Glossary of Six Sigma Terms: Letters M – O

M

• MANOVA in Six Sigma. 

  The acronym MANOVA stands for Multiple Analysis of Variance. It is a type of multivariate analysis. While it may sound similar to ANOVA at first, MANOVA deals with multiple outputs instead of just one. In MANOVA,  be sure to correlate your outputs according to your data. If you don’t properly correlate your outputs, then you will need to use a separate ANOVA analysis for each one. While ANOVA typically deals with only one input factor, MANOVA may have several. MANOVA is useful in Six Sigma practices as you can carry out several ANOVA analyses at once. If you do, however, and the factors don’t correlate, you will be more likely to discover a Type I error. There are some who prefer ANOVA over multiple pairwise t-tests, but both have their upsides. Black Belts practice MANOVA most often.

• Measurement Systems Analysis.

  There are many measurement systems in Six Sigma, all of which take variation and error into account. When you measure a characteristic in measurement systems analysis, your recorded readings will include measurement errors found. It can be difficult to notice the finer details of measurement systems analysis, with some results overlooked altogether. When this happens, Six Sigma practitioners tend to assume the measurements reflect the characteristic’s true value. This leads to further confusion, resulting in unproductive process improvement solutions.

• You should remember that Six Sigma demands accurate, justifiable data to be able to function at all. Six Sigma work relies on strong data to make educated decisions for process improvement. You may end up trying to improve process variation when the source of that variation is in your measurement system. This is where measuring systems analysis comes in, as it allows you to set the groundwork for statistical process control and experimental design. Six Sigma practitioners working with data and statistics will use measurement systems analysis when your measurements become faulty.

• Muda.

  In Lean and Lean Six Sigma, Muda is a Japanese term meaning ‘waste.’ It is a term every Lean Six Sigma practitioner should know, as it is your job to identify, eliminate, and prevent waste from accumulating. There are seven varieties of waste that appear in manufacturing. Overproduction (when a process is overworked for no added value, wasting time, effort, and resources). Waiting (when production delays unnecessarily). Transporting (when transportation of products takes an excessive amount of time). Inappropriate processing (when a process or task is performed incorrectly, leading to confusion, fault, and other problems). Unnecessary inventory (when your resources accumulate without being used, often wasting funds). Unnecessary and excess motion (when none-value-adding tasks slow down production). Defect (when products are faulty).

N

• Natural Process Limits.

  In Six Sigma and Lean, natural process limits are a way to measure variation in processes. You should arrange them at plus and minus three standard deviations from your goal. Therefore, provided the process is constant and unchanging, around 99.7% of your process output should land within your natural process limits. If, however, your natural tolerance limits are also within your specification limits, then the Cp value will be higher than 1. Your specification limits are essential to determining whether a process is capable of meeting specifications on a regular basis.

• You can define specification limits as confining dimensions of a characteristic, usually imposed by the consumer or the producer. It is up to the supplier to make certain the process can meet the specifications per the required process capability value. Cp values are one of the several indices used to predict performance. Specification limits and Cp values go hand in hand with natural process limits. Any Six Sigma Belt or Lean practitioner involved in process improvement will be familiar with these concepts.

• Nominal Group Technique.

  There are many teamwork tools in Six Sigma, one of which is nominal group technique. NGT allows you to create a short list of items that need to be acted on, very much like a ‘to do list.’ In team activities, Black Belts assume responsibility and administer tasks to complete. They will use nominal group technique as part of their project preparation. There are two stages when using nominal group technique.

• Stage 1: Ensure that everyone understands the purpose of your proposed activity. Why are you doing this? What are the benefits? This is important as NGT, like Six Sigma, relies on completely comprehensive team effort. You should also make sure that each group member works in silence, writing down ideas for tasks and responsibilities. The Black Belt will supervise the task. Then, once everyone has finished, your Black Belt should direct a discussion of everyone’s ideas. It’s important to provide time for clarification and understanding so that everyone knows how the project will proceed.

• Stage 2: Now you need to come up with a shortlist of tasks and ideas, ensuring it is a practical size. Then, hand out index cards to your group members, one for every five ideas listed. Have them write down one idea choice per card, ranking them in order of preference and Finally, you should summarize everyone’s scores on a flip pad or whiteboard. Work towards a unanimous agreement on which ideas you should use.

O

• OC Curve.

  Your OC curve is your operating characteristic curve. Six Sigma Green Belts use the OC curve in sampling inspection to plot the probability of accepting faulty product batches. Or at least one that goes against the batch’s standard quality level. You can also use your OC to determine the average run length in control charts. Operating characteristics are important sources of information in statistical Six Sigma work.

• For example, imagine you have an operating characteristic curve for a sample of 50 items out of 2000. When using the critical acceptance number ‘c’ of 2, this means a batch will be accepted with two or fewer defective items in your sample. From your operating curve, you can determine a probability percentage for accepting batches with defective items. Similarly, when creating your sample plan, be sure to decide on two points. The AQL and LQL, as well as the associated Producer’s and Consumer’s Risks, are also important here.