six sigma all
Walter Shewhart
Father of Statistical process control
Pros of using complaints
Feedback on unmet needs and requirements are very specific
Defects
Anytime you make a mistake and have to complete again
S
an esttimate of standard deviation. essentional the same thing as regualr standard deviation
operational definition
anything the team can identify and relate to as something to measure ( a starting point and stoppoing point, target)
Tracking Costs over time
companies can track figures, they should see prevention costs go up and then level out. internal/external failures and appraisal cost come down over time.
S
standard deviation
nominal is best
target is some middle value
smaller is better
zero is the target
External Failure
Customer Complaints, Penalities, Liability Suits, Repair costs, time spent evaluting things that have gone wront, time spent responding to upset customers
Data Observations
Dta obsertsions provide the most unfiltered and valuable data regarding customer needs and requirment . Think of it
Value Add Rule #1
THe first rule of value is that it has to change the form or the function
root cause analysis
root cause analysis is a tool to understand the underlying causes of a problem. not the one cause. The root is the underlying cause, it is below the survface and is not obvious.
samples
samples are a subset of populations. Good samples are randomly selected and are of sufficianet xie. in order to understand the population as a holw
Interquartile Range
shows the percentage of data fails within the certain range. when using mediums is not affected by the extreme values in the data set
E
sum find the sum of everything on the o ther side of this symbol
Once the purpose or use of the flowchart is established, the next step is to select the best type of diagram
to represent it. This could be a basic flowchart or a swim lane flowchart.
Data collextion Plan
what type of measer are we looking at, what type of data
alternative hypotheis
what you are testing for or what you think may happen
histogram
"A Histogram is a bar chart showing the frequency of an outcome. In Six Sigma, we can use a histogram to visualize what is going on. A Histogram can reflect the voice of the process.
Rule 1
"Never put specification limits on a contrl chart Why? Because if a point is above the upper control limit and below the spec limit, then it will likely be ignored as it is within specs.
Measuring more than the project Y
"Only monitoring the project Y can lead to the creation of a defect
Dr Genichi Tagushi
Approach to design of experiments. the wau to measure quality is through finicial loss it creates for customer, company & society. quality should be engineered into the product during the conceptial design phose
Rule 2
Never construct control charts from inspection record. Inspection recards are what happened in the pst, they are not things that you can change now ACT NOW!
process owners
influence over improvement team . accoutable for ts completing action items subject matter experts formal - supervisor or manager or informal leader- . can also be sponsor
causation
instead of dependedent and independent varibles we can aso say that x causes y an y depends on x this is causation
select solutions
selct the best solution, based on criteria important by customer, company, project, should be driven by data
Pp and Ppk are potential studies
showes performance past or present.
standard work
standard work is the ideas of having documented standard practices to that things are complerted the same way every time, no matter who is working on it
basic rules
start by considering how much data is available
rely on the team
team mebers should have sufficiaent knowledge and experience to rule out issues.
Champion and sponsor
"
Flat Distribution
"
Ordinal Scale
"
measurement scale charet
Scale Central Location. Dispersion. Significance test
SMART Goals
Specific Measurable Attainable Realistic Time Based
Mean
The average, middle of the curve, equal amount ofdata on the right and the left
Waiting
Waiting for something
Key rol& responsibilities
White belt, yellow, green, black
baseline performance mesaurement
can look at baseline of sigma
n
sample sizeo
Ratio Scale
"
Survey Cons
"
bi model distribution
"
flow chart
"
hypergeometric distribution
"
process out of control
"A Control Chart Indicates a Process is Out of Control When:
np chart
"np Chart
defects per million units
1. find defect rate 2. Get DPU. uSe sigma table. use calcualter to confirm sigma
Is the histogram normally distributed
A normal distribution indicates that random variotion is operating in the process which is different than when something systematic is occurring
Data collextion Plan
A plan to gather meaningfull data, What is meaningul is determined by peformance indicators
Internal Failure
Accidents, Work-arounds, Quick Fixes, Scrap, Rework , Defects, Mistakes, Employee Turnover, Equiptment Downtime, Sorting
Tollgate Review- Map the process
At the map the process tollgate review meeting the sponsor wil want the follwing
DPMO
DPMO = total defects in sample/ (total opportunities X 1,000,000)
Control
Explores how you can maintain the improved process
Value Add Rule #3
Ge it right the first time
Benefits of project charter
Helps keep everyone on the same page. is a point of reference or set of ground rules. Can bring everyone back whn they start drifting in the projet.
Potential root cause
Identify anything that might be contributing
Narrow root cause
Identify root cuases -largest contributer to problem, cant fix everything
fule five
If two out of three points are in zone c that may indicate special variation
Etra Processing
If you spend extra time inspecting & not adding value to the product
Obtaining the voice of the custemer
Interviews or focus groups require a company representative to ask prescribed questions to current or potential customers.
• Level I:
Macro process mapping
PP
PP is the voice of the custerom compared to the voice of the process. the equation for this metri. Pp USL-LSL/ 6XS
The Quincunx
Process generator that shows varition in the process. Red beads represent opportunities for data collection. Array of studs is the origi of variation in process ( the 6M's) Red slider at the top shows the centering of the process. Measurements need to be taken in order for these to be visible. Th bottom shows the variation created by ata collection efforts.
Overproduction
Produce more than you need or deliver sometning sooner than needed. - need somewhere to store product
types of data
Qaulitative vis quatitative
Dr. Joseph M Juran
Quality COntrol Handbook published in 1951 - detailed quality movment & tools
RTY
RTY= FPU1 X FPY2 X FPYn
SMED
SMED stnad for single minues exchange of dies. it is a form of quick changeover
map the process
Specify the steps to deliver the product or service
Inventory
Storing inventory costs money & space. if something happens to it you will have defective product
Lagging indicators
The outcomes of those leading things that go on the upstrea. example stepping on a scale
The SIx Blocker
The six blocker is a popular wau to present an entire charter on one page
tollgate review - continual improvement plan
This is the end of your six sigma projet. continual improvment is a core alue. it is what process improvemnt is all about. the sponsor will want the follwoing . Priorized list of future improvemtns that you are going to mak. some dates and goals for things you are going to keep continually improving. Commucate results . review lessons learned .
Leading Indicators
Upstream in the process. example diet and exercise
Use DPU
Use DPU when the cost of rework i high and the multple defects per unit are possible
Shewarts rules
Walter shewhart is know as the father of statistical contrl. Shewahrt and his colleques created egith rules to elp dtermine if a control chart variation was common variation or random cause varition
termination milestones
When termninating a project you also need to set some termination mislestones and take into consideration the necessary budget and cost allocations.
formula
see page 28. t = xbar 1 - xbar 2 divided by Sx1,x2 * square root of 1/n plus 1/n2
scatter plt
a scatter plot might tell us if a possile root causes has arelationshp to the data
quality characterisics realive to your quality system
certification- you know you are certified in a particualar system , is ist up to spped, are your internal audits and registrar audits performed in a timely manner n.
granularity
collect dta in its most granualr form helps set up for analyze. for continuous data need at least 30 data points for discrete data need about 100 data points
Contiual improvement plan
continual improvement is one core concept of quality. develp plann. do not stop
data collection
crare a data collection plan gather only the data that is needed
indirect
data comes from internal third party - compaint dept, service department , sales people
Key steps to hypothesis testing
defind the situation or problem you want to examine. State your null and alternative hypothesis. state the level of significance . selecxt the proper tool for analysis. collet data using that tool. calculate your test statistic look up your critical value. compare your test statisic to your critical value. make a conclusion
eatimate of standard deviation
see page 11-13
performance indicator
direct, idirect , third party
Sigma calculator
found on line or statistics program.
green belt working with process owner
green belt should aviod coming across as process owner boss- good relationship is whenn process owner knows green belt is there to help goals and timelines communicated . engage process owner early in define phose
defects per million opportunities - DPMO
opportuniteis for defect 2. defect rate 3. DPMO use six sigma table or use calulator
CP
page 12
Cpk
page 13.
Map of the process
picture for map - needs to be at the appropriate leve for the problem to be discussed. Spioc Chart. A quick and concise expplanation of the current process , evein if it is damaged. Early indicattors of what some of the causes might be Any modifications of scopeq
X
point of data
Process capabilit Pp/Ppk
process cabaility can be used to dtermine past perforance, present performance nad future performance
Rate impact high medium low
rate importance 1-5. Rate impact High Medium low.
determine the time fame you want to anlalyze
relevent to problem.
Using the formula
y = f(x) Y is the dependent variable, x is the independent varible F is the function x goes into the function and comes out as y
types of root cause tools
"Types of Root Cause Analysis Tools:
Voice of prcess in dmaic
"Use the voice of the process in each phase of DMAIC:
central tendancy
"Used in basic data analysis.
z scores
"We use Z Scores to transform a given standard distribution into something that is easy for us to calculate probabilities on. Why? So we can determine the likelihood of some event happening.
Quinex demnonstration
"What isn't obvious at the first is that it helps you understand the danger of adjusting your process while in the moment to the center. Watch the demonstration below. Doesn't that make it clear how specification limits are used to monitor and then control variation?
benefits of step/leaf plot
The benefits of a stem and leaf plot include the concise representations of the data. The ranges, mins, maxes, gaps, clusters and outliers are easily identified. With extremely large data sets, this tool is valuable to ensure that the data used will be preserved.
X-bar
the mean of the data - anything with a bar over it is the average of that thing
nominal scale
"
swm lane map
"
third party
"
what if z score is off the chart
"
Closing out the project
" . Extinction
measjre
" Measure
pareto chart
" Pareto Diagram or Chart is a bar chart that re-orders the categories so they are rank-ordered from the largest total occurrences to the smallest.
