Clustering data analysis that is robust to noise and is able to extract the most reliable information from sequential data comprises the ranking all of the measurement values across a third dimension of a 3D dataset in a selected one of an increasing order or a decreasing order and producing a three dimensional array of ranked values therefrom. It further comprises identifying coherent 3D patterns from the 3D array of ranked values, and counting the number of identified coherent 3D patterns. Each coherent 3D pattern parameters with a similar ranking and across subsets of the set of elements to a same group defines a cluster.