Evaluation of approach leg capacities at a signalized urban roundabout


Tezin Türü: Yüksek Lisans

Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Mühendislik Fakültesi, İnşaat Mühendisliği Bölümü, Türkiye

Tezin Onay Tarihi: 2018

Öğrenci: EGE CEM SALTIK

Danışman: HEDİYE TÜYDEŞ YAMAN

Özet:

Intersections are vital components of road networks, which provide ability to change directions serving many conflicting movements. When the traffic volumes are high, intersection control is necessary via either signalization or by design such as roundabouts. However, the even latter is mostly signalized in Turkey due to improper design aspects and driver behavior issues. When signalized, approach capacity at an intersection/roundabout is governed by the saturation flow rate, , as well as the allocated effective green time, . For the determination of , the literature has some formulae using design characteristics or site measurements, which are based on ideal and simple conditions, whereas on-site situations can be very complex and different. This study focused on evaluation of nature of traffic, more specifically saturation flow and inefficiency during green time on the approach legs of a signalized roundabout. The numerical results were obtained from digitization of time headways, , from fisheye camera recordings at a roundabout connecting a major and a minor arterial in Konya during morning (AM) and evening (PM) peak periods (including 115 cycles) as well as noon off-peak period (48 cycles). Headways are indexed for each lane, , on each approach leg during each cycle separately as , which enabled estimation of corresponding saturation flow rates of based on on Highway Capacity Manual (HCM) and German Highway Capacity Manual (HBS in German) methodologies. Analysis values from 163 cycles were used to develop a practical saturation flow rate, , which was later used in the development of inefficiency because of the fluctuations in the flow rates. During the green time on a lane of an approach ,3-moving average method was used to determine instantaneous flow rates, . The statistical analysis of the 19057 time headway data for 19057 vehicles passing through study intersection showed a lognormal distribution with an interquartile range (IQR) of [1.6s;3s] on major approaches during the peak periods, while IQR on minor legs were [2s;4s]. Time headways were more scattered during the off-peak noon with IQR values of [2s;3s] and [2s;5s] for major and minor approaches, respectively. Regression of headways with vehicular or flow characteristics (vehicle type, cycle length, green time, etc.) did not produce any statistically significant factor effect. Saturation flows based on design characteristics using HCM and HBS methods suggested values in the range of 1900 – 2000 veh/h/ln for different lanes in the major approaches. Estimation of HCM site-based observation method produced higher values reaching up to 2631 veh/h/ln. The practical S values, , were found within the range of 2400 veh/h/ln, which were reached mostly during the peak hour cycles, but also, some of the cycles in the noon, as well. 3-moving average flow rates showed values as high as 2900 veh/h/ln and as low as 600 veh/h/ln, even sometimes in the same cycle. Lane 2 and Lane 3 (the outermost lane) on all approaches generally experienced higher flows regardless of the time or assumed methodology. Average inefficiency during green periods was found as 30 percent even on the major approaches, which were due to traffic break in the upstream flow conditions.