|
|
Deciding which data to measure using six sigma: Featured Article
One of the most important parts of Six Sigma process improvement methodology is data collection and measurement.Data can almost always lead to the problems that are causing a decrease in product quality, quantity, or any other part of the production process that might be having trouble at the time.Without data it is extremely difficult to decide what things need to be changed to make improvements in the process.There are several key steps to make the best decisions about data collection in the Six Sigma process.Here are some pointers that will help you understand the process better and make the best decisions for your business.
There are all kinds of different data that might be useful to a business while they improve their products and processes.The following four steps will help ensure that the most valuable and relevant information will be gathered each time.First, decide what needs to be measured.Second, decide on how samples will be taken and evaluated.Third, come to agree on operational definitions.And fourth, carry out the actual measurement of the data also known as measurement system analysis (MSA).
|
|
Defining data
The defining phase is very important for the implementation of Six Sigma.In this phase a team charter is produced and then an overview of the process to be improved is also drafted.Then information is decided on what is critical for the quality of the product and for consumers.This step involves the most information gathering of the steps.The goal of this phase is to find and pinpoint the reasons or source of the problem and to define it thoroughly.
Helpful Resources: 1. Options in Measuring Data using Six Sigma This source offers a great deal of detail when it comes to the options that exist when measuring data using Six Sigma.Of all these options what you have left to do is decide which data measurements are most important. 2. Developing a Research Method 3. Online Books 5. DMADV Methodology 7. Pocket Guide 9. Data 10. Measuring Data |
It is essential that a factual understanding of the data be in place so that the action plan taken will actually be effective.This knowledge will help you to narrow down the range of potential causes and what you really need to investigate to analyze later.The DMADV (design, measure, analyze, design, and verify) methodology of Six Sigma demands that you define the problem through data.If you can't measure the data you should not even choose to collect or use it.When you are gathering data it is important to do many different things.You need to be able to identify possible solutions and also know how to select the most important and viable options.You should also decide which variables you are going to measure and know what they will give you in return for your efforts.You should them move on to a plan of how to collect the data and how to organize it once you have it.You should also know the different types of data that are available and be sure to find the best and most efficient way to collect the data and use the best operational definitions available.The form that you use to collect data should be useful whether it is variable or attribute data.You need to make sure that there is enough data to make a reliable reading of it, that it is relevant to the goals that have been set in place, that it is representative, and that it is measured and used in the right context.The data that you choose to collect should also clarify the goals that have been put in place by the organization.Anything that detracts or distracts from the goals of the company or causes more inefficiency than improvement should be reevaluated and then other options should be explored.The best information describes the problem that you are having and suggests possible ways to make the problem better.
What is an FMEA?
FMEA is an acronym for Failure Mode and Effect Analysis.This is another tool that can be used to help focus efforts made in data collection to be the most effective possible.It can be used in the measure phase as well as during the Improve phase of Six Sigma because the approach is to identify, estimate, prioritize and then evaluate the value of a risk.The whole reason that these steps are taken is to ensure that failures occur and if a problem does happen you can measure what they were and why the occurred.
Developing operational definitions for your data
When deciding which data to measure using Six Sigma, you want to make sure that you will be able to define and create procedures using the data in question.If you cannot accurately define the data then conducting a test and reviewing the results will lead to results based only in theory and not on concrete data.The next part in data collection is to develop operational definitions and procedures.The objective in doing this is to ensure that all of the date collected measures a universal manner or method.This avoids the problem of variation in the measurement of data and also helps to avoid confusion in the measurements.There are several things that need to be considered when developing these operational definitions.The features that you decide to include should be specific and easily definable.The also have to be measurable and also have to be useful or there is no point in collecting it in the first place at all.Of course there is not always just one right answer and there are many different approaches that can be explored.The more specific the definition used, the more useful it will be in helping increase efficiency.It has been found that the better the definition the better the results too.There may be several times when you need to refine or redo the definition so that it fits the goals that you have put in place.You also need to make sure that the people who will actually be gathering the data understand exactly what you want so that there is uniformity in the way that the data is collected.This ensures that you will have the most accurate information from day to day and from different manufacturing facilities.This is very important because if one person measures something will even a minute difference there could be a very dramatic difference in the kind of data that is collected.It could be shown that there is severe underperformance at a particular plant or during a certain shift and this could be blamed on many things.It is important to ensure that this is not happening just because of a measurement error.
There are many different things that you can test for and some of the most important ones are accuracy, repeatability, reproducibility, stability and adequate resolution.There are different ways that each of these can be approached and measured.To test for accuracy you need to measure throughout the product range things that have known quantities.For measuring repeatability you need to have the same sample measured by the same person using the same measurements so that you can determine any variability in the measurement even with only one person.Testing reproducibility demands that you have many different people measure the same sample repeatedly so that you can see if there are variations from one person to another.You can also have the same person measure the same sample over time to see if the measurements change as the sample is measured repeatedly to determine the stability of the measures.You also need to make sure that you measure at least 5 distinct values in the range of the product variation tests.This ensures that there is adequate resolution.
Before moving on.
When you have finished the defining stage and before you move on to the analyze phase, you should be sure to have defined what problems are occurring and what conditions make them occur.This will help those responsible for spotting these problems to know what to look for and how to approach finding the solution.You should have data available to you that helps you demonstrate the ability to measure important factors.At the end of this phase you should be able to know specifically what the main problem or problems are, show how you have prioritized the information about critical inputs, process and output measures.You should also show what you have done to validate the measurement process and what patterns are found in the data.Along with this you should also be able to show what the capability of the current process is.If you don't have these things in place you may not have been adequately thorough with your research or data collection that you have done.If you find that there are pieces missing and that you need to take extra measures or repeat the previous steps, this is worth the time that you will have to spend on it.It could cost you a lot of time and trouble in the future if you make sure that you have ensured the best possible implementation of each step before moving on to another.Another thing you need to be certain of is that you have kept the whole purpose of Six Sigma in your mind throughout the whole process so that you are working on things that will improve your business and result in more profits and greater efficiency.
Privacy Policy, Terms of Use