|
|
Designing the experiments for six sigma: Featured Article
Introduction
Six sigma is fast becoming a proven approach for businesses and organizations to use to improve their performance through taking advantage of some of the principles of lean management.Lean practices, which were once mainly practiced in manufacturing organizations, have earned a place in a variety of different industries.One who is unfamiliar with the possible benefits of six sigma may look at some of the terminology that is used and run the other way.It is true that six sigma is filled with process acronyms and that to the naked eye the process of doing something such as designing your own experiments can be daunting.However, the success of those who have made the efforts to implement a six sigma designed experiment find that the whole mentality of six sigma revolves around very fundamental elements that every organization has and can improve.The design of any experiment within six sigma starts with a simple vision of a goal that is to be achieved.That goal is the same for many: to deliver products and services to customers with no defects from the eyes of the customers.Once the vision is there, business leaders begin the experiment design process by defining their organization's objectives in numerical terms so that the aspects pertinent to the experiment can be accurately measured.As you start the process of designing your own experiments for six sigma remember that six sigma experiment design is all about the methodology that you choose.
In the paragraphs that follow you will be able to read more of an introduction and explanation of the process that is involved in designing experiments for six sigma.The design of the experiment is without a doubt a crucial part on the entire six sigma implementation process and careful study and preparation is needed.Keep in mind that there are specialists who have experience in designing the experiments for six sigma are available for consultation.Neglecting the design of your experiment is not something that you want to do as it can easily jeopardize the usefulness and accuracy of the information that you have worked so hard to analyze.
|
|
Methods of six sigma
There are two key methods that are used in most cases by organizations that are carrying out a six sigma project.These two methods are by no means the only options for your six sigma project, however, their simplicity in concept and history of yielding beneficial results make them an easy and obvious answer to the question of which methodologies to use.You always have the option of creating a method that is unique to your organization.There are many companies who have done this and have found great success in the implementation of a more original method.However, most companies just starting to implement six sigma are likely better off simply sticking with these basic efforts rather than trying to be original for originality's sake.
Helpful Resources: 1. Six Sigma Process This link provides helpful information regarding what the six sigma process entails and how you can design an experiment that utilizes the six sigma process and those trained in the methods. 2. Book on Designing Six Sigma Experiments 3. Developing Experiments 5. Designing the Experiments for Six Sigma 7. DOE and Six Sigma 9. Best DOE Practices 10. Course on Designing Experiments |
The first method of six sigma is called DMAIC or Define, Measure, Analyze, Improve and Control.This method is typically the easiest method to implement because its definition is so clear.This method is also the most appropriate method to implement if the problem that you are trying to resolve has to do with an inability to find appropriate solutions to concerns related to existing products, processes or services.A newer method for designing experiments in six sigma is called the Design for Six Sigma or DFSS.This design for six sigma is typically brought about by other sometimes more well known terms such as IDOV (Identify, Design, Optimize and Validate) and DMADV (Design, Measure, Analyze, Design and Verify).Whatever it is called, the point of implementing this method of experimentation design is to develop a new product, process or service that is free from defect and acceptable as a defect-free product by the consumer.
Six sigma Design of Experiments (DOE)
Once the method for the design experiment is chosen, those who will be responsible for the implementation of the methods are selected and appropriate training is given.It is this implementation process and the certification of Black and Green Belts that attracts a lot of attention by companies.Sometimes the six sigma emphasis is more on accomplishing the implementation rather than spending time on the design of the experiments.The design of the six sigma experiments is undoubtedly the most important aspect of any implementation strategy.It is the foundation by which all future chances are justified.The purpose of your experiment design is to create a process where you can control key variances known as inputs in order to achieve superior results (known as outputs).In DOE the term input is also referred to as a factor and the output is also referred to as responses.
For almost 100 years Design of Experiments has become one of the best known methods for discovering relationships between factors and responses because of its consistent ability to yield valid results.As you follow through with your DOE process keep in mind the purpose of the design of the experiment.Hopefully as you review statistic after statistic and see variable point charted on a variety of different graphs and charts, you will remember that these measurements of outputs and inputs are not imaginary.Outputs are the "y's" of the core statistical equations while inputs are called "x's."
If you have decided to seek out the advice of a consultant to help you in your experiment design, keep in mind that some outsourced curriculum providers will introduce response and optimization designs differently.Another alternative to the traditional experiment design approach that some consulting companies are teaching is called the Taguchi design and is a method for what is called "robust design."
What is OFAT?
OFAT is an acronym that stands for "one factor at a time."When designing experiments for six sigma a company will likely take an OFAT approach to their process.OFAT occurs when someone changes one thing in a process to see what it does.Depending on the results of doing that one thing, a project team is able to make conclusions regarding a number of things such as the effects of specific variable and the correlation between variables.Most people use an OFAT technique to solve their problems but they often do so without even thinking about how effective their method is.Although this approach is logical and will yield somewhat satisfactory results, a lot of time is wasted with trial and error.Additionally, you may find the best setting or options for each given operation but maximizing the effects of cooperative efforts among all of the operations in your organization would take an un-heard of amount of time and effort to accomplish if you plan on using an OFAT technique.
Six sigma implementation and the steps that are involved in the designing of experiments for six sigma are meant to utilize the benefits that exist with the OFAT approach, while eliminating the wasted time and effort.The problem with using OFAT is that you cannot use it (without a ridiculous amount of random trial and error sampling) determine how the various parameters of an operation interact with one another.It is this interaction understanding that makes DOE (Design of Experiments) such a powerful lean implementation tool.
Full factorial designs
The objective that should be kept in mind as you are designing your experiments for six sigma is to achieve a full factorial design.A full factorial design is an approach to identifying what the optimal parameters would be to design a simple full factorial experiment.In other words, instead of taking on random factors and trying to find a way to make them work together best, you will choose the factors and parameters of the project that are most important (or yield the most dramatic and positive effects) and work with only those factors to design an experiment.
When only a few key factors are taken into account, an experiment design can be created that tests all combinations of the variables making all interactions more easily seen.Naturally, care must be taken to ensure that your outputted data is accurate and this often entails creating a system of measurement that is consistent and reproducible.
Conclusion
Designing experiments for six sigma is a process that is here to stay, as more and more individuals are needed who have training in the designing and implementation of these practices.There is always a need for improved design no matter the company or the financial situation.Careful design of experiments is just as important to the small business owner as it is to the corporate executive who has his stock holders to answer to.Six sigma introduces the keys needed for a company to be able to adjust and to grow with the changes that come with variables such as expansion from within or economic challenges.Six sigma marries the statistical processes that are vital to consistency and accuracy of data, to the Design of Experiments methods which allows for the improved effectiveness of any operation area for which an experimental design can be created.
Privacy Policy, Terms of Use