Green Belt

Front 1

What drives companies?

Back 1

Every company is driven by quality, cost, and lead-time reduction goals.

Front 2

Who (1) focused in meeting customer demand for a product variety and (2) was the first to use strategy and marketing in industry?

Back 2

Alfred Sloan

Front 3

Why is Six Sigma the goal?

Back 3

For complex products and systems, six sigma is necessary to produce or perform defect-free more than 90% of the time.

Front 4

PLT = WIP X ER

Back 4

Little's Law: The more WIP (Work in Progress), the slower the PLT (Process Lead Time).

Fixed Capacity (Exit Rate) X Increased People (WIP) = slower PLT

Front 5

Goal Statements

Back 5

S (Specific)
M (Measureable)
A (Achievable)

R (Realistic)
T (Time-bound)

Front 6

Why measure?

Back 6

(1) To gain knowledge about the problem, process, customer, or organization.

(2) To establish the current performance level (baseline).

(3) To determine priorities for action.

Front 7

What is a measure?

Back 7

A measure is a qualified evaluation of characteristics and/or level of performance based on observable data.

Front 8

Non-Value-Added

Back 8

(1) Any activity that does not change the fit, form, or function of a product.

(2) An activity the customer is not willing to pay for.

*We focus on non-value activities for reduction or elimination in Six Sigma.

Front 9

RACI Chart

Back 9

--It is a communication tool.

(R) -- Who is Responsible (the doer)?

(A) -- Who is Accountable (the Champion)?

(C) -- Who is Consulted (before action is taken)?

(I) -- Who is Informed (after action is taken)?

Front 10

Stratification

Back 10

Important in data collection to ensure sample is representative of the true population.

Front 11

LEAN

Back 11

LEAN is a systematic approach (like DMAIC) of identifying and eliminating all non-value-added activities (wastes -- think TIM WOODS) for sustained improvements through a continuous-improvement philosophy.

Front 12

COPQ

Back 12

Cost of Poor Quality

Quality could be how quickly we are getting a student through a curriculum.

There are costs involved with poor quality. (Think about an iceberg--what you do not see that is beneath the water.)

Front 13

Control Chart

Back 13

-A statistical tool used to see if a procedure is stable.

-1st used in Measure Phase.

-They provide a graphical picture of the process over time.

-They help determine special cause & common cause components to variation.

Front 14

Champion

Back 14

The Champion "owns" the project.

He/She approves the membership of the Six Sigma Team.

He/She kicks off the team launch and should be the person who invites the members to the team launch.

Front 15

DMAIC

Back 15

Define

Measure

Analyze

Improve

Control

Front 16

DPMO

Back 16

Defects Per Million Opportunities

Front 17

Core Team Members

Back 17

Think about who has the relevant skills (as the relate to gathering and organizing data) and contacts.

Capabilities more relevant than just subject-matter expertise.

Front 18

DPU

Back 18

Defects Per Unit

Front 19

All work is a

Back 19

Process

Front 20

All processes have _________ and __________.

Back 20

variation and waste

Front 21

Variation causes ___________.

Back 21

defects

Front 22

FTY

Back 22

First-Time Yield

Front 23

Gate Review

Back 23

A gate review closes out one DMAIC phase and opens the door (or gate) to the next.

It could also be a stopping point for the DMAIC process.

Front 24

With what was Henry Ford most concerned?

Back 24

Time.

(Ex: Used black paint on cars because it dried faster.)

Front 25

What is the primary improvement vehicle in LEAN?

Back 25

Kaizen

Front 26

Kaizen

Back 26

The focused application of LEAN tools to reduce muda (waste) to improve cost, quality, delivery, speed, flexibility, and responsiveness to internal & external customer needs.

Implements "do now" solutions.

Front 27

Kaizen Events

Back 27

They have a strong bias toward action.

They focus on the immediate implementation of solutions.

Decisions are often made on basic data and tribal knowledge.

Front 28

KPOV

Back 28

Key Performance Output Variable

(Another term for a metric)

It is the "why."

Front 29

Mean

Back 29

It is our optimal or desired level of performance (when looking at the mean of a process).

It is also the average.

Front 30

Most dangerous waste?

Back 30

The one that cannot be identified.

Front 31

Muda

Back 31

Waste

Front 32

Mura

Back 32

Unevenness (being "in the weeds"); huge spikes.

Example: A very large order from a customer.

Front 33

PPM

Back 33

Parts Per Million

Front 34

SIPOC

Back 34

Suppliers

Input

Process Step

Output

Customer

A SIPOC is used in the Define Phase to help us understand the baseline condition before we move to the Measure Phase.

