Missouri Enterprise: Six Sigma & the Power of Data
By LauraLee Rose, ASQ-certified Six Sigma Black Belt & Project Manager for Missouri Enterprise
Six Sigma is a problem-solving methodology with the purpose of reducing variation in processes to reduce defects. It’s based on the use of proven statistical tools to analyze data. Here are 3 things you need to think about if you’re interested in using these powerful quality tools in the framework of the phases of Six Sigma: define, measure, analyze, improve and control (DMAIC).
1. One of the primary tenets of Six Sigma is that data should drive all decisions. Traditionally, often manufacturing decisions have been made by following gut instinct or “I think…, I feel…, I believe…”. Six Sigma at the manufacturer at which I received my Six Sigma Black Belt training, we received coins that said “In God we trust. Everyone else brings data.” So, what happens if you don’t have the data? You get it. I once worked with a client who was trying to prove that their CEO’s idea for a new product was a stinker. They hired me to come in and tell the emperor he was naked. I simply suggested that each of their salespeople call four or five of their best customers and ask them two questions after they explained the concept: Do you like this idea? and Would you buy it? They called 20 of their best customers, and 100% of them said that while it was an okay idea, they could accomplish the same end with their existing equipment. That’s data, and they could present it to the boss with confidence that it was valid. 2. Too much data can be just as bad as not enough. I’ve seen both ends of this spectrum in my 25+ years of manufacturing experience. From no data at all, to companies who are drowning in numbers, and not using any of it (see #3 below). Where does your company fall on this spectrum? Collecting data is not cheap. It takes time for operators to keep track of defects or record measurements, and in the lean world, all data collection is considered non-value-added from the viewpoint of the customer. Figuring out exactly what you need to measure, and then collecting only that required data is very important. A caveat to this: If you have all the data you need, tell those collecting it that they can stop tracking it. I’ve seen too many folks do short-term quality studies, but fail to tell the operators that they have finished the study, and don’t need it collected any more. 3. Data is useless unless it’s evaluated and analyzed. The ISO9001:2015 standard recognizes this and section 9.1 calls it out specifically. An exercise I like is to write down what data you’re tracking at all levels. Now ask “Why?”. For what purpose? How are we using the data? I am willing to bet that there is information being recorded somewhere in your organization that’s not being evaluated or analyzed. According to the website http://www.differencebetween.com, “The key difference between analysis and evaluation is that the evaluation is linked with testing whereas analysis is an in-depth study of a subject matter.”