Abstract: A computer-implemented method and system for identifying process steps for purposes of queue-time control and abnormality detection is provided. In one example, the method includes retrieving manufacturing information associated with a fabrication process, where the manufacturing information includes multiple process step pairs. The manufacturing information may be divided into at least a high group and a low group using a statistical clustering method. Values, such as P-values, may then be calculated for each process step pair by applying a non-parametric statistical method to the high and low groups. The process step pairs may be ranked based on their calculated values, and redundant process step pairs may be eliminated. The remaining process step pairs may then be analyzed to identify a particular process step or process step pair.

Abstract: A computer-implemented method and system for identifying process steps for purposes of queue-time control and abnormality detection is provided. In one example, the method includes retrieving manufacturing information associated with a fabrication process, where the manufacturing information includes multiple process step pairs. The manufacturing information may be divided into at least a high group and a low group using a statistical clustering method. Values, such as P-values, may then be calculated for each process step pair by applying a non-parametric statistical method to the high and low groups. The process step pairs may be ranked based on their calculated values, and redundant process step pairs may be eliminated. The remaining process step pairs may then be analyzed to identify a particular process step or process step pair.