Terminate checklist
" Plan a closeout meeting
how to create the pareto chart
"1. Define the problem or opportunity
6 m examples
"6 Ms in Manufacturing Example
Ishakawa 6m
"6M's of Six Sigma
binomial distribution
"A Binomial Distribution only has 2 possible outcomes, including replacement. Ex. Heads or tails.
the box and wikser plot
"A box-and-whisker plot is a graphical method to summarize data from multiple sources and display the results in a single graph. This tool visualizes the minimum value, twenty-fifth percentile, median value, seventy-fifth percentile, and the maximum value including any potential outliers. Box-and-whisker plots are very effective and easy to read. The benefits of box-and-whisker plots include allowing for comparison of data from different categories for easier, more effective decision-making.
defext location check sheet
"A defect location check sheet, also known as a defect map or a measles chart, is a structured form for collecting and analyzing data that provides a visual image of the item being evaluated. A defect location sheet is used when a visual and simple count is needed to determine the defectsA common practice in quality assurance and Six Sigma is to count the number of defects found. A visual cue such as a red dot would be placed on the image indicating a defect.
The respons Plan Purpose
"A response plan creates an ongoing action plan for the process. It explains where to go and what to do. A response plan:
run chart
"A run chart displays observed data as they evolve over time. Just a basic graph that displays data values in a time order. Can be useful for identifying trends or shifts in process but also allows you to test for randomness in the process.
Review & Feedback
"After the first round of interviews are conducted, process maps are reviewed for clarity, content, and continuity. You should always validate a process map before using it. A final round of feedback should be conducted to ensure the process map is valid.
Types of charts
"An attribute chart is a kind of control chart where you display information on defects and defectives. Helps you visualize the enemy - variation! Follows a philosophy of detection, not prevention; thus it is NOT really, truly part of Six Sigma which focuses on prevention.
when to use the pareto chart
"Answers the question: what are the critical few things driving performance in a process?
c chart
"C Charts
check sheet
"Check Sheets are a data collection tool that usually identifies where and how often problems appear in a product or service. It's specifically designed for the kind of process being investigated. Both variable and attribute data can be recorded on a checksheet.
introduction
"Collecting good data is an important part of the decision-making process. We need to ensure that we are collecting only the data that we actually need and that we are collecting it in a deliberate manner. This data helps us to understand the baseline performance of our process as well as the magnitude and frequency of our defects or issue.
Common Cause Variation
"Common cause variation are the usual, historical, quantifiable variation in a system. The best treatment is to look at all of the data available for the process and try to gain a better understanding of the system. You could then make basic, impactful changes to your process that would improve the whole.
continuous data tools and examples
"Continuous Data Tools & Examples
control
"Control
control chart vs run chart
"Control Chart vs a Run Chart
control limits
"Control Limits
control chart
"Control charts are a great tool to monitor your processes overtime. This way you can easily see variation. Control charts are a great tool that you can use to determine if your process is under statistical control, the level of variation inherent in the process, and point you in the direction of the nature of the variation (common cause or special cause).
internval scale
"Data that can be arranged by order and distance but is without origin. Differences in value are considered meaningful. For example, the difference between 20 to 30 degrees Celsius is ordered and qualitatively different in its measure of temperature. examples time between calendar date , average temperature
Types of data
"Discrete Data
gage repeatbility and reproduciblity
"For a product or service to work at an optimum level, its processes must be able to be repeated and reproduced. A gage R&R will answer the question: what percentage of the variation in the measurements is due to the measurement system as opposed to some other source? A gage R&R checks precision in the measurement system. It does not, however, check the accuracy of the measurement system. Students of Six Sigma must make a distinction between repeatability and reproducibility.
Sipoc Chart
"Identify the outputs.
Common Cause Variation vs Special Cause variation
"If you were to treat common cause like special cause, you would miss the trend of variation that is in your process and thus miss the opportunity to improve it over the long run. For example, a recurring problem of late deliveries all treated like a special case would cause you to miss out on the root cause of the problem.
improv e
"Improve
Cons of Research outputs
"It is very expensive
Cons of using complaints and kudos include
"It is very reactive; you must wait for a problem to arise before understanding the data
vilfrado parado
"Italian philosopher and economist
macro process map level 1
"Level 1: Macro Process Map
level 2 map
"Level 2: Process Map
how do you know if your process is following normal distribution
"Many of the analysis required in the Green Belt and Black Belt Body of knowledge assumes a normal distribution. There's a good reason for this; normal distributions have a nice, symmetrical shape that makes working with them much easier (like in an ANOVA for example!) But how do you know if your process is following a normal distribution? Use these tests to find out.
outline for data analysis
"Outline for Performing Data Analysis
Process mapping 2
"Process mapping is a technique used to create a visual representation of an existing or hypothetical process. Flowcharting provides a sequential view of the tasks and activities needed to carry out such processes. One of the first things to consider when designing a flowchart is its purpose.
normal distribution 2 standards from mean test question -= 95
"Question: For a normal distribution, two standard deviations on each side of the mean would include what percentage of the total population?
purpose of a control chart
"Question: The purpose of using control charts is to
importance of the response plan
"REsponse plan is designed to get thingsdone ins a systematic and quick way so that you do not havetoredo the project.
root cause analysis
"Root cause analysis is a collection of tools and processes we can use to determine the most important causes for an issue we are trying to resolve. This is an important function as one of the top 5 reasons for project failures is poor root causation / no root cause identified. If we can identify the root cause to an issue, we have a good chance of solving it.
Market Research Outputs
"Sales forecasts
sampling without replacement
"Sampling Without Replacement is a probability question that is an example of using the Hypergeometric Distribution.
continuually improve
"Six Sigma courses, one of the basic beliefs about quality is that the relentless pursuit of perfection, meaning continual improvement, is crucial. So that tollgate has been built into Lean Six Sigma, and it is an explicit action that you must undertake. Do not just stop the process after the Control phase.
standard deviation
"Standard deviation is used to measure the amount of variation in a process. Standard Deviation is one of the most common measures of variability in a data set or population.
Normal Distribution
"Statisticians refer to the normal curve as the Gaussian Probability distribution, named after Gauss. The Normal distribution is used to analyze data when there is an equally likely chance of being above or below the mean for continuous data whose histogram fits a bell curve.
3. Calculate the controle limits
"Step three of constructing a control chart is to calculate control limits. First, you need to calculate the grand average, or the average of the averages; these are also known as X-double bars.
Resistance
"Termination of a project requires a plan; otherwise, resistance and mismanagement can derail the entire process. The resistance can come from both the team and management - the team has a lot of anxiety about the future, and management does not want to be seen as a failure. And mismanagement can make things much worse than if you had just had a plan in the first place.
Quinex
"The Quincunx is a wonderful teaching tool as it is a great way visually to show the concept of randomness and with enough repetition, a bell-shaped curve. Obviously it helps you see the concept of the central tendency.
benfits /weakness of defect location sheet
"The benefits of a defect location check sheet include being able to collect data with minimal effort. The weakness of this type of checklist is being able to act upon the amount of errors found.
7 Elements of project charter
"The business case
Market Research
"The market demographics of an area
Truncated distribution
"The next histogram has a truncated distribution. It looks like someone threw away the left and right tails of the histogram, or as if some of the data is missing. This is not a normal distribution with random variation. A couple of things could have happened here:
normal probability plot
"The normal probability plot is a graphical technique for normality testing: assessing whether or not a data set is approximately normally distributed. The data are plotted against a theoretical normal distribution in such a way that the points should form an approximate straight line. Departures from this straight line indicate departures from normality.
step and leaf plot
"The purpose of a stem and leaf display is to clarify the shape of the distribution. A stem and leaf plot is a table where each data variable is split into a stem, which represents the first digit, and a leaf, which represents the last digit.The stem values are listed down and the leaf values go right or left from the stem values.The stem is used to group the scores. Each leaf shows the individual scores within each group. For example, assume we have heights of a group of two-year-olds. The heights are as follows: 2.3, 2.5, 2.5, 2.7, 2.8, 3.2, 3.6, 3.6, 4.5, and 5.0, For the value 2.3, the stem is 2 and the leaf is 3. Regarding the heights of the two-year-olds, the stem and leaf plot is as follows: for the stem, there are 2, 3, 4, and 5. For the leaf, there are 3, 5, 5, 7, 8, 2, 6, 6, 5, and 0. The stem and leaf plot is indicated in the image at the right.
Sipoc Chart
"There are different ways to map a process. One popular method in Six Sigma is known as a SIPOC (suppliers, inputs, process, outputs, customer) chart. Steps to create a SIPOC chart include:
constructiong control charts
"There are six steps to constructing control charts:
Survey Pros
"They are inexpensive
Interviews/focus groups Cons
"They are relatively expensive in both money and time
Level 3 map
"This flowchart is sometimes known as the "improvement view." It provides the maximum level of detail and identifies the location of root causes of problems. At this level, one looks for redundancies, workarounds, and any variations from standard operating procedure.
Process mapping 1
"When mapping the process, we want to capture the second perspective, or what is actually happening. We can't determine where the problem is occurring and what is contributing to that problem if we do not map the process as it actually is. We want to capture everything—the good and the bad. The other two perspectives are biased and could distort the picture. Additionally, there needs to be a process focus. That is the only way to understand how and why work gets done. It helps characterize our customer and supplier relations and forces us to see the process "as is" from start to finish.
Interviews/focus groups Pros
"You are able to access nonverbal cues such as body language and tone
Market Research Inputs
"You get a large amount of quantitative data
mapping tools
"here are some basic mapping tools that we can start with and add to if necessary. At a minimum, use one map that provides a start and stop point, and one that traces all process activities to the end. Use different mapping symbols to help move through the process:
6 M's
"n the Analyze phase of DMAIC we want to identifying root cause to the defects that are causing variation in our process. Sometimes the culprits are easy to identify. More often we have to make a hypothesis about what is causing the variation and then test that hypothesis to see if our hunch was right. But how do you come up with a hypothesis for what is causing variation?
p chart
"p Chart (proportion chart)
Team memebers
"sometmes you can choose your team members; sometimes you cannot. Team members should include people who actually touch the process. They can include internal staff, direct customers who are the recipient of the process output, or suppliers.
