Six Sigma and Statistical Modeling
Without care, a leak may ultimately ¢ood an entire process. As Benjamin Franklin notes, even a small leak in a process can mushroom into a big gap that can lead to project failure. Taking care of a process requires understanding it fully. Statistical modeling is one good way to achieve a better understanding of a process that can then lead to implementation of improvement techniques such as Six Sigma. Statistics involves tabulating, depicting, and describing data sets. It constitutes a body of formal techniques that attempt to infer the properties of a large collection of data from inspection of a sample from the collection. The basics of statistics involve measures of central tendency and variation. The tools include SIPOC (suppliers, inputs, processes, outputs, customers), key process inputs and outputs, value stream mapping, process maps, cause and effect (C&E) matrices, FMEA, continuous measurement system analysis, attribute measurement system analysis, and process capabilities. Variation measures are central tendency, shape, and spread.