One of the challenges in manufacturing is when a product defect occurs intermittently and its source is unknown. The defect could be from a given process, or from one of the machines or operators within a process step. There could also be more sources. The natural inclination is to create a designed experiment to determine the source of the defect. However, this can be impractical for several reasons, mainly because of the intermittent aspect of the defect.
For batch processing the experimental design technique "Swap Tests" can be used to successfully design experiments to isolate the problem area creating the defect. It can be employed for any number of processes and is only limited by a minimum batch size required based upon the design. As an added benefit, it does not shut down operations in order to conduct the testing nor does it require qualification of test material prior to shipment.
Kevin Keller has over 35 years of industrial experience. He spent the first 20 years in the electronic materials world making high tech circuit boards for military equipment as well as silicon wafers for the semiconductor industry. Kevin retired a few years ago from Anheuser-Busch InBev, where he served as a Master Black Belt. Kevin was responsible for Quality Systems in the manufacture of cans and lids for the beverage market. He now serves as an independent consultant for both global and local companies in Lean Six Sigma, Discrete Event Simulation, Quality, and Quality Management. He has a BS degree in Chemical Engineering from the University of Missouri-Rolla and a Masters in Applied Statistics from The Ohio State University. He and his wife, Ann, reside in St. Peters, MO.