Conclusion of the control phase
"t the end of the Control phase, you should have a plan for monitoring the process to make sure it works well, new procedures and processes in place to keep it running well, a list of items that were not addressed in the Analyze and Improve phases, and, finally, the continual improvement plan.
u chart
"u Chart
Cons of using data animation
"you might observe problems that customers might not recognize as problems
1. Project Charter
- Business Case - uses nonquantifiable terms. Describes how the project is giong to be good for the businessk, how it impacts the key business objectives & why the project is worth doing
80/20 rule
*0 percent of the wealth belongs to 20 percent of the populiation. 80 percent of porcess performance or issues are driven by 20 percent of the activity
pt ratio
0.10 or less which is good, 0.30 or greater whic hs not acceptable. in between we need to make arrangements
calculating DPMO
1. Determine the number of defects ( D). 2. Determine the number of units processed or services offerend (U) 3. Dtermine opportunities per unit (OP) 4. Calculate the total opportunities (TOP) 5. Divide the total defects by the total opportunities (DPO). 6. Multiply that answer by 1 million
Improve
1. Generate possible solutions 2. select solutions 3. implement solutions
Tolgates of control
1. Monitor and control plan 2. Response plan 3. Continual improvement plan
Measure
1. Performance indicator identification - upstream leanind indicators 2. data collection plan 3. baseline performance measurement
Analyze
1. Potential root cause 2. Narrow root cause 3. Critical root cause
calcualting the standard deviation
1. find the man of the data. Subtranct the mean freom each data point. 3. square those numbers 4. add up the squared numbers 5 Devide that wole sum by n-1 6. take the square root
tollgates of analyze
1. potential root cause 2. narrow root case 3. critical route cause.
Define
1. project charter- reason for project, 2. customer needs & requitements 3. Map the process use sipoc
Calculating the interquartile range
1. sort the data from low to high. 2. find the middle or median of the data set. if there is an odd number, eliminate that number. if there is an even number just devide the data set in ahlf 3. find the medium of the upper half q3 and lower half q1 of the data the medium is the simply the middle value. Subract Q1 from Q3
creating a kano model
1. the voice of the customer it is critical that you drive your decisions based on what the customer is saying. lisen carefully to what is communicated verbally and non-verbally 2. be sure you identify the proper category for their requirements dissatisfier, satisfier , delighter. 3. prioritize your time according to the categores identified . you deal with the dissatisfires first and then the satisfiers. if you do not habe the dissatisfires all taken care of and a number of ther satisfirest then having a delighter will not help.
sux sugna
3.4 deffects per millon opportunities
Baseline performance measurement
A before icture. Allows ou to show mangement the progress made on a project
planning for continual improvement
A continual improvement plan is a long term plan that incudes multiple improvement projets all aimed at a common goal. Theseprojects keep the team focused on short term and long term goals nd set the expectation for continually improving.
Statistacal Difference
A difference or a change to he process that probably, with some defined degree of confidence did not happen by chance. We are looking for a statistical difference when we use hypothesis testing
what is a process
A process is a repetitive and systematic series of steps or activities where inputs are modified to achieve a value-added output. The intent of each task or activity is to add value as perceived by the customer. We want to create something that has value-add so that a customer will buy it. We look at each step, task, or activity to determine whether it adds value. Does it change the input—form, fit, or function—into something that the customer will consider valuable and be willing to pay for? To find out, we need to adequately map the process. Mapping the process helps determine where we are spending our time and what we are doing to create what the customer will label as value-add.
The project charter
A project charter is lika a contract terms and onditions, explicitly states expextations , clarifies what the projext is Defines the support needed in order to complete the projext. Is a living document that changes only when the projext requires it
Project Charter
A project charter is like a contract with terms and conditions. The project charter includes everybody's expectations for the project, clarifies exactly what the project is all about, and defines the kind of support the project is going to need in order to reach its goals.
Respnse plans and defects
A response plan outlines those processes that need to be followed when there is a defect or out of control condition that is being tracked in monitoring plans. These include Out of control point on control chart . Non Random behaviors are present. There are failures within the check sheets. Check sheet failure. Automation Failure
Rule One
Any data point that is pas the upper or kiwer control limit is probably special cause variation. Most of your data 99.73 percent of it , should be in zone c
Tollgate Review
At the customer needs & requirements tollgate the sponser wants the following information . A description of the customer grop and speixfic segments that are being impacted by the probleme identified in the charter. - bring supporting documemnts, customer survey , market reports, customer interview. anytning foung out up to the tollgate mpacted the charter. tell the sponser scope, contracting anytning imperative to tell the sponser.
Tollgate review- the project charter
At the project charter tollgate review , it is expected the greembelt will: follow the agenda for the meeting. Provide a lear conncise and conplete charter document. the charter document should be sent to the poject sponser a few days ahead of time for his or her review. Proesent a compelling business case that shows why this needs to be a SIx Sigma project and wht will happen if it is not done. Focus on issues that directlly impact custoers and/or will either safve the omany money or prevent it from losing money. Provide a clear problem statment and well defined scope statment. The buisness case needs to be soldig th the green belt needs to tell the porjectsponser what her or shee nees to get informatin that will be completed. The green must take time and effort and create a clear consise charter.
tollgate review - response plan
At the response plan tollgatrereview meeting the sponsor wil need: What things could possibley go wrong with the process. what might happen thatcould derali the performance of the new process. A documented list of prioritzed issues. Prepaarations for how to deal with those things when they happen. A failure mode and effect analysis. (FMEA) would be an excellent things to use to help you do that.
Apraisal Costs
Audits, Testing, Inspections, Equiptment calibration
cellular design
Cellular design allows you to get things closer together and be able to flow from one step to the other without poles of inventory in between. it helps reduce transportation waste and waste of inventory by more quickly identifying ptential defects
Dr Willaim Edwards Deming (1900-1993)
Champion the work of Walter shewhart. used shwharts tools such as statistica process control, operationl definitions and envolved vesion of the shewart cycle called Plan-do study act or PDSA
Data Collection Plan
Clarify how data will be collectied. Clear defintions/procedures
Normal distribution: In the Middle
Classic bell shape and falls right down the middle of the customer specifications indicated by the dotted lines. It has some wiggle room on either side and is not in danger of breaching the upper or lower specs. Thi is the type of distributin we like to see in six sigma
Complaints and Kudos
Complaints and kudos use data to determine common problems customers face with a product or service. Information is gathered through various means such as returned product information, customer letters, customer phone calls, and customer reviews. Companies may have customer management systems that collect and track customer complaints and kudos.
CPR
Consistent, predictable, and repeatable- targets with as little variation as possible around those targets
Constraints
Constraints are things that could limit the project's progress or derail it entirely. Adding constraints to your project charter gives your champion and sponsors a heads-up on potential project roadblocks. It also lets your champion and sponsors know whose help you might need; you may need IT's support in order to collect data, and you may need the help of the financial department because they know where the money is. Types of constraints might include lack of adequate resources or unavailability of people, resources, or raw materials.
Define
Covers the project charter , custmer needs and requirements and process mapping
Value Add Rule #2
Customer must be willing to pay for it
Features
Does the process provide what the custemer expects and needs, how do you know, what are you using to help you understand customer needs
Defects Per Million Opportunities - DPMO
DPMO is a good met ric to use for more complex products and services
DPU
DPU = total number of defects / numbers of units processed
DPU VS RTY
DPU considers how many total defects exist in a given number of units. If multiple defects casn exist in one unit the calculated DPU can be greater than one. RTU only considers that the unit was defetive , regardless of how many defects exits. A unit with one defect can have the same impact on RTY as a unit with 10 defects.
Pros of using data animation
Data observations provide the most unfiltered and valuable data regarding customer needs and requirements. Think of data observations like anthropologist Jane Goodall studying chimpanzees; you are studying your customers and their behavior
Normal Curve, normal distribution, bell curve
Data that is distributed normally exhibits very specific characteristics. These charateristics allow us to make predictiosn or conslusions of outcomes for the population
DOWNTIME
Defects , Oveproduction, Waiting, Non-Utilized Talent, Transportation, Inventory, Motion Extra Processing
Conclusion
Defets per Unit- PU total defects oer total units producted , simple metric for assessing the cost of poor quality, doesnt account for poor quality. Rolled throuput yield (RTY) the yield of process multiplied together - only acdounts for defective products or services , not if the product has multiple defets per process step. Defects Per Million Opportunities (DPMO) accounts for opportunites for failure. Allows an organization to compare products or services varying complexity across the same process steps.
The most important phse
Define is the foundation for everything else. There are a lot of projects in the real world that go wrong right out of the gates, but SIx Sigma. Letting projects happen where management never supports it , or they support a bad idea without a business case. Letting the voice of the business drive the actions instead of the voice of the customer. Having no process of knowing what the process is so improvements are just made at random
Customer requirements
Define what customer wants, voice of customer, specific & detailed terms
DMAIC
Define, Measure, Analyze, Improve, Control
Key Metrics used to monitor a process
Definew effectieness or success- the metric eliminates thequality issue in the beginning. Identify records that need to be monitored - the type of data collected needs to be capable of detecting the quality issue should it reoccur. Identify the minimal sample size and determine the length o fthe observation period. Verify that the change did not introduce new concerns - this metric has a direct relationship to the measure phase
Improve
Describes how you can improve that condition
Measure
Descripbes how to identify the right thing to measure how to create a plan to collect that data, and how to baseline the process
Descriptive Statistics
Descriptive statistics Describes what data looks like
Dr Kaoru Ishakawas
Developed the Ishikawa diagram referred to as fishbone diagram
rule eight
Eight points in a row on either sie with no points in Zone A could indicate special cause varation. Expect 68 pecent of data in Zone A
Prevention Costs
Error Proofing, Capability Studies, Procedure Writing, Education and training
Business case
Every charter starts off with a business case. The business case justifies why the prjoet is needed and needed at this particular time. Aligns the reason for doing the project with the organizations strategic objective. Justifies why the project is needed and needed at this particular time. Alignes the reason for doing the project with the organizations strategic objectives. Identifies a specific business goal that will be impacted by completing the project. Identifies the consequences if the project is not completed should be 2-3 sentences. Needs to gcrate a sense of urgency; express urgency in terms of money. If urgency is not attached to money then none of the senior supports wil buy into it
Analyze
Expains how to come up with a long list of possible root causes to problem, narrow tha list down to a short list, and isolte it to one or two root causes
The cost of Poor Quality
Experts maintain that up to 35 to 30 percent of annual revenue is lost to cost of poor quality
Control
Finalized control systems are determined and long term capablities are verified for sustainalbe and long term sucess. Understand how process will continue to be measured once implemented. this can be performed through statistical process control and contrl pans that determined who and what should be meausred. Standardized process put in place so that old way of doing things do not happen agina. Improvements are quantified to make sure the goals of the project were achieved.
