Thesis (Ph. D. in City and Regional Planning)--University of California, Berkeley, 2003. / "Fall 2003." Includes bibliographical references (p. 173-183). Also available online via the ITS Berkeley web site (www.its.berkeley.edu).

The thesis evaluates mobility based on mobile phone positions. The aim is to develop and assess different methods for travel demand estimation based on CDR data. Besides this estimation location data in cellular data is explained in more detail and a previous work based on mobile phone data and travel demand estimation is reviewed. The different methods of travel time estimation include both static and dynamic estimation. The static travel demand estimation evaluates movements in the city based on predefined time periods, whereas the dynamic estimations are based on different definitions of a trip. A trip can be defined as movements between important places, or just simply count a trip between each position, or a filtering of active states to create more accurate origin-destination matrices. The second part of the thesis includes evaluation of travel time based on CDR data before the final conclusions are drawn. The main finding of the thesis is that it is possible to assess mobility in a city based on CDR data, even if there are no validation data available.

Long-range demand planning and capacity management play an important role for policy makers and airline managers alike. Each makes decisions regarding allocating appropriate levels of funds to align capacity with forecasted demand. Decisions today can have long lasting effects. Reducing forecast errors for long-range range demand forecasting will improve resource allocation decision making. This research paper will focus on improving long-range demand planning and forecasting errors of passenger traffic in the U.S. domestic airline industry. This paper will look to build upon current forecasting models being used for U.S. domestic airline passenger traffic with the aim of improving forecast errors published by Federal Aviation Administration (FAA). Using historical data, this study will retroactively forecast U.S. domestic passenger traffic and then compare it to actual passenger traffic, then comparing forecast errors. Forecasting methods will be tested extensively in order to identify new trends and causal factors that will enhance forecast accuracy thus increasing the likelihood of better capacity management and funding decisions.

The financial and environmental effects of traffic congestion and automobile-centric air pollution continue to be problems that must be addressed within the United States. In response, travel demand management (TDM) has emerged as a potential way to reduce automobile-based travel in order to minimize these effects. TDM strategies are highly dependent on specific urban form characteristics such as bicycle lanes, sidewalks, or transit facilities. A current gap exists in the analytical tools available to transportation planners when evaluating TDM projects. The standard transportation models do not take into account urban form characteristics in a systematic way. These characteristics play an import role in an individual's selection of walking, bicycling, or transit based travel modes. This gap needs to be filled in order to evaluate TDM projects with the same decision-making rigor that is applied to road expansion projects. The purpose of this project is to develop an enhanced transportation mode choice model that presents a systematic approach for incorporating urban form characteristics. This approach determines which elements of urban form have the strongest influence on transportation mode choice behavior. This work is being done in conjunction with the Roanoke Valley Allegheny Metropolitan Planning Organization as a way to evaluate the potential of TDM projects in promoting non-automobile forms of travel within the Roanoke region. This approach to developing an enhanced transportation mode choice model is a step forward in address the gap between TDM strategies and the tools needed to evaluate them. / Master of Urban and Regional Planning

As transportation policies are changing to encourage alternative modes of transportation to reduce congestion problems and air quality impacts, more planning organizations are considering or implementing activity-based travel demand models to forecast future travel patterns. The proclivity towards operating activity-based models is the capability to model disaggregate travel data to better understand the model results that are generated with respect to the latest transportation policy implementations. This thesis first examines the differences between the two major modeling techniques used in the United States and then describes the assessment tool that was developed to recommend whether a region should convert to the advanced modeling procedures. This tool consists of parameters that were decided upon based on their known linkages to the advantages of activity-based models.

This paper presents a two-stage model of optimization as a quick method to choose the best potential links for implementing urban travel demand management (UTDM) strategy like road pricing. The model is optimized by minimizing the hidden cost of congestion based on user equilibrium (MHCCUE). It forecasts the exact amount of flows and tolls for links in user equilibrium condition to determine the hidden cost for each link to optimize the link selection based on the network congestion priority. The results show that not only the amount of total cost is decreased, but also the number of selected links for pricing is reduced as compared with the previous toll minimization methods. Moreover, as this model just uses the traffic assignment data for calculation, it could be considered as a quick and optimum solution for choosing the potential links.

