|
Lean Six Sigma Black Belt
Program Outline
Lean
Six Sigma Flow Chart
Contact
Information
Program Outline
• Program Participants
• Black Belt Training
• Instructors
• Cancellation and Refund Policy
Program Delivery
Program
Outline
Program
Participants
There will be a maximum of 20 participants for each of the
Black Belt training sessions.
Black Belt Training
A prerequisite for the black Belt training is completion
of the Green Belt program or its equivalent. Participants
will be required to complete a project, which likely will
continue beyond the training. These projects frequently return
$50,000 or more in value to the organization, providing an
immediate payoff that more than covers the total expense of
the Lean Six Sigma training.
The following topics will be presented during the Black Belt
training, building on what already has been covered during
the Green Belt training:
LSS Black Belt Introduction - Projects
This session focuses on the stages and skills for project
management and includes a brief review of the key elements
of Lean Six Sigma. Participants will learn how to manage a
project during its various stages, prioritize project opportunities,
develop the project charter, prepare a work breakdown structure,
determine the critical path for activity sequencing, identify
resources and requirements, and track performance.
Statistical Inference
Inference is a method by which we try to generalize from the
specific to the more general. Statistical inference incorporates
statistical thinking and methods into this inference to help
ensure that it is sound, and to point out any weaknesses in
making inference. Taking measurements on a random sample of
units from a large population, and generalizing these results
to the population, involves statistical inference.
Hypothesis Testing
A fundamental technique in making broader inferences from
a sample of values involves a hypothesis test. For example,
if a change is made in a process, we would want to know if
the change had any noticeable effect. For example we may want
to test if the mean of some process feature (such as length,
or time to delivery) changed. Because of natural process variation,
this question may be difficult to answer. Hypothesis testing
provides a mechanism for testing this, and related, ideas
in a statistically objective manner.
Participants will learn about types of errors when a decision
is made about a statistical property of a population or a
process. They will learn about the meaning of p-values and
how to use them. They will perform power analysis as a tool
for planning an efficient sample size.
Confidence Intervals
Confidence intervals are, in a sense, an extension of hypothesis
tests. A hypothesis test often tests one particular value—for
example, to see if process mean changed at all, we would test
if the mean change might equal zero. In this example, a confidence
interval is intended to provide a range that includes all
the mean changes that are consistent with the data. For example,
a hypothesis test might find that the mean time to delivery
did change. In a confidence interval, we might claim that
“we are 95% confident that the mean change in the time
to delivery was a decrease of 3 to 5 days.” Participants
will learn how to find confidence intervals in a wide range
of practical situations and what the relationship between
a confidence interval and a hypothesis test is.
Correlation and Multiple Linear Regression
When both input and output variables are continuous, these
methods can be used to see whether the input variables can
predict the output. In multiple linear regression, a number
of inputs can be used to predict the output. Participants
will learn fundamental techniques of regression and how to
apply them in process investigations. They will learn about
confidence and prediction intervals, how to correct problems
with a regression model through transformations, and how to
deal with qualitative variables by using dummy variables.
Project Leadership, Organization Learning and Presenting
Skills
Black Belts are often required to not only lead projects,
but also to increase understanding of Lean Six Sigma throughout
the organization and make presentations. This hands-on session
will introduce the participants with the skills required to
not only be a good leader, but also to be able to develop
PowerPoint presentations and deliver training modules. As
homework, the participants are required to develop a brief
training module.
Project Report (Charter)
Participants will present the current status of their projects
using the project template to indicate the tools utilized,
the data gathered and analyzed, and the next steps that are
planned. This also is an opportunity to receive feedback from
the program faculty and to ask questions about the project
and/or tools.
Measurement Systems Analysis (MSA)
Participants will review the qualities of a good measurement
system, such as good operational definitions, accuracy and
precision. They will also analyze data from a number of measurement-system
studies, including the most common studies, gage R&R (repeatability
and reproducibility) studies, and will discuss their plans
for conducting a measurement system analysis for their project.
Components of Variance/Multi-Vari Charts
Multi-Vari charts provide a graphical way to examine different
sources of variation. These sources of variance are known
as components of variance. Participants will analyze a number
of data sets. For each data set they will perform an informal,
graphical, analysis with Multi-Vari charts, and then a formal,
statistical, analysis using a statistical technique known
as analysis of variance, or ANOVA. Participants will also
learn the connection between these analyses and gage R&R
studies, and will learn and use important statistical ideas
such as crossed and nested, and fixed and random, factors.
