Several statistical techniques have been introduced in zoning sequential data such as well-log data or blocking experimental data. In general, these techniques lack the ability to optimize the predicted zones as per the desired number of blocks or zones. In addition, all these techniques' results depend on the amount of data inclusion in the blocking or zonation process. In this paper, a new method has been introduced to address the blocks or zones optimization and data-amount dependency. The new method also helps in establishing a new approach in estimating the data random error, if not known. This paper does not compare all techniques with the proposed approach. One technique has been selected to highlight the advantages of the proposed method. An example of a Canadian oil-sand well has been used to demonstrate some applications of the new blocking or zoning method.