DEVELOPMENT OF MODELS FOR PREDICTING SPEED ON HORIZONTAL CURVES FOR TWO-LANE RURAL HIGHWAYS

A variety of physical factors may influence speeds on two-lane highways. Horizontal curvature has the greatest impact on speed. The objective of this study was to develop curve–speed prediction equations using data collected at 28 sites in northern Iraq. Four equations that include the effect of curve radius and grade on the predicted 85th percentile curve speed were developed and subsequently validated using observed values from another 20 sites. The speed of more than 7000 passenger cars was measured during the study. A series of regression models were developed. The main independent variables chosen for use in the regression analyses were radius of the curve and approach speed. The speed prediction equations performed well, in that the mean absolute percent error ranged between 7 and 9 percent and there was an overall mean absolute percent error of 8 percent. Box plots were developed to demonstrate graphically the range of error between the predicted speeds and the field measurements