. However, the static nature of the approach raises questions as it pertains to its relevance in producing medium and long range forecasts. This thesis investigates the validity of the hypothesis of temporal stability of the parameters of the first three stages of the traditional forecasting sequence. To do this, taking the Lyon conurbation as our case study, we coded the road and transit networks at different points in time (1985, 1995 and 2006). We then combine this temporal data, which is generally unavailable, with the corresponding household travel surveys in order to calibrate the first three steps of the traditional model and test their predictive ability. For the generation models tested, we note acceptable performance for a 10-year forecast. For a 20-year forecast, some changes in lifestyles have resulted in a decrease in the average number of work trips that traditional models do not predict accurately. Regarding trip distribution, the increase in travel distances observed for certain purposes is reproduced fairly well by the gravity model. At the modal split level, the parameters are not stable and the estimated models would be unable to predict accurately the recent increase in the market share of public transport.

O objetivo deste trabalho é testar um modelo de distribuição de viagens do tipo gravitacional para a calibração de matrizes origem/destino (O/D) em linhas de transporte público por ônibus a partir de dados de pesquisa de contagem de embarque/desembarque (E/D). A metodologia proposta possibilita a obtenção de matrizes O/D de forma rápida e barata, pois combina um método de pesquisa simples e de baixo custo (pesquisa de contagem de usuários) com um modelo de distribuição de viagens. O modelo associa a cada ponto de origem um valor de parâmetro. A utilização de um valor de parâmetro associado a cada origem busca neste caso reproduzir o custo médio de distribuição de viagens de uma origem em relação aos seus diferentes destinos. O modelo incorpora ainda como restrição a probabilidade de um passageiro desembarcar no ponto seguinte ao seu ponto de embarque. Os dados de pesquisa foram cedidos pelo Departamento de Transportes da Universidade Federal do Paraná. O teste de desempenho do modelo foi realizado através da comparação entre matrizes O/D observadas e simuladas para as cidades de Curitiba e Paranaguá. / The aim of this work is to test a gravity model for trip distribution designed to estimate bus routes origin/destination (O/D) matrices based on boarding and alighting data. The proposed method combines a simple and low-cost survey method (on/off passenger counting) with a mathematical model for trip distribution, which enables the estimation of an O/D matrix in a fast and inexpensive manner. The model assumption that each origin point is associated to a parameter value tries to reproduce the average costs of the actual trip distribution from each origin to every single destination along the bus route. The model brings also as a built-in restriction an expected (usually low) probability of passengers getting off the vehicle in the bus stop following the boarding point. The survey data used in this work have been collected by researchers of the Transportation Department at the Federal University of Paraná. The model performance has been tested by the comparison of observed and simulated O/D matrices in the cities of Curitiba and Paranaguá. The results found in most of the simulations showed that for an estimated trip frequencies did not statistically differ from the actual values for a required level of significance.

Transportation planners are often faced with the problem of estimating passenger and freight flows between regions. In the literature there are many models for passenger flows. However, models about freight flows are more limited. Modeling freight flow is also more complex than modeling passenger flow and there are many agents related with freight flows. In addition, data availability is a critical factor. In this research, freight flows between provinces in T&uuml / rkiye are forecasted by demand analysis. Transportation is one of the important activities of human beings and plays an important role for spatial interactions in economic growth. In other words, there is a very strong linkage between economic growth and the freight flow, thus transportation demand. Regional trade as spatial flow appears on transportation systems as freight flows. In this study, using the existing limited data and surveys in T&uuml / rkiye, nationwide origin-destination (O-D) matrix of freight flows between provinces is obtained. Using this empirical matrix, the generation of freight flows of provinces is formulated depending on the socioeconomic and demographic variables by means of multiple linear regression analysis. In addition, interactions of freight flows between provinces and economic growth of regions are investigated. The generations and attractions of provinces as freight flow are distributed between provinces with traditional gravity model. By comparing observed O-D matrix and simulated O-D matrix, gravity model is calibrated. Calibration is also performed by freight trip length distribution. In this research, two steps of traditional &ldquo / four-step analysis&rdquo / , &ldquo / trip generation&rdquo / and &ldquo / trip distribution&rdquo / , are applied to develop nationwide freight demand model between the provinces in T&uuml / rkiye. The developed model is single-mode, single commodity and nationwide.

