Cyclic Weighted k-means Method with Application to Time-of-Day Interval Partition

Time-of-day interval partition (TIP) at a signalized intersection is of great importance in traffic control. There are two shortcomings the traditional clustering algorithms based on distance definitions (such as Euclidean distance) flows. First, some continuous time intervals usually divided into small segments. Second, 0 o’clock (24 o’clock) selected breakpoint. It follows that relationship between TIP and signal control neglected. To this end, novel cyclic flows defined, which can make end last cycle day) beginning current (0 cluster one group. Next, weighted k-means method proposed, with centroid initialization, number selection, breakpoint adjustment. Lastly, proposed applied to real evaluate benefits The conclusion empirical study confirms feasibility effectiveness method.