Front 35

TIM WOODS

Back 35

TIM WOODS -- This helps us remember the eight wastes LEAN focuses on eliminating.

T (transportation)

I (inventory)

M (movement)

W (Waiting)

O (Overproduction)

O (Overprocessing)

D (Defects/Rework)

S (Subutilized Innovation/Crativity)

Front 36

Value-Added

Back 36

An activity than changes the fit, form, or function of a product.

The customer must be willing to pay for these activities.

Front 37

Standard Deviation

Back 37

Standard deviation (sigma) is the average distance between the data points and the mean on a normal curve.

Front 38

RTY

Back 38

Rolled Throughput Yield

The probability that a product will pass through the entire process without any rework and without any defects.

Front 39

TPS

Back 39

Toyota Production System

Formation of just-in-time (JIT) and LEAN production movements in the US and the world.

Front 40

Time Trap

Back 40

The process step that takes the most amount of time.

The time trap determines the exit rate.

Front 41

Time

Back 41

The foundation of LEAN. We cannot get time back (once it is gone; it is gone).

Front 42

Team Launch

Back 42

--Part of the Define Phase

--Charter is very important

--This is where the team gets together for the first time.

Front 43

Range

Back 43

The difference between the max and the mean.

range = max-min

Front 44

Mode

Back 44

The mode is the most frequently observed value in a data set.

Front 45

Attribute Data

Back 45

Has two main subsets: Binary Data and Discrete Data.

No gauge is used to measure this type of data.

Examples: Pass/Fail, Agree/Disagree, Win/Loss

More subjective than continuous data.

All of the values are positive integers.

Front 46

Process Z

Back 46

Sigma Level

Front 47

Continuous Data

Back 47

Any data or number than you can use a gauge to measure. A set of numbers that can potentially take on any value from minus infinity to positive infinity including zero.

Also called variable data.

Money = continuous data.

Continuous data can be divided into subsets.

Front 48

Y = f (Xi)

Back 48

Y = to function of Xs

X refers to the inputs that are associated with or create the output.

Front 49

stakeholders

Back 49

Parties who affect or can be affected by an organization's actions.

Front 50

Six Sigma

Back 50

Six Sigma is where the mean of the process equals six standard deviations away from the customer expectations of the process.

To determine the sigma value, we want to obtain the distance from the mean to the closest specification limit.

Front 51

Six Sigma Quality Level

Back 51

Sigma describes variability.

It is used to indicate how likely errors (defects) are to occur.

Six Sigma quality level equates to 3.4 PPM defects (99.9997).

Front 52

Bell Curve

Back 52

The Bell Curve is the VOP. The process determines the size ofthe Bell Curve. The customer determinesthe vertical lines (VOC). A defect is outside of the vertical lines (the upperand lower spec limits).

Spec limits are based on customer wants/needs.

Front 53

Define Phase
(Key Elements)

Back 53

ProjectCharter

Validationof Voice of the Customer (Firstwe need to determine who our internal and external customers are as related toour project.)

TeamLaunch

Highlevel SIPOC

Front 54

CCR

Back 54

Critical Customer Requirement

Front 55

Boyle's Law

Back 55

Asthe pressure increases, the volume never changes. (The more space we have to“put stuff,” the more stuff we will store.)

Boyle's law states that at constant temperature for a fixed mass, the absolute pressure and the volume of a gas are inversely proportional. (In other words, the product of absolute pressure and volume is always constant.)

Front 56

Central Tendancy

Back 56

Central Tendency is the property thatdata tends to group around a “center”point– This “center” may be themathematical average, most frequent observation, or data point in the center ofthe groupMedian, mean, andmode are common measures of central tendency. Central Tendency relates to accuracy.

Front 57

Bias

Back 57

The term given to the distance between theobserved average measurement and the true value, or “right” answer.

Bias isthe term given to the distance between the observed average measurement and thetrue value, or “right” answer.

Front 58

Precision

Back 58

The extent to which we are able to get the same data values when independent measurements are made on the same entity.

Front 59

Dr. Shewhart

Back 59

Dr. Shewhart of Bell Laboratories developed a theoryof variation that states there are two components to variation: Common Cause& Special Cause. He iscredited with the development of standard control chart based on +/- 3 standarddeviation limits to separate common cause variation from special causevariation.

Front 60

Control Charts for Continuous Data

Back 60

–I-MR,Xbar-R, Xbar-S

Front 61

Control Charts for Binomially Distributed Attribute Data

Back 61

P, NP

Front 62

Western Electric Rules

Back 62

They are a common method of identifying special cause variation.