O presente estudo procura identificar grupos homogêneos de indivíduos quanto aos padrões de seqüências de atividades diárias que estes realizam. As atividades são caracterizadas por múltiplos atributos, fazendo com que as seqüências sejam multidimensionais. Como atributos, ou características, são considerados a natureza da atividade realizada, ou motivo da viagem, e o período de realização da mesma, ambos separados em categorias. É estudado o efeito da inclusão da forma de acesso à atividade, ou modo de viagem, como uma terceira dimensão. Este atributo, entretanto, dados os resultados obtidos, não é utilizado nas análises finais. É também considerada a adoção de diferentes categorizações para a dimensão motivo. São usados dados da pesquisa Origem e Destino realizada em 1997, na Região Metropolitana de São Paulo. No trabalho são considerados os indivíduos com 12 anos ou mais, com pelo menos duas viagens diárias e com seqüência de viagens iniciada e terminada em sua residência, sem inconsistências internas. O número de indivíduos que atende a estes critérios é 49.616. A classificação, ou agrupamento, das seqüências de atividades em classes ou grupos é feita considerando uma medida de distância ou dissimilaridade calculada entre as seqüências, que é baseada no esforço necessário para igualá-las. Esta medida é chamada de OT-MDSAM (uni-dimensional Optimum Trajectories-based MultiDimensional Sequence Alignment Method). A partir da matriz de dissimilaridades é executado um processo estatístico de agrupamento hierárquico aglomerativo usando o Método de Ward. Os grupos de seqüências formados são analisados considerando características das próprias seqüências e atributos sóciodemográficas e econômicas dos indivíduos que os compõem, e usados em um modelo de segmentação do tipo árvore de decisão, usando o CHAID (Chi-square Automatic Interaction Detector). Resultados indicam que os grupos formados são bastante homogêneos quanto aos padrões de seqüências de atividades que representam e aos indivíduos associados a eles. / The main objective of the dissertation is to identify homogeneous groups of individuals, with regard to the daily activity/travel sequences performed in a weekday. Activities are characterized by multiple attributes, thus generating mutidimensional seguences. In this study, the nature of the activity (travel purpose) and the starting period of engagement in the activity (ending time of a trip) were the dimensions considered in the characterization of activities. Access mode to the activity was also considered as a third dimension, but the results had led to the decision not to include it in the final analysis. Alternative categorizations of the activity nature dimension were also studied, that resulted in further disaggregation than adopted in previous analyses of the same data. The study used data from the 1997 Origin-Destination household survey of the Sao Paulo Metropolitan Area. The analysis considered all individuals aged 12 or over that conducted two or more trips (starting and ending at home) on the survey day, resulting in a sample of 49,616 individuals. A sequence alignment method - OT-MDSUM (uni-dimensional Optimum Trajectories-based MultiDimensional Sequence Alignment Method) - was used to compare and calculate distances between pairs of different activity/travel sequences. These distances were then fed into a Ward hierarchical clustering algorithm to create classes of groups of activity/travel patterns. These groups were then analyzed according to the characteristics of the activity/travel sequences included and to the sociodemographic and economic characteristics of individuals who performed these patterns. The data were then utilized to develop a decision tree model using CHAID - Chi-Squared Automatic Interaction Detector, having the group of activity/travel sequences as the response variable and the characteristics of individuals and their families as independent variables. The results indicate that the groups formed through this procedure present a good degree of homogeneity regarding the activity patterns they represent and that they can be clearly associated to the characteristics of the individuals which perform these patterns.

A contínua ampliação da infraestrutura viária já se provou ineficaz em resolver os problemas de congestionamento nos grandes centros urbanos e uma solução não sustentável tendo em vista os altos investimentos envolvidos. Em contraposição ao tradicional aumento da capacidade viária, surge a Gestão da Demanda de Viagens (GDV). Uma das medidas englobadas pela GDV é a mobilidade corporativa que visa a promover o uso de opções de transporte mais sustentáveis e eficientes nos deslocamentos casa-trabalho das pessoas. Aproximadamente 50% dos deslocamentos diários nas cidades brasileiras ocorrem por motivo de trabalho. Por isso organizações públicas e privadas desempenham um papel fundamental em questões atreladas ao transporte. Embora não controlem a forma como seus funcionários vão ao trabalho, as organizações, por muitas vezes, podem estimular a mudança de hábitos de deslocamento ao prover informações e incentivos para isso. Este trabalho tem por objetivo verificar a aplicabilidade da adoção de estratégias de mobilidade corporativa em organizações situadas no Brasil. Em um primeiro momento, apresenta-se mais profundamente o conceito, os benefícios e as medidas de mobilidade corporativa que podem ser implementadas. A seguir, são comparados e analisados diferentes métodos existentes para a elaboração de planos de mobilidade corporativa. Propõe-se um novo método adaptado à realidade local composto por sete passos que totalizam 26 atividades que devem ser cumpridas para a construção de um plano bem-sucedido. Realiza-se uma pesquisa-ação em um complexo administrativo que reúne 17 mil funcionários e está localizado a 20 km do centro da cidade. Investiga-se o padrão de deslocamento casa-trabalho dos funcionários da organização e são analisadas que medidas de mobilidade corporativa poderiam ser adotadas para estimular o transporte sustentável entre os funcionários deste local. / The continuous expansion of road infrastructure has proven to be ineffective in solving the problem of congestion in large urban areas. It is also an unsustainable solution due to the high investments involved. Instead of increasing road capacity, Travel Demand Management (TDM) emerges as an alternative to deal with the problem. One of TDM’s measures is corporate mobility, which aims to promote more sustainable and efficient transport options for commuting to work. Approximately 50% of daily trips in Brazilian cities are work related. Therefore public and private organizations play a key role in issues related to transportation. Although they do not control how employees commute to work, organizations have the ability to stimulate travel behavior change by providing information and incentives. This study aims to verify the applicability of the adoption of corporate mobility strategies in organizations located in Brazil. At first, a more in depth concept is explained in addition to the benefits and corporate mobility measures that can be implemented. Then, we compare and analyze different existing methods for the construction of corporate mobility plans. We propose a new method based on the local context. The method is composed by seven steps totalizing 26 activities which must be followed for developing of a successful plan. Finally, we describe the application of the method that was carried out in an Administrative Center which gathers 17 thousand employees and is located 20 km away from the city center. We investigate employees’ commute patterns and analyze which corporate mobility measures could be implemented to promote a more sustainable commute pattern to work.