Reliability Indices
Many MSA’s are performed with continuous data, but when
the data are discrete (such as pass/fail), many standard MSA
techniques can not be used. A method of testing the quality
of a measurement system is introduced for working with discrete
data. Participants will understand how to study discrete data,
how to summarize the quality of the measurements with a measure
called the reliability index (also known as kappa). They will
examine this through a series of data sets that they will
analyze.
Project Status Review/Consultation
Design of Experiments
Control charts and Multi-Vari charts are excellent tools for
passive analyses of a current process. Design of Experiments
is a powerful active technique to improve processes. Participants
will review fundamental ideas about experimentation.
Two-Level Factorial Design
A two level full factorial experiment is one in which each
factor is studied at exactly two levels and in which all combinations
of factor levels are studied. The value of this approach over
standard one-factor-at-a-time methods can be enormous. The
full-factorial approach allows the variability of the process
to be taken into account, while at the same time reducing
its impact and allowing so-called interactions (features of
process complexity) to be measured. Participants will learn
key terminology, when and how to design an experiment in an
organization, and how to analyze and summarize data from an
experiment. Both graphical and numerical techniques will be
emphasized. Participant will analyze numerous experimental
case studies, and design and analyze an experiment in a simulated
study.
Two-Level Factorial Design, Two-Level Fractional Factorial
In the early stages of many experiments, the large number
of factors one would like to study precludes the use of a
full factorial design. For example, to study eight factors
each at two levels would require 28 = 256 runs in a full factorial
design. This problem can be solved by using fractional factorial
designs, in which only a carefully selected fraction of the
total number of runs is made. For example, eight factors can
often be studied quite well in only 16 runs. Participants
will learn the idea behind fractional factorial designs, how
to construct these designs, the advantages and disadvantages
of different fractional factorial designs, and how to analyze
experiments from these designs through a number of examples.
Training Module Presentation
Each participant will present a training module that he/she
has developed, and will receive feedback from the instructors.
Two-Level Fractional Factorial, Response Surface Methods
Two-level designs are useful to gain a good understanding
of which factors are important and which factors interact.
Sometimes more detailed information on these factors is needed.
Response surface methods are useful to learn more information
about these factors when they are continuous. Participants
will learn the value of these designs, see the connection
between two-level designs and these designs, learn how to
design response-surface experiments, and how to analyze them.
They will also learn how to make intelligent tradeoffs among
several responses using the method of desirability functions.
SPC/Control Charts
Participants will learn the fundamentals of control charts
for variables (continuous) data and for count data, including
how to set up control charts for their processes. The effect
of size, sampling frequency, and subgrouping will be illustrated
with examples. Alternatives to classical control charts when
productions runs are short, or defects rates are low, will
also be discussed.
Capability Analysis
Participants will learn to calculate metrics for potential
and actual capability of a process from a sample, or from
control charts, for the situation when the process is in statistical
control. The difficulty of conducting such an analysis on
a process for which special causes are still present, or for
a process whose distribution is non-normal, will be discussed.
Project Reports
This is the final presentation of the project. Participants’
supervisors are invited to this session. For the presentation,
participants will present the actions taken following the
DMAIC process, the Lean Six Sigma tools that were used, the
data that was gathered and any analyses that were performed,
the improvement strategies that were developed with the resulting
financial benefit, and a plan for any steps that remain to
be taken.
Instructors
Cancellation and Refund Policy
Full tuition is refunded if cancellations are made more
than 30 days before the program begins. A $50 fee
is charged for cancellations made between seven and 30 days
before the program begins. Refunds are not given for
cancellations made less than seven days before the program
begins or for nonattendance or withdrawal after the program
begins. Substitute participants are welcome. RIT
reserves the right to cancel programs, substitute speakers
and modify content. Program fees will be refunded when RIT
cancels a program.
For additional information
contact:
Greg Evershed
Director of Business Development
KGCOE
585-475-5442
greg.evershed@rit.edu
Donald Baker
Director
CQAS
585-475-5070
ddbcqa@rit.edu
[ Back to top ]
|