Front 63

Normal Distribution
(Empirical Rule)

Back 63

99.73% of the data is within three standard deviations of the mean; 95.46% within two SD; and 68.26% within one SD.

Front 64

Rational Subgrouping

Back 64

The process of selecting a subgroup based upon “logical” grouping criteria or statistical considerations

Front 65

Minitab Xbar-R

Back 65

Used with continuous data.

Stat

Control Charts

Variable Charts for Subgroups

Xbar-R

Will need to enter variable(s) and subgroup size.

Front 66

Minitab P Chart

Back 66

Charts the proportion of defectives in each subgroup. Used with binomial attribute data.

Stat

Control Charts

Attributes Charts

P

Will need to enter variable(s) and subgroup size.

Front 67

Special Cause Variation

Back 67

It is due to assignable causes that we canidentify. It is not random and changesover time. It is usually caused by an force acting upon the process from theoutside. This variation can be eliminated by eliminating the outside force acting on the process. Before reducing or eliminating, first clearly identify the source and root cause of the variation.

Front 68

Common Cause Variation

Back 68

It is random, stable, and consistent over time. It is an inherent part of the process itself and can only be changed bychanging the process itself. Sincemanagement owns and creates the process, it is up to management to change theprocess to minimize this variation.

Front 69

Tampering

Back 69

Tampering,or over-control, is when adjustments are constantly made to a process based onindividual measurements, test results, etc.

Front 70

Statistical Control

Back 70

Statistical control means a process iscontinuing to operate as expected, exhibiting only a stable, predictable amountof common cause variation.

Front 71

Descriptive Statistics

Back 71

The field of statistics that defines or characterizes a population based uponthe data points ( i.e., values) taken from that population.

Front 72

Parameters

Back 72

The terms used to describe the key characteristics of a population.

Front 73

N

Back 73

The letter N is used to describe the number of values in apopulation (the population size) when the population size is not infinite.

Front 74

Time Series Charts

Back 74

Time series charts showthe individual data values plotted in the sequential order in which they weregenerated–Run charts and Control Charts arecommonly used time series graphs

Front 75

Accuracy

Back 75

Theability to stay on target as measured by the mean

Front 76

Precision

Back 76

Consistencyof a process as measured by the standard deviation

Front 77

Right and Left Hand of Six Sigma

Back 77

Central Tendency and Variability

Front 78

Simple Time Series Plot

Back 78

Used when we have only one variable and no groups.

GRAPH

TIME SERIES PLOT

SIMPLE

OK

Front 79

Charts Used with Continuous Data

Back 79

I-MR and X-bar-R

Front 80

Charts used with Binomial Attribute Data

Back 80

P and NP Charts

Front 81

Charts used with Discrete Attribute Data

Back 81

C and U Charts

Front 82

Process Capability

Back 82

Quantifiable comparison of VOC (Spec Limits) and VOP (Control Limits)

Most measures have some target value and acceptable limits of variation around the target. The extent to which the "expected" values fall within these limits determines how capable the process is of meeting its requirements.

Front 83

MINITAB CHEAT SHEET

Back 83

http://beauchampsixsigma.com/wp-content/uploads/cheat-sheets.pdf.

Front 84

CP/CPK

PP/PPK

Back 84

(These are terms used to discussprocess capability. CP – example, thermostat on wall. Temp should be 70 +/- 2 degrees.Use thermostat to adjust on the fly. Doesn’t have to stop anything to make the adjustment. If he actually has to go to the AC unit andshut it down to make the adjustment, use CPK (The K = the constant. Thatrelates to the process shifting up to 1.5 standard deviations.). CP = short term (Collected data over a certain aperiod of time); PP = long term (What is the process capability over the longterm?) PP extrapolates from the current process to see what the long-termpredicted stability is.

Front 85

Control Limits

Back 85

The horizontal lines above/below the mean on a control chart. They come from our historical data.

Front 86

Measurement System Analysis

Back 86

To validate measure system is sound. We want to make sure we do not have too much variation.

Key part of the measure phase of the DMAIC Process.

Helps us to verify and validate how much variance we have in our measurement system. (Exam Question)

Front 87

Why measure?

Back 87

Establish the current process level (baseline).

Determine priorities for action -- and whether or not to take action; substantiate the magnitude of the problem.

To gain insight into potential causes of problems and changes in the process.

Prevent problems and predict future performance.

Front 88

Measurement System

Back 88

Helps us make a data-driven decision about a process and change the process.

They help us to see processes clearly.