During Tollgate Review Meeting
Greet sponsor, remind which proect is being discussed review prior phases/tollgages. Mention updates Inform sponsor of issues, changes to proejct charter. Brifeler review current information. What the goal and tollgate were, what you planned on doing & what was done, tools & findings. Answer questions. Reuest approval to move to next tollgate. REquest resources if needed. Sponsor approves. thank sponsor
steps to crearte a continual improvement plan
HOw do you create a continuual improvement plan 1. determie the true north, a defined goal. 2. break down the plan into short term medium term and long term solutions. 3. determien leading and lagging indicators so you can make sure the improvements you are making are getting the desred reults. 4. decide on a cadance of progress reviews with leadership meaning choosing daily, weekly or monthly reviews
2 Customer needs and requirements
Identify who the customer is
hypothesis testing type 1 and type II errors
If I rejext the null hypothesis and the null hypothesis was actualliy wrong I should have rejected its- type II error or beta error. If I reject the null hypothesis when in fact the null was true- Type 1 error or alpha error. Type 1 errors can occur due to just chance alone, badluck
Waste
If it is not adding value then it is waste - Eight ypes of wast- Downtime
Ppk
If the cetner is closer to the lower specification- Xbar - LSL/3XS
Control chart
Is the data behaving normlly - display the data on a control chart. Allows us to see process over time . Provides more predicability than raw data. are there any trends, are any points out of control?
General electric
Jack welch CEO committed to six sigma in 1995. formalized aas an expectation of every employee. focused on the cutomer and intensified its stategic elements. created a si sigma culture. every employee was trained and apointed one full time team leader for every 100 employees 40 percent of manager bonus was connected to involvement with six sigma. Had to serve a six sigma black belf for promotion. Added D forDefine. the new methodolgy DMAIC is integral part of six sigma.
Critica to Quality characterisics- CTQ
KEY Measurable Charachterisics of a product/service process whose performance standards are setblished by the customer in order to satify the customer's needs. CTQ can e dificult to define if you do not have measure to help you understand.
Imporance of process Capbility
Knowledge of the process determined by process capabilit. Allows product develpment engineers to selct the right equiptment for a proposed design. Provides the abilit to scheule jobs better with the right machines. Lowers work in process. Reduces stoppages. Ensures that process problems are visable and quatifiable. provides means to communicate with the customers. helps develop realistic tolerances based upon actual performance. Helps set pricing of products
Allied Signal
Lawrence Bossidy becaue CEO of Allied signal in 1991 - used six sigma as business enabling tool . used six sigma to improve operating margin. INcreased operating margin from sigle digits to 14.1 percent Each projectwas handpicked by an executive trained in six sigma. three core processes - the people process, the strategy process, the operations process
Leading and lagging indicators
Leading indicators are called the X's - these are checkpoins along the process. these should ber reported frequently to ensure stability without the X it is difficult to know if the y is being accomplished
what are we looking for when using the Quincuns
Look for bell-shaped curve. Most values are in the center and fewr tail out to either side. represents randomness. if the curve is not bell shaped then there is opportunity for improment
6. Act on what the control chart is telling you
Look forshifts or trunds in the upper chart. a trend is seven points in a row upward or downward. a shift is seven points above or below the averate.
Identirying customer groups
Make a list of customers, segment into groups, prioritize grops based on what customer says is the most important characteristic to tehm
what are our cririca to quality characteristics- Quantititive method
Measure the number of corrective and preventive actions that we put in place this is the quantitiative method
Measure Phase
Measurement systems are developed and evakluated. CUrrent process performance measured. Key performance indicators determined. Data collected. Quality of data checked for accuracy. Distributions of data can be determed to see how the process is behaving. Amounts of defects and variation present are determined.
Measures of location/Central Tendencies
Measuring the mean, median or mode of the data set , this is also called central tendencies
Medium
Middle number in the data set. To find the medium put the numbers in order from least to greatest or greatest to least. then count an equal number of times from both sides untiil you find the midde example 2.3.4.5.5.6.7.8. Medium is 5. Use the medium whebn you have an outlier in the data set ths mediumat would radically change the mean. Real estate often use
Defects include
Mistakes, Flaws, Customer. complaints, warrenty returns, Liability exposure, all non-qualitu otes
Motorola
Motorola -1985 - trademarked by motorola. Bill smith memo focused on relatinship between product field life and rework in manufactoring
Companies instrumental to six sigman
Motoroloa, Unisys, texis instruments and ABB, allied signal, general electric
Transportation
Moving things, changoing location - but not changing the products form or function
The voice of the customer
Needs of the customer are the most basic problems that customers are trying to solve REmember nees changes. THings that cause needs to change include new problems, external pressure, new compettion , customer awareness and understanding of products , markets or competitors, new needs
Rukle two
Nine ponts in a row on either side of the mean is called a shift - investigate why the shift occurred , it could be special cause variation
Normal distribution: Off to one side
Not right down the middle. All the data is within the customer specifications but the distribution is nearing one of the sped liits. this indicates a defect might occur soon
Six Sigma and Statistics
Numerical summary of data so that we can analyze it, provide the asis for making inferences about the future, provide a foundation for assessingour process capability and provid common language to describe our process
Determants of quality
On the product side: Performance , features, conformance to guidelines or targets, timeliness, reliability , servicibility. On service side- competence, reliability , accuracy , tiomeliness, responsiveness, access
The next phase of DMAIC
Only carry the few critical issues needed to make improvements. there will rarely be just one vital root cause. hang on to the root cause you didng identify as the criitical few. we will revis those later in the control phase. dont' throw it out becuase you are not focuin on it
Villfredo Pareto
Pareto Chart, derivative of the pareto principle developed late eighteen hundreds. The pareto principle identifies the separation of significant few the 20 perent from the trivial many 80 percent.
Pareto Chart
Pareto chart is an excellent way to cisually graph large amounts of data. they are used to answer the question what are the critical few things driving performance in a process
Non- Utilized Taelnt
People who are underemployed or overemployed
Green Belt
Pivital role in six sigma project. keep team on task. facilitate six sigma tools . mentored by black belt coach for first 3 projects. dedicated part time to six sigma projects
Population
Populations are comlete collextion of people things or data that you can anllyze and make conclusions on
Sponsors Resposibility at meeint
Provide gidance insige & suggestions, Key ey on goal and direction of team. Provide support approve tollgage stop project
Project Charter
Provides a business case, problem statement, scope of project. Critical foundation step of process.
pull systems
Pull systems signal when its okey for you to produce. eliminate wase ina lean system by pullingting through to get flow
short term fixes
RCA seeks the true casue of the problem but most peole engae i short term fixes. the problem still occurs with short term fixes because measures are only put in [lace to try to eliminate the wast. prevention of theproblem does not occur. even though assumptions are utlized for RCA they should be backed up with documentation and dta.
preventive actions
Reactive - a result of finding a problem and fixing the issue. Proactive - potential for issue or defecrxt or provlem where you put in place a prevetive action before the problem occurs
After the Tollgate review meeting
Report back to team outcome, follow up requests from sponsor
Requirements
Requirements are features or characteristics that customers expect in prducts and services. Requirements can change so producxts and proxesses will also need to change in order t meet customer expectations,
Generate Possible solutions
Review all possible solutions to each of the critical root causes. Need Options in order to choose the best solutions
rolled throughput yield
Rolled througput Yiled RTY accounts for both scrap and rework. is the probility that a product or service will go through all the steps of process without a difect. Rolled throuput yield is equal to the first pass yield (FPY) of the first process step multiplied by the first pass yeld of the second , for all process steps
Before tollgate review meeting
Send out critical documents to sponsor. Preapre backup documents including collecte data and details about collection process. Find out sponsors preferred format for presntation.
5. plot and connect the ponts on the upper & lower chart
Step five of constructing a control chart is to plot and connect the dots on your charts. The dots for your upper chart are your values, which monitor the centering
Process Improvement
Six Sigma improves process- Quality improvement , speed improvment , elimintates waste
rule three
Six pointsin a row , increasing or decreasing is called trend. - investigate why trend is occuring it could be special cause variation. If the trend contiues , points may end up outside of your control limit
special cause variation
Special cause variation are unusual, not previously observed, non-quantifiable variation. Best practices for fighting special cause variation include obtaining process information quickly, taking immediate action on them, and develop a quick solution to prevent recurrence.
Performance Indicator identification
Specify what needs to be mesarured, focus on leading measures upstream in process, measure the steps of the process that will ensure we get it right the first time
Motion
Spend More time looking for the object than you need to do your job, then thre is less time actually doing your job
Standard Deviation
Standard deviation is the basis for all of six sigma. standard deviation is the most important of all the measures of variation. It is the measure of dispersion or varation that goes with the mean
1. Gather Data
Step one in constructing control charts is to gather data in the order of production. Make sure that the data is gathered and recorded in order; if it is not in order, then the data will not make sense. If you cut up a film strip, mixed them up, taped them back together again, and then tried to play the movie, it would not make sense. The same principle applies to your data.
2.Calculate subgroup average and range.
Step two of constructing a control chart is to calculate subgroup averages and ranges. In our example, we have a subgroup size of five. This means that we are going to collect five pieces of data now and then five pieces of data at a later time period of our choosing. We are going to keep collecting this data across time.
Surveys
Surveys are used to collect information by sending out a questionnaire to a large sample of customers. Surveys can be either paper or electronic.
