#
All entries for December 2012

## December 23, 2012

### Bell Curve

To analyze the process in the company, it is essential to learn about process first, then to illustrate the process by charts and diagrams. Thus, after learning and describing the process, the individual and moving range charts were created to illustrate the process. Then to identify how process suits to customers’’ requirements the process capabilities were calculated. At this stage I face a problem – how to illustrate those calculations?

As I understood, this is a six-sigma logic, if you want, - first, to calculate and then illustrate results. Unfortunately, I didn’t know how to illustrate the process capabilities. I knew how picture should look like with Cp = 0,351 but couldn’t find the formula that helped me to build the graph. Luckily, I found an article where the application of Gaussian Bell Curve to six sigma project was described.

So what does, basically, Gaussian Bell Curve can show? The main aim of this function is to consider the distribution of the process. If the points of our process is distributed “Normally”, then around 68% of the points should lie within one standard deviation of the mean, 95% of the points should lie within two standard deviations of the mean, and 99.7% of the points should lie within three standard deviations of the mean.

Although the Gaussian Bell Curve is useful tool in six sigma analysis, its calculation is complicated, as was mentioned in the article. The Cp and Cpk calculations are much simpler and provide you the same results. Moreover, by calculating process capabilities you can also get picture of the process’s distribution. Maybe, it will be not as clear as using Gaussian Bell Curve but its calculation will be, definitely, less time-consuming.

## December 12, 2012

### Presenting data in context

Almost every company compares one number with another number. It is so easy to do and to their point of view this comparison can provide the evidence that process is below or above the target. Is it? Can we really judge how good our business operating by such simple comparison?

All information that we get from others we put in the context. So why then figures and numbers should be considered independently? Statistical Process Control suggests to build a flow chart and then also put natural limits of the process, thus, to present figures and numbers in the context. This is reasonable and it makes sense, BUT…what I cannot understand and agree with is that this traditional ways of communication with numbers are considered to be useless.

As Wheeler states this comparison cannot filter out the noise of the process. Why not? For example, if we develop target with taking into account economic environment and seasonality (two noise factors) and compare it versus actual numbers in this way we do filter out the noise of the process. After that we can use obtained data and put it into the context by using control chars. I do believe that this traditional comparison is useful. If not, then why everybody is still using it?

## December 09, 2012

### Identifying quality characteristics

Which tool to use to identify quality characteristics of the process? And what do you mean by quality characteristics? These are two main questions which I faced during PMA writing.

The quality characteristics are usually a measurement of the process. And **SIPOC** diagram involves the measurement of the process identification consequently I can use it as a tool. But SIPOC is commonly considered as a tool for describing the process, not identifying how it can be measured. In that point of view SIPOC is not the best choice. Can we then use **fishbone diagram** logic to identify the characteristics? Fishbone diagram enable to develop those characteristics through 5 factors: methods, materials, people, environment, and equipment. But it usually applied when the need of cause and effect analysis appeared. A good combination of SIPOC and fishbone principles is presented in **CTQ tree. **Using this tree you focus on customers’ expectations, which can be taken from SIPOC diagram, then for each customer expectation you develop the key characteristics which are essential to satisfy the customer. And this CTQ tree uses the same logic as fishbone diagram – it describes the characteristics by combining them in different groups (drivers). In case with fishbone these are 5 factors.

CTQ tree seems to be most appropriate tool that helps to identify quality characteristics of the process, there are lots of others tools, but CTQ provides a highly descriptive approach with a good visualization.

## December 02, 2012

### Filter out the noise

Never thought that creating a Control Chart can take so much time. You have to consider almost everything:

- The nature of the process. If it have subgroups or not.
- The size of the sample. Sometimes 10 values cannot show the whole picture, so it is better to take at least 20
- Finally, noises of the process. Before building a chart, make sure that you filter out the noise factors.

The last requirement really drove me into a corner and I stuck. How can I take into account such noise as *seasonal character of the sales*? Or, for example, I took a small selling company which opened its branch not so far ago, so I also had to consider the *growth rate of the company*, but how??? After two days of agony I still didn’t have any answer, but then I sudenly stumble across the report, where company described how they developed sales targets. They were wrighting that each monthly target was developed according to expected growth of the business and sales seasonality. After that I decided to take figures of sales performance comparing monthly targets (sales/target). In that case, the noise factors were filtered as it was taken into account earlier in targets development.

This time I managed this task somehow, but I really believe that noises in some processes cannot be filtered out so easily.