Front 89

FMEA

Back 89

0 no control
1 remote

3 moderate

9 strong

Front 90

Discrimination or Resolution

Back 90

Dependent upon how much accuracy we need. Ex: Olympic swimmer v. regular swimmer. We want our MS to be able to distinguish changes in 1/10 of range of the specification (LSL and USL).

Ex: LSL = .80; USL = 1.20 Range = .40 (Acceptable Discrimination in this example is .04)

Minitab (Number of Distinct Categories must be greater than four.)

Front 91

bias

Back 91

It is an effect. Another term for this is error. The definition between the true value and the value captured by your measurement system.

Linearity and stability can help offset bias. Calibrate. Can come from stability, linearity, or the person doing the process.

Front 92

stability

Back 92

The stability of the gauge itself over time. (Ex: bathroom scale with spring)

Front 93

linearity

Back 93

Ex: I know what my true weight is. I step on scale and it shows I am two pounds lighter than I actually am. (Always off by a set amount.)

Front 94

Calibration

Back 94

To adjust linearity and stability.

Front 95

Gauge (Gage) R & R

Back 95

It is a controlled study. Used with continuous data. Addresses the Precision of the Process. (It gauges the precision of our measurement system.)

Reproducibility (Takes more than one person).

It is expressed as a percent. Over 30% is unacceptable.

Front 96

Repeatability

Back 96

The inherent variability of the measurement system. It is the variation that occurs when successive measurements are made under the same conditions: same part, same characteristic, same person, same instrument, same set-up, and same environmental conditions. (Ex: Jack measuring same highlighter ten times; he does not know he is measuring same highlighter.) Want to remove as much bias as possible.

Front 97

MSA Standards for Gage R & R

Back 97

• The AutomotiveIndustry Action Group (AIAG) has two recognized MSA standards for Gage R&R:

– Short Form – Five parts measured two times by two differentoperators

– Long Form – Ten parts measured three time each by threedifferent operators

• For good insightinto Gage R&R, go to www.aiag.org

• Remember that theMeasurement System is acceptable if the Gage R&R variability is smallcompared to the Process or Study Variation seen.

Front 98

Distinct Categories

Back 98

In a Gage R & R Report, the number of distinct categories should be greater than four. The distinct categories show the capability of the gauge to recognize differences.

Front 99

Historical Standard Deviation

Back 99

An option in Gage R & R. It references the part we are making not our measurement system.

Front 100

Gage R & R Study (Crossed)

Back 100

This is our go-to feature in Minitab. Gage R & R Study (Nested) is for doing a destruction test (where you can only measure something once). Ex. How much pressure can this part take before it breaks.

Front 101

Gauge Study

Back 101

Stats
Quality Tool

Gauge Study

Gauge R & R Study (Crossed)

Front 102

Analyze Phase

Back 102

Purpose is to drill down to the root cause. Tip: Focus on the time trap.

-identifythe few critical cause-and-effect relationships that explain most of therelationship between the key process input variables and the key process outputvariable.

-Start with subjective analysis and move to statistical analysis.

Front 103

Cause and Effect Matrix

Back 103

Used to develop an understanding of the greatest sources of variation within the process; pinpoints the critical few key process input variables (the X's) that must be addressed to improve the key process output variables. It is a matrix not a diagram. A subjective, stand-alone tool. Helps us get to root cause. Helps us determine what to focus on.

Front 104

Fishbone Diagram

Back 104

Cause and Effect (Remember 6M -- helps us categorize)

The 6 M’s are a mnemonic tool used primarily during the creation of a cause & effect diagram.

The 6 M’s are:

* Machines
* Methods
* Materials
* Mother Nature
* Manpower (People Power)
* Measurements

Front 105

Five Whys

Back 105

By repeatedly asking the question “Why” (five is a good rule of thumb), you can peel away the layers of symptoms which can lead to the root cause of a problem. (Can apply with fishbone diagram, or can stand alone.)

Front 106

Brainstorming

Back 106

Used to generate ideas.

Front 107

FMEA
(It is a form but is also a process.)

Back 107

Failure Mode Effects Analysis

(Started in our country in WWII, Manhattan Project. Used to develop atomic bomb and chemical warfare program.) Purpose is to head off Murphy's Law. (Severity, Occurrence, and Detection -- all assigned a value. RPM (Risk Priority Measure) -- the number we get when we multiply these three values.