The 6 M's
The 6m's are six categories used to facilitate brainstomring effects Man(in a generic sense0 Machine Material Method Measurement Mother Nature
what the control plan does
The control plan explains - what we need to do in order to sustain our solution. WHich variables we need to keep an eye on for monitoring. WHen to respond / what triggers a response to action
Ownership transfer
The control plan makes it easier to transfer ownership of the project and its deliverables
the ends of the wiskers represents
The ends of the whiskers represent the minimum and maximum values. The individual dots beyond the whiskers represent any outliers that could be present in the data. Indiviudal dots beyond the whiskers represents the outliers
Goal Statement
The goal statement must link to the problem statement. The goal statmen is what fixes the problem. target condition or desired future state that will fix everything. Defines how much of the problem this project will fix. Never promis more than you can deliver. When defining how much of the problem you are going tofix, start at 50 percent and work your way up. Promise less, deliver more
Tollgte review Meeting Partipants
The green belt, spnsor, black belt mentor team members when appropriate
Interquartile Range IQR
The interquartile range (IQR) is comprised of the middle quartiles (the seventy-fifth percentile minus the twenty-fifth percentile).The middle part of the plot, or the interquartile range, represents the middle quartiles, or the seventy-fifth minus the twenty-fifth percentile. The line near the middle of the box represents the median or middle value of the data set. The whiskers (lines) on either side of the IQR represent the lowest and highest quartiles of the data.
control charts and stability
The process must be stable in or der to determine process capabiligy. what does stability look like. Point flutuate randomly between the control limits. A proces that is in control is predictable. a process that is not in control demonstrants intability and we are unable to predict how it will behave next
extreme data sensativity
The mean is sensative to extreme data unlike the mediam. the interquartile range is not sensative to extreme data. the standard deviations is sentsitive to extreme data
Standard Deviation
The measure of variablility that goes with the mean . Determines how close to middle most of data is. Shows what percentage of data should be certain distances away from the mean. Impacted by extreme data and not easy to calculate
milestones
The milestone section is a high-level team action plan. It can be created in alignment with the tollgate deliverables or each of the DMAIC phases. The milestones should be kept short due to space constrictions; just include the highlights.
Analyze Phase
The potential causes of problems are identified. Current process are analyzed . relationships between inputs process and outputs are identified. data analysis is carried out. waste and route cause are determined. root causes are verified to understand cause and effect
Define Phase
The project is initiated, the problem is defined, goals & scope are identified as well as the customer, a clar project charter is develeped based on real life problems that affect the customer. business case is defined, benefits are determined by asking basic questions. Project management is determed including project plan the stakeholders, managment support and the tams
Range
The range is the distance from the highest point of data to the lowest point of data.
training and communcation
The response plan creaes an ongoing strategy for taking action. Essentioally it is the teams how to instructions for esponding to out of control conditions and defects. It allows em[loyees to be trained so they know how to resopnd to the rocess once it has been subjected to DMAIC metholdolyt. The response plan should be completed prior to wriring any work instructions and procedures. Thereesponse plas is used to make sure that it alignes with your work instructions and procedures. Iti is impertarive that hte respons plan is visible acceptable, accessible and learly communicateed to everyon
SCOPE
The scope statement should be brief and identify the problem process. Once you've identified the problem process, set boundaries around it. Where does the process start and stop? What is not included in the project?
missing the middle
The tails are present, but the middle is missing. The data is within the spec limits, but the tails are somewhat close to edge. This distribution commonly occurs when somebody is pulling items from the middle data. This situation can occur when a customer wants a high level of consistency and pays a premium to pull the parts at the center of the specification. This histogram represents the leftovers.
Z score table
This can be done
PT Ration- precision to tolerance
This checks precision in a measurement system. It does not, however, demonstrate how accurate an overall measurement system is. It is understood as the ratio of variation from the measurement systems as a percentage of the total tolerance.
to creat e the plot
To create a box-and-whisker plot, you start by ordering your data by putting the values in numerical order. Then you find the median of your data. For example, assume that you have three points — the first middle point, which is the median, and the middle points of the two halves. The middle points of the two halves divide the entire data set into quarters called quartiles.
Tollgate Reivew
Tollgates are key to the success of the project. each one feeds to the next and they build upon each other. Never skip a tollgate. Each of the fie phases of DMAIC has 3 tollgates. Approval to move on to the next tollgate comes from the teams sponsor who is also a member of the management team . Approval happens during the tollgate review meeting
Conclusion of define Phase
Tollgates of define- Created a charter, compelling business case , clear problem statement , defined scope, management spupport. 2. Gatherded voice of customer- who customer is, needs, requirements. 3. mapped process. The process that delivers the goods or the service to your customers. SIPOC chart and one or more detailed flow charts specifying the steps to delivering the goods and the services.
Sigma Table
USe THIS - 1.5 sigma shift - six sigma is 3.4 defects per million, WIthout Sigma shift- around two defects per bilion.
Philip Crosby
Unconditional involvement of senior management which addreses the culture tha must be adopted for the new philosopy to take affect
Unisys texas instruments & ABB
Unisyse & texas instruments were first to inquire about motorola methods & used it as a defect removal system in manufacturing The Mikel Harry & Richard Shroaeder formed a private consultant company and worked wit h ABB
Standard Deviation
Unit of measure for distance from the mean. Equiuvalent to the average deviation of values from the mean for distribution. Use this when data is symmetrical
USE RTY
Use RTY if there are checks and yields from multiple process steps and if multiple defects per uint are rare and not impactful to the cost of qulity.
Measures of Variation/Dispersion
Use the range , inter quartile range, standard deviatin or varience to understand the dispersion of the data -
Hypothesis Testing
Used when we are making inferences about what our data is telling us. Helps integrate the voice of the process with the voice of the business
Tollgate Review Meeting
Villanoval lean six sigma recommens having 15 smaller indiviual metings one for each tollgate.
voie of customet
While mapping the voice of the process, think about the voice of the customer. The voice of the process considers the activities within the process, the order they occur in, level of quality, cycle times, and more. At the same time, consider the customer. What are the customer's specifications, targets, requirements, and tolerances? We need to separate these perspectives at this point, but eventually the voice of the process and the voice of the customer will be married when we do a capability analysis.
Contents of a response plan
Who will be taking actin once a defect is found. what kind of actions will be taken. be very specific becuase there could be some ambiquity in som of the terms. it is a good ideas tocreae operation defintions for any terms thatm might have different interpretations. A schedule that explains the timeline for inestigations and implementation. any troubleshooting procedures and best prectice approches to correct the issue
Normal distributin: too far off to the side
Within normal distributed but misses the target. It is so far of the side that defects have occured. Someting needs to be done in order to satisfy the customer and move the distribution back toward the middle.
Central Tendency
a measure of central tendeny is one number that represnetns the middle of the data that could be the mean medium or mode. The dispersion or variation is the spread of the data or fatness. in six sigma we always consider the variation or distribution
cp and cpk are capability studies
are capability studies- estimate of the future process
Multi-vari chart
a multi vari chart enables you to see the variation of the part. measure the item in multibple ares. mark down the high and low vaues. connect the dots and draw a line. th variation will show. if the lenghs of ines are the same it indicates of within piece variation. if the lengh of the lines that the within variion are very small wile the pice to piec is juming all over the place it shows smething is changing.
hypothesis testing
a scientific process for examining assumptions. Provide the framework agains which you will test those assumptions. Provides the level of precision you will need in your reults to draw certain conclusions. includes the hypothesis statements themselves , the assumptions you begin with and what you are testing for.
tally sheet
a tool to keep track of data
The Critical quality Tree
a tree digram used to drive anlaysis from the voice of the customer back to the activities requierd to help you meet the nees of the custome. this ties in diretly with the hous of quality or quality function deployment. Data should always com first from the vice of the customer even internal data received from our experts
add
add each proess column
Mean -average
add the numbers and divide by the numbers you add
third party
advantage - gain competitor data, gain researh data reach more pople less company time spent gathering data. Disadvantage - expensie,
Direct
advantage - gain first hand info, target your group, face to face contact / disadvantage expense , renat an area for interview, group think, low respnse rate to survey
Indirect
advantage - immediate resolution, control over turnaround time, complaint or conscern is expliciety express, direct communication with cutomer. Disadvantage - unhappy customers dont complain, they go away
affinitize the list
affinitize or group the items. Many casuses may be the same or very similar , aggregate the list so that only unique items appear. Allow the team to rule out the obvious items if the team comes to a unanumous or high consensus eliminite the item
Improve
all potential solutions are evaluated through data driven exercise o that awssumtions are mitigated. Error proofing exercises are put into place to eliminate accidents. risks are assessed. famulre modes and effect analyis is performed. pilot plans are conducted to ensure proper rollout beforme implementation occurs. Implementation occurs. validatoin of tatistical methods takes place.
eleiminate potential root cause
allow team to rule out the obvious potential root cause, consider trying a graphical tool, use a priority matrix on the cuases that are left over. set up the criteria, define an impact or importance rating, score each possible cause, ask the team what will eliminate or elevate the possivble casue. Elimination techniques wil vary for each. select a tool that will eliminate or elvate the possible cause.
What is in the problem statment
answer the questiosn what is the primary issue my team will focus on- idetify the primary issue my team will focus on. this means identiying the primary project Y. Identify the process output that is not living up to the expextation. This means identifying the seondary projext Y. Y's are expressed in terms of xost gaps or something that is going to bring pain. this is what will give you buy in and support.
Performance indicators identified
are those things measureale, what affect do they have on a customer, what are the customer requirements are there leading indicators
tollgate review. potential root causes
at eh potential causes tollgate review meeting the sponsor will waht thefollowing information. waht are some of the potential reasons for gaps in preforanc, how did the team identify those. the list of potential root causes might be an overwhelimnglily large list. if tha is the case group them into meanigful categories. the goal of the potential root cause tollgate is not to eliminate any root causes, soe of which must be reasonalbe . if they are not reasonable or the list is too small you can alwyas go and find more and add them as necessary.
Tollgate Review : Critical root causes
at the critical root causes tollgate review meeting the sponsor will want the following informatin. A short list of critical root casues. the ritical few from the trivial many. How the team identified these critical root ases- use data driven tools to make this final selecxtion of critical root casues. An estmiation of how much gap in perforamance can be closed by eliminating each of these root casues. A hard copy of the official list of critical root causes.