Travel to outdoor recreational spaces belongs to a general class of research questions for understanding destination and travel mode choices. In travel demand modelling, discrete choice models (DCMs) have been applied to understand and predict a wide range of choices, such as how people choose among alternative destinations for jobs, homes, shopping, personal services etc. Surprisingly, DCMs have rarely been used to understand and model travel to outdoor recreational spaces. In the current literature for modelling travel to outdoor recreational spaces, the established models are Negative Binomial Regression (NBR) models, such as what was used in the UK NEA studies. However, these NBR models were developed to assess the effects of travel to outdoor recreational spaces at a national level, and they are not intended for assessing choices of individual sites. One reason for this is, as identified by previous studies, is that compared with the DCMs, the NBR models have certain limits on estimating people's choice behaviours. There is, therefore, no existing model that can represent and predict how people choose to travel to outdoor recreational spaces. Given the importance of outdoor recreational activities to urban land use planning and public health, this is a clear gap in the field. The aim of this study is to develop a new travel demand model capable of representing and predicting travel to individual outdoor recreational sites. This is achieved by answering four main research questions: First, how to build the new model for outdoor recreational travel? Secondly, is the estimation accurate enough? Thirdly, to what extent can the new model be transferred to destinations outside the case study area? And, finally, how can city planners and designers use this new method? The new model draws upon ideas from random utility theory that underlies the conventional travel demand models to represent trip generation, trip distribution and mode choice. This research follows the standard modelling procedure: data collection and preliminary analysis, model calibration, model validation and model application. The data are collated from a wide range of sources that, importantly for model transferability, cover all areas in England. The new model has been calibrated for a case study area which spanned 14 selected districts in the North-West region. Validation of the new model is based on estimating the numbers of trips to two outdoor recreational sites (Wigg Island and Wigan Flashes) and to nine English National Parks where data on visitor trips exist. In the final stage of the research, the new model is applied to estimate the changes that would arise from planning and design interventions in existing (Wigg Island and Moore Nature Reserve) and proposed (Arpley Country Park) sites. At the end of this process, it is possible to show that the new model can predict the number of trips to individual destinations and that the model can be transferred to other outdoor recreation sites. Furthermore, the new model presented here is capable of predicting the changes in the volume and catchment of visits to an existing green space after land use planning or urban ecological interventions. This is a completely new theoretical model that is focused on understanding and quantifying the travel choices to outdoor recreation sites, which can inform decision makers by forecasting changes in outdoor recreational travel demand, according to different planning scenarios.

Modeling travel demand by time of day is gaining increasing attention in travel demand forecasting practice. This is because time of day choice has important implications for mode choice and for quantifying potential modal and time of day shifts in response to traffic congestion and peak period travel demand management strategies. In this context, understanding the causal relationship between time of day (departure time) choice and mode choice behavior would be useful in the development of time of day based travel demand modeling systems both within the four-step modeling paradigm and within newer tour-based and activity-based microsimulation paradigms. This thesis investigates the relationship between departure time choice and mode choice for non-work trips as work trips tend to be constrained with respect to time of day choice. Two alternative causal structures are considered in this thesis: one structure in which departure time choice is determined first and mode choice is subsequently influenced by departure time choice and a second structure in which mode choice is determined first and affects departure time choice. These two causal structures are analyzed in a recursive bivariate probit modeling framework that allows random error covariance. The estimation is performed separately for worker and non-worker samples drawn from the 1999 Southeast Florida Regional Household Travel Survey. For workers, model estimation results show that the causal structure in which departure time choice precedes mode choice performs significantly better. For non-workers, the reverse causal relationship in which mode choice precedes departure time choice is found to be a more suitable joint modeling structure. These two findings can be reasonably explained from a travel behavior perspective and have important implications for advanced travel demand model development and application.