This paper describes a simple study of measurement of saturation flow at signalised intersections, and the development of the prediction formula which can be used for Malaysian traffic conditions. The purpose of this study is to determine the validity of Webster Cobbe's (1966) method as to whether it can still be applied in Malaysia since many factors were different there and then. Data were collected from 50 road sites in various cities in Malaysia. In particular, the influence of lane width, gradient and turning radii on saturation flow was studied. The average saturation flow on a flat approach with a lane width of 3.0 metres was found to be 1930 pcu/hr, about 5% higher than that implied by Webster's work for his `normal good to average' sites. The factors for lane width, gradient, turning radius and the proportion of turning vehicles were also derived using linear regression and correlation analyses on the data collected. Using the 1930 pcu/hr which corresponds to a 3 metres lane width as the base saturation flow, predicted saturation flow can be obtained by factorising it for lane width, gradient and turning radius. Other factors which yet need further studies for the local conditions are opposed turning movement, vehicle composition and the appropriate pcu values. (A)
Abstract