Tollgagate review- montot and control plan
at the monitor and control plan tollgate revidew meeting: performance charts and graphs. a document measurement plan- this lets the sponsor kow that the process is being watched so that is does not get out of control again. Cost savings or improvements i customer responses. The new process map. Gives this to the sponsor a few days before the tollgate review meeting. A control Document - staff should be using this. It shows the sponsr that the standard oprating procedures are being executed.
Attribute and Variable data
attribute data - whenver you are investigating if diving in half makes sense then you are using continuous data things like time, pressure, speed, feed rates. If diving in half does not make sense then it is attributre data
why people do not use the project
believe ist more work, do not see benefits. as a team leader make sure everyone understands the benefits and function of the project charter
Dr Armoand V Feigenbaurm
believed for quality improvement to be successful everyone involved had to be completly commited to success dubbed his stratgy TQC toatl Qualilty Control. Products should be made and sold , higher quality lower osts.
sum up and sort the characterisics
by had or with a computer add up the occurences of each charteristic and sort from largest to smallest
The value of process capbaility using cp and cpk includes
confidence that process is performing well. ability to perdict an outcome. better ability to communicate with customers. developing realistic expectations . establishing pricing for product or service. better ability to plan. potential the most valuable part of process capbility is that i is part of the control phse of DMAIC cp and Cpk are great metrics to monitor the process.
Monitor and cotrol plan
continue to measure performance. Ensure the new proces and procedures are being executed correctlu. avoid slipping back oito doing things the old way
kaizen
continuous imrovement - to constantly review and make improvements so that you/your products get beter, faster, easier, safer, cheaper
Fonal Project Report
contorl lans are essential portion of the final projectreport. The project report comlies what you need to give the process owners. it is their document to keep the projectgoing. it is used to refernece to keep the team on track. source of lessons learned and ginding information
Business case
creaet a complelling argument for why the project needs to be completed. - Short sweet and to the pont.
defect location
create a picture of the part, make aform give it to someone that is inspecting and have them put a dot on the part they see as the deefect. if there is a cluster they can look at procss to see the cause of defects of that area.
Response plan
create plans to address potential problems
critical quality characteristics
critical to quality characteristics CTQ are lean tools that directly address quality , cost and delivery
internal
cusatomers within your organization. thay are not buying the outputs of your process but they do receive the output of your work. They ahve requirements and specifications that need to be met , ortherwise it will impare their abilit y to complete their portion of the process tha will be bought by external customers
matching needs/requirements
customer nees are quivalent to process output. customer needs and requirements do not always match up. requirements are supposed to identify a specific haracteristic of an output. clarify that the need and requirment match up
customer requiremnets
customer requierments must be clearly defined - upper specifriction limit & lower specifiction
Where does it fit relative to the customers specifications
customer specifications are anything important to the customer, data is collected on the spexifications examples include size, speed or cost of a product or service
Customers and the IPO Model
customers can be the next person working on the product or service. The model called ipo model that isolates internal custmers by identifying inputs, process and outputs.
feedback challenges
customers sometimes give solutions not feedback. needs go unspoken called laten needs, customers man yot even know their needs. presumptions of your product / service may not match what the customer needs/requires. you may run into competing critical to quality charachteristics.
Indirect external
customers who do not receive or pay for the output of your services but have vested interest in what you are doing. - example government agencies.
Operational definitions and target types
define things aghad of time in order to avoid confusion and prefent introduction of varition. its not the most exact definition but is the definition everyone can agree to
Dissatisfiers
dissatisfiers ared the basic requirments the custer must have. you will not get their busines without these requirements.
ratio
distance, tempature or wieght - has a fixed starting point
Normal distribution: Wide in the middle
distribution is wide in the iddle. this signifies a lot of variation in the process. if the variation becomes any larger, it may be in trouble of failing outside the customer specifications. Whenever the data breaches those limits, a defect has occured
quantiative
divisiions of values name, order interval orgin
Practical difference
does it make sesnse, does this dirreence result in an improvement that will have economic or practical value.
who should seek yellow or white belt training
entry level employees or execs who need foundation to properly support an improvement team.
poka-yoke
error proofing is the idea of preventing devects from ever happening
nominal scale
example - color,
ordinal scale
example days of the week since they follow a sequence
types of customers- external/ direct externa.
external . direct external customers- customers wo receive the prodess sucha as a product or service, source of our resume
black belt
facilitate large improvement projects. Coach/mentor green belts. enure tollgates are completed correclly before moving forward manage the improveire tplan for entire e product line . dedicated full time to six sigma
four options for hypothesis testing
fail to rjext the null hypothoses - becuase the null hypotheesis was actually true.
White belt
few hours of. training, basic understanding
rule seven
fiftween ponts in zone A abouve or below the mean could indicate special cause variation. Expect 68 percent of data in zone A . Operators may be throwing out defets instead ofrecording them. If the process is running that well, recacualte the new control limits
Check mark
find the square root
Ferdrick Windlow Taylor
first scholar of modern business - used adherence to detailed documentatin to identify patterns in the behavior of workers. used common measurement tools three core pronipals of project managerhip and leadership: Benchmarking , Process Redesign and work out. Strong proponent of standardized processes and measured performance.
workout
for less complex problems use six sigma version called workout - define the problem. decide what you need to measure, get your baseline, affinitize root cause, priritize root cause , find root cause, brainstorm solutions, affinitize solutions, prioritize solution, implement solution, measure again
decision drivers
four decision drivers are tied to teh voice of the customer. Features, quality , deivery Cost
Critical to quality
four decision drivers help you understand what your customers want and why they buy from you
levles of measurement
four levels of scale:nominoal , ordinal , interval, ratio
rule six
four out of five points in zone b or c could indicate special cause variation
rule four
fourteen ponts in a row alternating up and down may indicate special cause variatrion - investicgate why the alternating points are occuring
master black belt
full time position manage six sigma for entire business, create a 5 year strategic plan - coach & mentor black belts
how to check for accuracy
gare R & R will check for precision no acurracy. whe hae a prcise measurement system you have to have amaster gage or some kind of an expert or accurante gage that is know to be exact. they you can measure your parts or services with that master gae. then you check with your gages to make sure that there is a small amount or difference between the expert and the mster gage and we're seeing from our gages that we are using in our measurement system.
how to narrow root causes
gaterh a list of potential root cause , they need to be narrowed down to the critical few or one
value of process capability
get some idea as to whether the process is performing well. somehat predict the outcome . communicate better with customers. develp more realistic expectatoins . plan better than if there wasno data.
looking up the critical value
go to the t=table and look up the critical value. most t tables show both one and two tailed tes values on the same chart so be sure that when you checkt the column that you are using the two tailed column
colleting good data
guessing - worst way. finance can regulate how much data to collect as it may be expensive. governmental agencies , regulatory rgroups may tell you minium sample. industry standards will tell you if you are abiding by requriements . forumlas are avialbe to help determine minimum sample . you need a sampling plan based on how data is generated to ensutput variavble priorized re you only collect as much data is needed. destructive/nondestructive. return on inviestiment. input/ output priorize the collected data , determine how often to collect , how much , process to collect
force field analysis
helps identify- the forces pushing you twards a certain action the forces pusihg ou awy for a centain action. Information could be useful in trying to develop a charter.
voice of the process
helps you understand what is giong on within our processes. helps undertand what is wrong and why the other two voices are complaining. balance 3 voices voicess of process, voicess of businnes, voice of customer
Criteria for selecting first wave projects
hgh impact tied to key business objectives. low performing process. low impact on resources . potential cost savings.
SIPOC
high level process map tools -C in sipoc stands for customer- will help you find a list of key internal and eternal customers. Map relates the customers to specific processes your project has tartgetd
skewed distribution
histogram is not normally distributed the data is not bell shaped but piles up on one side. this is called skewed distribution. since this curve is not normally distributed we know that something other than random variation is at work. the mean is not in the center of the two specifications.
Quality
how does custerom define value-customers define value because they are the ones purchasing the product /service
u
how many out of a thousond, the bar is continually rased. sigma help;s break it down into sizable chunks
cost
how much doesa the product cost, does the custemer percieve the value for that cost
Pp
how wide the bell curve is compared to the window of opportunity from the customer
Critical to quality Charateristics CTQC
howe are we going to address or measure our sucdess, what is critica to quality dirve all the way back to the voice of the customer. what is critical for meeting that customers needs. What are the whats or needs.
3. Critical root causes
identify the significan few casues andverify to be source of problem. not just symptoms but actual causes. use statistical or data driven tools address these casues will close the gap
1. Possible root causes
identify using ideas gernation tools, ideal to generate a large number, keep in ase you want to revisit
type 11 error
if you make the region to large it can give you a false rejectin. a type 11 error can occur due to chance or bad luck. most of the type we get a type 11 error failing to rejext the null even though is falce because we made the significance level to. small.
benefits of box and whisker plot
include alowing for the comparison ofdata from different categories for easier more effetive decision making
independence
independence in actual events and how they are related or in this case if they are independent not related. two events are independent the outcom eof one event does not impact the outcoe of the other. dependent. the outcome of one does impact the outcome of the other
Ppk
indicates the centering of the data or the curve
baseline performance and data collection -
introduction-ollecting good data is an important part of the decision-making process. We need to ensure that we are collecting only the data that we actually need and that we are collecting it in a deliberate manner. This data helps us to understand the baseline performance of our process as well as the magnitude and frequency of our defects or issue.
