A MULTIVARIATE ANALYSIS OF FREEWAY SPEED AND HEADWAY DATA

摘要: The knowledge of speed and head way distributions is essent ial in m icroscopic traffic flow studies because an d headway are both fundamental microscopic characteristics flow. For simulation models, one key process the generation entry vehicle speeds arrival times. It helpful to find desirable mathematical odel individual values, tim e simulations usu ally generated based on som form models. Traditionally, for investigated separately independent each other. However, this traditional approach ignores possible dependence between headway. To address issue, research pr esents a ethodology construct bivariate describe characteri stics Based investigation freeway da ta measured from loop detector data IH-35 Austin, it shown that there exists weak betw een first proposes skew-t ixture models capture he terogeneity distribution. Finite can significantly im prove goodness fit data. develop distribut ion headway, study