Hız etüdü: İstanbul O-2 otoyolunun temel bir kesiminin örneği

Speed is one of the fundamental variables of road traffic flow that stands as an important evaluation criterion for traffic analyses in several aspects. In particular, varieties of speed variable, such as average speed, free flow speed, optimum speed (capacity speed), acceleration/deceleration speed and so on, have been explicitly considered in the analysis of not only road safety but also road capacity. In the purpose of analyzing speeds and realizing speed differentiations on freeway traffic, this study presents a case study conducted on a basic freeway segment of O-2 in Istanbul. Chapter one comprises a brief introduction to the study. The fundamentals of road traffic flow are explained in chapter two, briefly including traffic flow measures and variables of traffic flow. Then, the fundamental relationships of traffic flow variables are presented with graphs. Chapter three of this thesis contains a brief literature review. Temporal and spatial variability of free speed along freeway, speed variation and freeway traffic congestion, and freeway speeds and speed variations preceding crashes are the main points of this chapter. The case study of this thesis is explained in chapter four. The numerical implementations for field observations speed data are the major points of chapter four. This chapter encompasses statistical evaluations for speed data of field observations. The speed data are analyzed into two main areas such as uncongested and congested areas. In the context of this thesis, Kruskal-Wallis (K-W) tests were performed to evaluate the effects of changes in lanes, weeks, and days of week on the speeds at uncongested and congested flow areas separately. In this study, Kolmogorov-Smirnov (K-S) tests were implemented to determine the distribution of lane and road speed data of free flow speed region. The analyzed distributions are normal, lognormal, gamma, gumbel, exponential and uniform distributions. Furthemore, the regression analyses were carried out in order to investigate the relationship between maximum speed difference across lanes and road speed. These investigations were performed at uncongested and congested areas, separately. Moreover, the relationship between maximum speed difference across lanes and road flow rate were evaluated by applying regression analyses for both uncongested and congested areas separately. In the final chapter, chapter five, the conclusions of this thesis are given. The results of K-W tests of speed data at uncongested flow area show that the speeds generally vary significantly across lanes; however, the variation of speeds across weeks and days are not that much obvious. The results of K-W tests of speed data at congested flow area show that the speeds do not vary significantly from one lane to the others, from one specific day of week to the others, and from one day of week to the other at some analyzed locations. Moreover, the conclusions of 35 K-S tests at road sections indicate that the free flow speed data fit to normal distribution 20 times, gamma distribution 19 times, lognormal distribution 18 times and gumbel distribution 3 times. It is concluded from the results of regression analyses for uncongested and congested areas that there is moderate relationship between maximum speed difference across lanes and road speed in %50 cases. Additionally, the results of regression analyses for uncongested and congested areas indicate that there is moderate relationship between maximum speed difference across lanes and road flow rate in %30 cases. The maximum speed difference across lanes decreases as the road flow rate increases.