control plan
is a written summary - describes the systems used to montor and control our. - processess, products, service quality. Is a formal record of approaches used to meet project goals. Defines everything that needs to happen. Keeps everyone informed as to what is going on in the project. Starts with define phase nad flows throught DMAIC preocess
delivery
is custopmer getting what they need ain a timely manner, what is the customer time frame, how do you know that you are delivering within the customer time frame
ask these questions
is data normally distribured. Is the data consistent , predictable, and repeatable- CPR- predictability is most important
LSL
lower specification limit
scatter diagram
is there a relationshp between one of the critical x's and the y or outcome variable
choose the charactherisic to analyze
it should lead to a conclusion that is actionalble
Kaisen Events
kaizen events make very substatntial changes and make improvements in operations in a small number of days instead of months
The purpose of Lean
lean is focuing on eliminating waste and creating flow. in order to increas flow you should decrease cost because time is money. the longer an items tays with you the more it costs
Lean and DMAIC
lean tools can be found throughout DMAIC - Measure - Lean metrics are about measuing. Analyze - cellular design, tpm, kaizen. Improve - Kaizen Control- standard work
Points of data/number of bins recommended
less /equal to 50 = 5-7. 51-99 = 6-10. 100-25 7-12,, greater than 250 12-20
Project collection Matrix Framework
list criteria down left side. List five process across top with a column for ranking criterias importance 1. order taking 2. preparation 3.Initial step 4. Finishing Step 5. Delivery
Symbols
look at page 22 for symbos
how to capture voice of customer
make a lit of what you need to know, decide how much data you need to. capture, decide how you are going to use that data. select tools for data collection, collect data
black belt skill
manage multiple projects. complete advanced statistical analayses Expert facilitator. communicate ideas ensuring individuals understand ideas
applying force field analysis to the project charter
many forces ushing toward doing the project. many forces pushing agains the project. issues tha the champion or sponsor might appreciate being addressed during the charter tollgate.
carl friedsrich Gauss
mathemetician from 19 century europe- introduced coceptual normal curve metric as an error analysis tool for measurement probability and hypothesis testing. initial principal - for six sigma
bell shape
means there is no unexpected varition from any of the 6M's. we cannot learn much from looking at just one measurement every so often the more the process is adjusted to the center the more it increases in varioation
parameter
measurement sucha as mean or standard devition wihich represents an entire poulation. if the poulations is small enough to be analyzed then it is a parameter. parameters are denoted by a greek letter. sigma represents the standard deviation of a poulation
statistics
measurement that erpresents a randmon sample or a subset of that popuation. Statistics are denoted by a regular letter. example S represents the standard deviatin ofr a statistic or sample
measurement discrimination
measuring fine enough to see the variation.
six sigma definition
meaure of quality that strives for near perfection. Six sigma is a disciplined data driven approach and methodologh for eliminating defects. company must reduce their defects to 3.4 defects for every million opportunities
champions / sponsors
mid level or sr managagers. authority to deploy resource & remove barriers. chartering & selecting process owners review results , make course corrections for moving forward
Mode
mode is the number that most appears in the data set. 2,3,4,555,7,8 5 is the mode. 2,3,4,555,6,7,888 9. bi model distribution- when there are 2 numbers that are most often in the data set
Multiply
multiply the importance number with each number in the row
t test formula
n is the sample size. x-bar is the mean of the sample data. S is the sample standard deviation. U is th population mean. for a t-test the mean of the population the data is drawn from is known
tollgate review: narrow root causes
narrow root casue tollgate review meeting the sponsor will want an official listof root cuases. how the team identified the narrowed root caseues. what tools the team usesd. datashoulddrive the decision to eliminate potentail root causes. any new information or updatses that might change the scope of the project or charter
Examples of needs vs requirements
need : people need to eat, requirement perso wants a fresh saled with frout.Understanding customer needs and requirements are critical to ensuring that products and processes meet customer expectations. If they don't meet customer expectations, then we don't have a business.
Conssequences of not using the project charter
no one will be working off the same expectations
measurement scales
nominal ordinal interval and ration, dispersion. central tendicy mean medium mode
non-rational subgroups
non rational subgroups have many changes in regards to the 6 M's example manufacturing a jet airline.
k
number of bins
n
number of items in the sample
Common mistakes of green belt
preconceived assumption of solution. Poor definitiaon of problem statment. Should follow SMART- Specific, measurable, attainable, relevent and time bound. Lack sponsor/champion buy in. Exclude key functions from team- should be cross functional, include front line employees
Proecess capability CP and Cpk
process capability can be used to dtermine past perfoemanxe, presnet performance, and furture performance.
process capability
process capbility is the ability of process to meet customer requirements. The upper and lower specifications limits must be clearly defined. process cability involves 1. voice of the process , or understanding what the process can do. 2. Voice of the customer or knowing what the customer wants.
3. Process Mapping
put the entire process on one large map incuding the inputs and outpust who the suppliesr are who the customers are
standard deviation
quantifies the varibility in the curve. Translates to how broad or how narrow the curve is
define the problem
remember the SMART Goals- Specific , measurable, attainable, relevent and time-bound
Eliminating Defects and slowdowns
remove defects and slowness
Market Research
research uses a company's internal data as well as industry data to understand customer needs, services, and competitor information. Third-parties will often provide market research solutions because their specialty is complex data analysis.
satisfiers
satisfiers ares variables that are not sttic. the more of those satisfires you can meet the better. they are not static. the more those satifiers you can meet the better. the customer will noteice when you meet these and you have to meet a certain number of them in order to keep the customer happy.
scatter diagram
scatter diagram show positive/ negative correlation. correlation does not necessarily mean causation. it shows you the variation that is hard to see if you are just looking at thenumbers but it does not show causation
nunmber of bins 2K + 1
see oage 32-33 for exampes
using different scales
see page for pic
Vaidating and prioritizing the critical to quality characteristics
set specific requirements and validate them wioth your customers, review existing peformance to ensure there are no gaps in what customers say they want and what you are acturlly doing confirm with your customers
data collection plan
should be easy to read and understand. Define all coluns. ensure it presents the right kind of data in the right amount. identify what is already in the document control for data collections. using
two sample t test
six sigma practitioners oftew want to know if tow means are statistically different from each other. so different that it was likely that not just chance casued thedifference.
Kaizen principal
small incremental continuuous improvements every day.
implement solution
soloutions put in place, plan drives execution
tollgate review - data collection plan
sponser will want : a plan teling the sponser exactly what data you are going to collect, how you are going to collect it and how muhc of it you are going to need. an estimate of the cost and time that it will take to go get thedata. Give the plan worksheet to the sponsor ahead of time. be prepared to give a consise clear explanation of each measure at the review meeting. come prepared for the items the sponser may find undreasonable . have a backup recommendation, measure ready to go e especially if some of the data is going to be expensive, time consuming to gather or distruptive to business
formula to find sx1x2
square root of sx2 + sx2. /2. - see page 29. This formular works only when sample sizes are equal n1=n2. assumes equal variance.
What des standard deviation tell us
standard deviation shows the varibility or dispersion of the data. If there isa normal bell-shaped curve whe know a lot more than just the average point away from the mean. when there is a normal distribution of the randam data there are certain expectations about how much data falls under varius points under that curve. 68 percent of the data between plus or minus one s t andard deviation. 95 percent pof thedata plus or minus two standard deviations. 99.7 percent of the data plus or minsu three standard deviations.
flexibility in six sigma
strenthths - clarity, purposeful, sytematic, logical, ensures that you have what you need to be successful
taguchi signal to noise ratio
taguchi created a concept of signal to noise ratio which is about the vial few verses the trivial many. The vital few are the signals, they are things that cause an output. X's or independent variabls are the vital few. The trivial many are additional things that are happening that do not cuase an impact
present performance
teams can gaterh real time data to draw the process and get an ideas of how the process is performing
Team
tean lists sponser/champion, process owner green belt/black belt team mmebers. also lists what each teams members ole and how much of their time is dedicated to the team
The Kano Model
the Kano model is a measure of customer needs. its is used to identify where we are meeting or not meeting the customer needs data is categories into three sections dissatisfiesr, satifires and deligters
Z-scores
the Z distribution is another name for the normal curve plus or minus one astadard deviation is 68 percent , plus or minus two standard deviations is 95 per cent . plus or minus three standard deviations is 99.73 percent or almost all of the datal.
process capability
the ability of the process to meet customer requirements process capability is made up of voice of the process or understand what the process can do voice of the customer or knowing what the customer wants
68 percent of data- 1 standard deviation
the area plus or minus one standard deviation
95 percent of data - 2 Standard deviation
the area plus or minus two standard deviation
key characteristic of a normal curve
the area under the curve is relative to how many stand deviations we are away from the mean
99.8 percent of data - 3 standard deviatin
the areal plus or minsus three standard deviation
Xbar
the average of datate
create a bar chart of occurances
the bar chart must pass a commonsense test, is the data relevent to the problem and is it actionable
rejection region
the confidence level and signifcance level must alwys add up to 100 percent. they are complementary. the significance level is the total amount of data in the tail or tails depending on the type of test. the tails are also know as the rejetion regions
Living Document
the control lan is a living set of documnets. Update when necessary adding what you need for continuous improvment - new measurement systems, new control methods, - standard operating procedures - adjustemetns to process maps - adjustments to training appraoches
Normal distribution too wide in the middle
the data goes down the middle but there is far too much variation. It is so wide that the histogram crosses the spec limits on both sides. we need to determine why here is so much varition in the process and decreast that variation in order to narrow the distribution
The 15 DMAIC Tollgates
the five phases of DMAIC and 15 tollgates are clear logical and throrough. DMAIC and 15 tollgates are systematic repeatable reliable process to drive success during improvement projects tat can be applied to any project.
interquartile range
the measure of variation that goes with the mediam. notimpacted bh extreme data indicates where the middle 50 percent of the data is . A little more difficult to calculte and only gieves imormation about the middle bend. Most useful is used in confuction with a box plot.
pareto diagram
the pareto diagrem is a great option to rank issues. it will tell us which things occur the most often. collecting data is usually simple for pareto diagram
the vetting process
the peoptential causes will be verified before being designated as the critical root cause
The beneifits of lean tools
the purpose of lean systems is to create a flow and eliminate waste. WHen discussing wast lean focues on owntime and downtime. six sigma is about efficiency speed and elimination of varition. when discussion waste six sigma focusos on the elemination fo defects which is the d indowtime
monitoring phase
the purpose of the monitoring process is to perform checks in order to ensure that a new pocess is performing as designed
qualitative
the quality of something can be describe as the divisions between soures or categories 6 M's .
tollgate review baseline performance measurement
the sponser will want the following- charts and raphs that depict the measure the target and how big the gap is. This might involve calculating the actual sigma measurement and or conducting a capability study by measuring process capability PP and Ppk. Any necessary inforation to make some final updates to the charter. An up to date version of the charter.
