Six Sigma Black Belt is 8 days of training for advanced process improvement leaders. The focus of the projects, at this level of proficiency, is typically on optimizing the process - establishing the best settings to maximize and provide predictable process performance.
Black Belt is an extension of the Green Belt cadre of tools. The topics presented are to extend the breadth and depth of methods across the various phases of the roadmap.
Black Belt development is conducted in two 4-day sessions of training with a 3-4 week break between sessions for a total of 60 hours of instruction.
CEU and PDU credits will be provided to students after successfully completing the training and required exam or a live project for certification. In addition, attendees must attend all training and participate in all classroom exercises. If a student misses more than one day of training, make up work must be completed before the final exam may be taken and a training certificate, CEU’s or PDU’s will be issued at this time or after the successful completion of all the training and a live project. Project should be submitted to Six Sigma.us within 6 months of the last day of training for certification. Projects should illustrate an understanding of the tools and concepts of the appropriate training level (i.e. Black Belt). Individuals interested in receiving PDU’s or CEU’s only my take a comprehensive online exam in lieu of a submitting a project with the instructor’s approval. The instructor will determine, if testing or completion of a project is the appropriate means of assessing the student’s understanding of the subject matter. Students wanting to receive Six Sigma certification will be required to complete and submit a project for Master Black Belt review.
Upon completing this course, students will be able to:
In many processes, the underlying distribution is not Normal but estimating the mean is still desired. Understanding the Central Limit Theorem and it’s applicability to sampling theory can aide in many aspects of reducing variability effects.
There are various methods for measuring process performance, and some are more difficult to analyze due to the nature of the data being collected or the type of measurement being made. Participants will learn various techniques to deal with unusual measurement system challenges.
There are various conditions where a standard Capability Analysis does not apply or provide accurate index values. Participants will learn how and when to apply transformation techniques to data for accurate analysis of capability as well as how to evaluate single-sided specifications.
Many situations exist where a process performance measure is affected by more than one continuous factor. Multiple regression techniques will be introduced to determine which factors provide the best prediction of the response, and how to deal with non-linear relationships.
Just as with Multiple Regression, there are also occurrences where process performance is affected by more than one discrete factor. This method will provide the ability to evaluate and select the dominant factors in combination with one another to predict the response.
Many designed experiments fail to provide the desired understanding due to lack of poor planning prior to execution. Participants will understand the critical aspects of designing effective and efficient experiments to investigate non-linearity, impacts on variability, and how to incorporate and evaluate sources of uncontrolled variability during the experiment.
A common issue with many experiments is an unavoidable yet predictable noise factor. Participants will learn how to properly include and evaluate the impact of a noise factor within an experiment.
An efficient method for evaluating the contribution of factors and their interactions on the process performance measure is through the use of a Full Factorial design. The methods for designing, analyzing and interpreting these designs are presented.
When there are a larger numbers of factors to be investigated but a limited budget (time and resources), fractionating the design can provide insight on which factors to investigate more deeply with Full Factorial methods. Participants will learn the pros and cons of fractionating a design and the applicability to the DOE roadmap.
Occasionally there is a need to evaluate the effects of varying levels of multiple factors on a response. Participants will learn the methods for designing, analyzing, and the pros and cons of these designed experiments.
All hypothesis tests require adequate amounts of data to provide insight on the relationships between the factor(s) and the response. However, the sample size chosen can impact the confidence in the decision based on the data. Participants will learn the relationship between the various risk factors, the difference desired to be seen, and the sample size required for various hypothesis tests.
Once the main contributing factors are established, and their interactions evaluated, there is often a desire to optimize the performance of the process. Participants will be introduced and practice techniques for determining the optimum through the use of some specialized designed experiments in a strategically determined pattern. This technique is typically applied to manufacturing processes, but has applications to the IT world as well.
To optimize a single process performance measure could be done at the detriment of other performance measures. Participants will learn how to establish the optimum process settings for multiple factors to maintain goals / targets for multiple performance measures at the same time.
EvOp was developed in the chemical industry for optimizing processes while they are in use, producing sellable product. This methodology can be utilized in any process where radical change in factors is difficult and / or dangerous, and slow steady improvement is desired and supported.
Not all process performance measures are quantitative - many are qualitative. This poses serious challenges for statistically evaluating the significance of factors on the performance without application of some specialized data transformation techniques. Participants will learn the methods for transforming two styles of qualitative response data for the purposes of designed experiments.
The introduction to SPC module (taught in Green Belt) dealt with the most common and most versatile Control Chart. In this segment, participants will expand their knowledge of specialized Control Chart methods for attribute data as well as sampling methods for continuous data.
Lean Six Sigma is focused on changing the ways processes function to make them perform at better rates than in the past. With any change comes resistance at various levels within the organization. Participants will learn some of the common signs of various types of resistance, and methods to aid in overcoming these forces.
Once critical factors and their optimum levels are determined, appropriate and effective methods of controlling the process must be instituted in order to maintain long term, predictable performance. Participants will identify various effective ways to document and control the process to provide closure to the project