Phd Degree / Doktora

Permanent URI for this collectionhttps://hdl.handle.net/11147/2869

Browse

Filters

2000 - 20092

Settings

Author: search.filters.author.Duvarcı, YavuzLanguage: search.filters.language.en

Search Results

Now showing 1 - 2 of 2
  • Doctoral Thesis
    Traffic Accident Predictions Based on Fuzzy Logic Approach for Safer Urban Environments, Case Study: Izmir Metropolitan Area
    (Izmir Institute of Technology, 2009) Selvi, Ömer; Duvarcı, Yavuz
    Dissertation has dealt with one of the most chaotic events of an urban life that is the traffic accidents. This study is a preliminary and an explorative effort to establish an Accident Prediction Model (APM) for road safety in İzmir urban environment. Aim of the dissertation is to prevent or decrease the amount of possible future traffic accidents in İzmir metropolitan region, by the help of the developed APM. Urban traffic accidents have spatial and other external reasons independent from the vehicles or drivers, and these reasons can be predicted by mathematical models. The study deals with the factors of the traffic accidents, which are not based on the human behavior or vehicle characteristics. Therefore the prediction model is established through the following external factors, such as traffic volume, rain status and the geometry of the roads. Fuzzy Logic Modeling (FLM) is applied as a prediction tool in the study. Familiarizing fuzzy logic approach to the planning discipline is the secondary aim of the thesis and contribution to the literature. The conformity of fuzzy logic enables modeling through verbal data and intuitive approach, which is important to achieve uncertainties of planning issues.
  • Doctoral Thesis
    A Specaial Transportation Modelling Approach for the Disadvantaged Groups in Urban Traffic
    (Izmir Institute of Technology, 2000) Duvarcı, Yavuz; Süel, Akın
    It has been known that conventional transportation models and studies have been inadequate to solve the acute transportation problems in the urban areas to date that overwhelmingly those disadvantaged groups face. The major reason is probably that the qualitative and social parameters that could be the real factors in the explanation of the "disadvantagedness" have not been efficiently considered in the modelling because of the uncertainty, and non-normative nature of the models.This thesis study as a normative one offers basically two methodological approach that can be integrated to the normal models: First, the clear-cut defInition of so-called disadvantaged by the cluster analysis method, and second, application of the modelling procedures both for normal case and for the disadvantaged simulatenously by which the determination of policy packages (policy capturing) is probable on the basis ofthe differences between the two models. The improvement of disadvantaged means getting closer to the normality in transportation conditions. With this, it is meant that policy-making to improve the disadvantaged can even start from the modelling stages.As an innovative approach, vanous correlated variables are grouped into "Major" variables in the form of function formulations, which are thought best represent the social/qualitative parameters.In the model runs, latest version of TRANUS (6.0) was used to speed up the modelling process (especially the Trip Assignments). With TRANUS., categorical handling is possible. It is found that there appears a remarkable discrepancy between the two models at the level of Trip Generations (productions), and some different variables could be used in the model for disadvantaged.Finally, in the Correspondence Module (or, category analysis), as of the adopted equity principles, the association of those disadvantaged categories (as transportation categories) with the "disadvantagedness" levels is maintained. This matching process also provided a gauge with which the policies could be produced for the matching transportation (disadvantaged) categories. To see the effectiveness of the method, three simulations are run based on the three policy scenarios where any move towards betterment in the condition of disadvantaged is welcome.As the result of this study, though less than expected, an improvement was observed in the travel conditions of the disadvantaged. It is observed not surprisingly that policy formulations playing around the income related and vehicle ownership variables can be more successful in obtaining better results. More frequent trials with better scenario formulations as well would have ended with better results.Key Words: Transportation Planning, Equity, Transportation Disadvantaged, Cluster Analysis, Category (Correspondence) Analysis, Transportation Ethics, Modelling-, Fuzzy Sets Analysis, Sensitivity Analysis, Simulation-, Gini Index.