The Portfolio Analysis Management Office (PAMO) for the Aeronautics Research Mission Directorate (ARMD) at NASA Headquarters tasked the Systems Analysis and Concepts Directorate at NASA Langley to combine efforts with Virginia Tech to develop a global demand model with the capability to predict future demand in the air transportation field. A previous study (Alsalous, 2015) started the development of the Global Demand Mode (GDM) to predict air travel demand based on Gross Domestic Product (GDP) and population trends for 3,974 airports worldwide. The study was done from year 2016 to year 2040. This research project intends to enhance the GDM capabilities. A Fratar model is implemented for the distribution of the forecast demand during each year. The Fratar model uses a 3,974 by 3,974 origin-destination matrix to distribute the demand among 55,612 unique routes in the network. Moreover, the GDM is capable to estimate the aircraft fleet mix per route and the number of flights per aircraft that are needed to satisfy the forecast demand. The model adopts the aircraft fleet mix from the Official Airline Guide data for the year 2015. Once the aircraft types are distributed and flights are assigned, the GDM runs an aircraft retirement and replacement analysis to remove older generation aircraft from the network and replace them with existing or newer aircraft. The GDM continues to evolve worldwide aircraft fleet by introducing 14 new generation aircraft from Airbus, Boeing, Bombardier, and Embraer and 5 Advanced Technology Aircraft from NASA. / Master of Science / The Portfolio Analysis Management Office (PAMO) for the Aeronautics Research Mission Directorate (ARMD) at NASA Headquarters tasked the Systems Analysis and Concepts Directorate at NASA Langley to combine efforts with Virginia Tech to develop a global demand model with the capability to predict future demand in the air transportation field. A previous study (Alsalous, 2015) started the development of the Global Demand Mode (GDM) to predict air travel demand based on Gross Domestic Product (GDP) and population trends for 3,974 airports worldwide. The study was done from year 2016 to year 2040. The previous study done by Alsaous, predicts how many seats will be departing out of the 3,974 airports worldwide. This project intends to use the outputs of the GDM and distribute the seats predicted among the airports. The objective is to predict how many seats will be offered that will be departing from airport “A” and arriving at airport “B”. For this, a Fratar model was implemented. The second objective of this project is to estimate what will the aircraft fleet be in the future and how many flights will be needed to satisfy the predicted air travel demand. If the number of seats going from airport A to airport B is known, then, by analyzing real data it can be estimated what type of aircraft will be flying from airport “A” to airport “B” and how many flights each aircraft will have to perform in order to satisfy the forecasted demand. Besides of estimating the type of aircraft that will be used in the future, the modeled created is capable of introducing new aircraft that are not part of the network yet. Fourteen new generation aircraft from Airbus, Boeing, Bombardier, and Embraer and 5 Advanced Technology Aircraft from NASA.

Predicting the passenger flow inside a city is a vital component of the intelligent transportation management system. The proposal for a new residential area, an office space, post­pandemic policy implications for work from home, behavioral changes for revised traffic patterns, infrastructural improvements, require a visual and analytical backing which can be provided through a macro simulation model. This research explores the performance of the Machine learning (ML) based transport model against the predictions provided by the traditional Spatial Interaction Models (SIM) for the city of Oslo. The transport models and their parameters are analyzed for sensitivity analysis and scenario analysis to derive city character. Furthermore, the derived model is deployed over an interactive dashboard for analytical and their practical visualizations through infographics. The results show that the ML model outperforms the SIM. Although the traditional SIM has a clear advantage of being interpreted by design and requiring a few parameters, it suffers from its inability to accurately capture the structure of real flows and greater variability as compared to the ML model. Extensive statistical analyses are conducted to obtain significant results and realize the pros and cons of both the models which question the validity of results for the ML model over SIM. With this thesis, we discuss the potential of ML model detected trends of passenger flows, andtheir capacity to simulate city development­related scenarios for the traffic flows within the city.

