A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science of the Asian Institute of Technology, Bangkok, Thailand

Thesis (M.Sc.) - Asian Institute of Technology, 1978

For determining the saturation flow at signalized intersections in Bangkok, 17 approaches at 9 signalized intersections in the city central area were observed. These observations were performed from the beginning of green phase until the end of saturated flow condition for each cycle. Data were gathered by a single observer by photographing the site from a vantage point using a normal speed video-camera. Projecting the data tapes on a television screen, the headways were measured by individual lane. At the same time, the vehicles were classified as cars, commercial vehicles" samlors and motorcycles, and the turning movements were recorded. The estimation theory of statistical inference was used to determine when the flow of traffic became saturated, and the Student t-test was used to investigate whether the saturation flow was affected by lane position. The effects of lane widths and vehicle composition were determined quantitatively by simple regression analysis and multiple regression analysis, respectively. The effects of other factors on saturation flow, such as turning vehicles, gradient, and kerb lane use were determined by comparison. This study also calculated the passenger-car-unit equivalents for various straight-ahead vehicles and the through-car-unit equivalents for turning vehicles. The lost time at the start of the green period was also estimated.