Definition of standard deviation
the square root of the average of the sum of the squared deviations. . it can also be thought of as the average distance any pont of data is from the mean.
probability
the student t distribution helps find the probability of a vlue being on one side or the otherof the test statistic. as a result the first stepis to calculate the test statisic. then use the student t-table to make a determination about the probability that statisic occurred due to just chance alone. it is importnat to know tha the table being used provides the probability that the vaue lies to the left of the right.
standerd curve vs t curve
the t distribution looks like a standard normal or Z distribution and fits a bell shaped curve. However the tails in the t distribution have a little more data and are ghiger than a standar normal curve which comes down a little lower first and then coes across.
larger is better
the target could be inifitint or 100 percent
the t distribution
the te distributin is a probability distribution used to estimate the population mean or test a claim about the population mean based on the fact that the poulation distribution is approximately normal and the population variance or standard deviation(eitherone) is unknown or the population distribtion is approximately normal and the sample size is less then 30. If the sample size is less then 30 we wantto do a test of means we use the student t-distribution.
Defects per unit- DPU
the total number of defects found within a specific period of time divided by the number of units produced in that time
customers and six sigma
the voice of the businees and the voicew of the customer ar not opposting voies. keep customer satified with the producrt so you can generate revenue
assumptions
there aer three main assumtions of the two sample t test. 1. the populations samples come from are normally distributed. 2. the varience in the populations is equal or can be assumed to be equal 3. the samples are independent. Independence is critical data needs to be geterated in one group or other not both. the t test is really great when smaller size samples less then 30.
parts of box and wisker plot
there are three points that create the box onthe box and wisher plot the first iddle point is called the mediam thi ssplits the dta in half. The second point is the medium of the first half f the data. the thrid point is the medium of the second halp of the data . the tree points create quarters which are called quartiles
4. plot the control limits -
there are two parts o of a control chart the upper paret monitors the centering the lower part monitoras dispersion
mutually exclusive
there is no way for outcome in event a and the outcome in event b to occure at the same time. two independen events could be mutually exlusive.
rejecting the null hypothesis
there is statistically significant differencebetween the mean . if it is beyond the crxitical valuen the null hypothesis must be requected. That means the difference between them was greater than or far less then.
delighters are the latent requirement
these are the requirment that the customer may not have even known that they wanted or desired. they do not expext them but if you meet them the customer definetly notices and they are delighted.
Drawbacks of DPU
they do not consider the complexity of the product or service being analyzed. the more complex a product or service the worse it DPU or RTY will be when compared to similar products that are less complex. Only consier DPU or RTY may cause an organization to choose the wrong priority when trying to decide which service or product to deploy resources for improvement
attribure
things that are counted , discrete data answer to these questions are yes/no, go or no go , counted data
variable
things that are measured - distance, height , time
leading/lagging indicator
things you can change in order to get better reults in your lagging indictors. lagging indicators cannot be changes so focus more on leading
bell shape is too wide
this means there is too much dispersion in the process. The data isn't quite centered within the customer specfications - the mean falls witin the specificatoins , but part of the tail is ouside the upper limit
complementary
thowe wher it has to be one or antoehr example die it is either odd or even, wont be both but has to be one.
twofers
three ares you need data for , customers reuire quality. - inpouts from suppliers, - process designs - outputs. in each of these arease there are two types of measures that why they are called twofers. there are two measures for each data area
internval scale
time- does not have adefining starting point
future performance
to deterine process capbility for future performance use Cp and Cpk
run chart
to use a run charg you need data. the run chart will tell you about trends. if there are 7 points in a row either in a upward direction or 7 points in a row in a downward direction that tells you that something unexpectedly happened . 7 points above or below the lne it tells you that something shifted in the process up or down. why 7 ponts in a row - that brings you o nearly a 1 percent problitiy that the items heppened just by chance is alone You need to go through the 6 M's to figure out what made you unexpectedly better on that particular outing same as unexpectedly worlse. Run chartes give you a running reacord and illustrate shifts and trends.
tollgate review - performance indicator identification
tollgate review - - what you are going to measure, clear explantion of performance meausres you chose and wjhy the are good choices
total productive maintenance
total productive maintance (TPM) includes having the right people involved at the right levels. for example oerators who use the equiptment everyday should be invloved in the process
Conclusion
totals help you determine which process you might base your first project around - pick one with highest total
Yellow belt
two to 3 days of training. help facilitate basic tools. cannot manage a project on th
qualities of a good gren belt
unbiased, no stake in problem basic understanding of proecess, excellent soft skills know how to communicate. good Project management skills
USL
upper specification limit
milestones
use DMAIC as your 5 major milestones. Each tollgate takes about a week. since each phase has three sub-tollgates you would have 15 tollgates all tether. if anything modify the charter is necessar
how to addrtess the hows
use corrective actions and preventive action. create a whole system by which we maintin good quality in our production. continuous improvement activities and emphasis on maintaining the flow of product through our vacility facilitates delivery in a timely manner
2.Narrow root causes
use team decision tools or graphical tools (pareto chart- Reduce possible casues to big issues.
project selction matrix
use the criteria to make a proect selection. list of potential projects
value stream mapping
value stream mapping helps us see the flow of material and information throught the process. complete a current state map of wereh you are today. complae a future state map to see where you want to be use the future state map to note the improvements you need to make
variation
variation always exists if you measure fine anough
Variation
variation is a fundemental principal of six sigma six sigma is all about how variation causes defects and focues on reducing or eliminating the sources of vartion to increas quality measures of varition or dispersion allow us to see where variation occurs
Variation
variation is the number one thing to eliminate when brainstorming and fixing a problem. special cause variation must be reduced with RCA . The problem must be undertood thouroughly in orcer to ensure that a permanent fix is put in place instead of a quick fix. RCA is the process of finding and eliminating the cause which would prevent the problem from returning.
Critical Root Cause
verify which root causes will make a differnce, the critical few we will focuson on in the next phase
visual management
visual mangement indludes the 5s and 6s systems. it allows you to organize your workpace get things cleaned up and have a method of sustainability
calulating sigma
we need to know two things are we calculating sigma based on defexts per million units - DPMU or defects per million opportunities DPMO . with DPMU the produt or service is either completely right or defetive. DPMO considers multiple opportunities for a defect
lead time
what about lead time-for continuous improvement we are always after a lead time reduction. are we meausing the lead time and are we. seeing it giong in the right direction. Is there improvement in on-time delivery what are we documenting - if we are documenting 100 percent on time delivery is that the standard being mainting. What about flow. how do we put in place pull systems with a single point of schedule that is at the end of the time contrlled by the customer demand. how do we measure our invetory. Inventory does not equal shortned lead time. it may shorten the lead time for one delivery but the prudt must work all the way through the system and wait in line behind all other invoentory in pipline. so more inventory we have, we decreaseflow and increase lead time
Performance indicator IDentifiation
what do we need to collect data on
null hypothesis
what you would expect to happen due to chance alone if nothing really influenced the process out of the ordinary. Includes all of the things that are not included in hte alternative hypotheis. when you are testing if you rejext the null hypothesis you are foreced to acept the alternative hypothesis.
alpha percentage
what's in the tail or rejection region. selecting the level of precision when selecting the level for alpha becomes important because alpha determines wehere the critical values goes and where the critical value goes determines where we rejet the null hypotheesis orfail to reject our null hypothesis or fail to rejext based on the rlationship to the test start
keep a record
when eliminating the items do not permanently delet them. keep a complete list somewhere.
correlation
when one value gets larger the other gets larger, when one value gets smaller the other gets smaller, when one value gets larger, the other gets smaller, when one value gets smaller , the other gets larger
using images
when possibou can record actual cyle times.convert check sheet data into picuture- pareto chart is good tool . it priorizes for you. Use cycle time information as continuous data. this check sheet will divide up time blocks where you can record cyle times this shows frequency or distrubution pattersn of dta. use data to support your decioin making
Problem statment
where you state the problem and what the urrent process is. Keep it short 2-3 sentencess at most
goals
where you state what you want to accomlish. remember the thre key items go from here to there by when. if you have financial expectaions add it here. keep it short
pareto chart
which of hese bad things happens most often
the charter
why is the project important , the charter helps define why the project is important. not necessarly the only tool you can or should use.
conclusion
with a t distrubution we can explain how probailities occur Deterine what the probility is of a particular value occuring based on the test statistic
Process mapping 1
workrk in an organization is done via some process. In fact, everything we do, every day, is done via a process. All outputs are the result of activities that have transformed inputs into those outputs. That transformation happens in a series of process steps.
Range
works even if we have only two points of dta. immediately shows the highest point and the lowest point in the data set. fast and easy to calculate deos not indicatre where the data piles up arouhd the middle. if shape is importnat range might not be the right measure to use
values check sheets
wrote the values on the bottowm. make form and give it to the people in the process. people can make a frequencey distribution. show the process variation the spec spread
rational subgroups
x-bar R charts require a rational subgroup. Rational subgroups have continuous productin that is fast without much change. not much chage in terms of the 6 M's. Example: baking cookies the whole batch is exposed to the smae processed components
Functions and variables formula
y = f(x) relaes to independent and dependent variables and by extension leand and continuous improvement
direct
you are directly in conteact with the customer. - conduct interviews, focus groups, survey, observation, become the customer, scorecards
using the t test during the define phase
you can use the t test during the define phase to determine if two items are statistically different for radium variation or not
sample size
you cannot always survey your entire population that is why us use a random sampling
actionalble list of measures
you need an actionalbe measure when your investigate customer needs/requirementts.
control charts are divided into three zones
zone A - plus or minus one standard deviation from the Mean. Zone B plus or minus two standard deviations from the mean Zone C plus or minus three standard deviations from the mean