Generalised linear models (GLMs) provide a flexible and sound basis for calibrating gravity models for trip distribution, for a wide range of deterrence functions (from steps to splines), with K factors and geographic segmentation. The Tanner function fitted Wellington Transport Strategy Model data as well as more complex functions and was insensitive to the formulation of intrazonal and external costs. Weighting from variable expansion factors and interpretation of the deviance under sparsity are addressed. An observed trip matrix is disaggregated and fitted at the household, person and trip levels with consistent results. Hierarchical GLMs (HGLMs) are formulated to fit mixed logit models, but were unable to reproduce the coefficients of simple nested logit models. Geospatial analysis by HGLM showed no evidence of spatial error patterns, either as random K factors or as correlations between them. Equivalence with hierarchical mode choice, duality with trip distribution, regularisation, lorelograms, and the modifiable areal unit problem are considered. Trip distribution is calibrated from aggregate data by the MVESTM matrix estimation package, incorporating period and direction factors in the intercepts. Counts across four screenlines showed a significance similar to a thousand-household travel survey. Calibration was possible only in conjuction with trip end data. Criteria for validation against screenline counts were met, but only if allowance was made for error in the trip end data.

Neste trabalho é proposto um modelo para estimar a geração e a distribuição de viagens intraurbanas baseado em agentes, denominado Modelo GDA. Neste modelo foram aplicadas simulações em Sistemas Multiagentes (SMA), nas quais foram usadas, como dados de entrada, informações relativas ao morador e ao uso do solo. Na estimativa da geração de viagens, a simulação SMA foi usada para estimar uma população sintética baseada nas informações sociodemográficas dos moradores e para obter um plano de atividades associado a cada morador. Na estimativa da distribuição de viagens, a simulação SMA foi usada para obter uma matriz Origem-Destino (OD) com base no plano de atividades dos moradores e nos atributos do uso do solo. Para definir os critérios da distribuição de viagens, foram testadas teorias alternativas à força gravitacional, como a teoria das redes livres de escala e o conceito de path dependence. Foi feita uma aplicação do Modelo GDA na cidade de São Carlos (SP), cujos resultados estimados foram comparados aos resultados observados, provenientes da pesquisa Origem-Destino (OD), e aos resultados estimados obtidos pela aplicação de modelos gravitacionais nesta cidade. Os resultados mostraram que os resultados estimados obtidos pelo Modelo GDA são tão acurados quanto aos do modelo gravitacional. Conclui-se que o Modelo GDA, comparativamente ao modelo gravitacional, possui vantagens quanto à sua aplicabilidade, pois em vez de serem utilizados pelo modelo dados provenientes de pesquisas de tráfego, geralmente onerosos e de difícil aquisição; são empregados dados acerca do morador e do uso do solo, de fácil coleta e atualização periódica. / In this work, an agent-based model in order to estimate trip generation and trip distribution in an intra-urban context (GDA model) is proposed. Simulations using Multiagent Systems (MAS), with input data concerning dwellers and land use were applied in this model. To estimate the trip generation, the MAS simulation was used to elaborate a synthetic population based on sociodemographic information of the dwellers and to obtain an activity plan of each dweller. To estimate the trip distribution, the MAS simulation was used to obtain an Origin-Destiny (OD) matrix based on the dwellers activity plans and the land use characteristics. To define the trip distribution rules, alternative theories to gravitational force like free scale networks and path dependence theories were tested. The GDA model was applied in the urban area of São Carlos (Brazil), whose estimates was compared to the observed data from the OD survey and the estimate data from the Gravity model applied in this same area. The results showed that the estimates from the GDA Model are as accurate as from the Gravity Model. It was observed that the GDA Model presents advantages in relation to the Gravity Model because instead of using traffic survey data, which often is expensive and difficult to get, it uses dwellers and land use information, which is periodically collected from government researches, making it easy for government agencies to obtain this information.