Front 108

5S

Back 108

1.1 Sort

1.2 Systematic Arrangement (Set in Order)
(A place for everything; everything in its place.)

1.3 Shine (keeping things clean, painted, etc.)

1.4 Standardize (labels, signage, color-coding)

1.5 Sustain (Make part of our culture; keep in proper order) -- hardest "s"

Front 109

Pearson Correlation Value

Back 109

In statistics, the Pearson product-moment correlation coefficient (sometimes referred to as the PPMCC or PCC or Pearson's r) is a measure of the linear correlation between two variables X and Y, giving a value between +1 and −1 inclusive, where 1 is total positive correlation, 0 is no correlation, and −1

Front 110

Fitted Line Plot

Back 110

The "fitted line" is the running average of the data points.

We need to select "storage" when we run this and store (residuals and fits) so that we can later run a residual analysis.

Front 111

Adjusted R Sq

Back 111

Adjusted R Sq is to R Sq as PP is to CP

(More appropriate to use Adjusted R Sq)

Front 112

Simple Regression

Back 112

Input/output = continuous.
Looks at the correlation between the X and the Y. Determines the slope.

Front 113

Residual Analysis

Back 113

Once we run a regression analysis, we need to do a residual analysis just to verify the model we use is healthy.

Front 114

Hypothesis Testing

Back 114

Allows us to determine statistically whether or not a value is cause for alarm (or is simply due to random variation).

Sample size is important.

Front 115

Alpha Risk

(On Exam)

Back 115

.05 is representative.

Type I error.

"I've discovered something that really isn't here."

Innocent man convicted.

(producer risk)

Front 116

Beta Risk

(On Exam)

Back 116

Type II error.

"I've missed a significant effect."

Guilty man set free.

(Consumer risk)

Reverse of Type-I error.

Front 117

ANOVA

Back 117

Input is categorical and output is continuous.

Testing more than two things with ANOVA.

One factor with more than two levels. (Factor and Levels = terms that will be on the exam.)

Checks variation within (ex. individual team), between (ex. between teams), and total.

Front 118

Sir Ronald Fisher

Back 118

Developed the ANOVA mathematical model to help us determine which X's have more impact on the Y.

Front 119

Chi-Square Test for Association
(AKA -- Pearson Chi Square Test)

Back 119

Chi-squareTest for Association is a (non-parametric, therefore can be used fornominal data) test of statistical significance widely used bivariate tabularassociation analysis. Typically,the hypothesis is whether or not two different populations are different enough in some characteristic or aspect of their behavior based on two random samples.

Front 120

Chi -Square Goodness-of-Fit Test

Back 120

Chi-square Goodness-of-fit Test is used to test if anobserved distribution conforms to any particular distribution. Calculation ofthis goodness of fit test is by comparison of observed data with data expectedbased on the particular distribution.

Front 121

Null Hypothesis (for Chi-square)

Back 121

• Ho:The null hypotheses (P-Value > 0.05) means the populations have the sameproportions.

Front 122

Alternate Hypothesis (for Chi-square)

Back 122

• Ha:The alternate hypotheses (P-Value <= 0.05) means the populations do NOThave the same proportions.

Front 123

FEMA

Back 123

Failure, Mode, Effects, Analysis

RPN = P (probability of occurrence) X S (severity) X D (detection)

The higher the number, the more risk.

(FEMA is all about reducing risks.)

Front 124

FMECA (same as FEMA)

Back 124

Failure, Mode, Effects, Criticality, Analysis

Front 125

VOC (Voice of Customer)

Back 125

VOC = Voice of Customer = House of Quality = QFD (Quality Function Deployment)

The VOC is the upper and lower spec limits.

The VOC sets the expectations.

Front 126

Field of Quality

Back 126

Is about reducing risks.

Front 127

Gage R & R

Back 127

Tells us how effective our measurement system is.

Front 128

ONE-WAY ANOVA

Back 128

Type of hypothesis testing. With this type of test, we are really trying to decide is whether any of the means are different. It is a statistical method for comparing the means as of more than two levels when a single factor is varied.

Front 129

Pearson Correlation

Back 129

1 = perfect "positive" correlation

0 = no correlation

-1 = perfect negative correlation

Front 130

Capacity

Back 130

Themaximum amount of service a process can deliver over a continuous period oftime6&qg=728&qi=1366&qj=728&qk=0&8#3B

Front 131

Constraint

Back 131

A time trap that is unable to produce at the exit rate required to meet customer demand (internal or external).

Front 132

Time Trap

Back 132

Anyprocess step that inserts delay time into a process. We are concerned with the time trap thatinjects the MOST delay.–Example: our property appraisers evaluate 120properties per day, all other process steps can process 145 applications per day

Front 133

Three Basic Principles of LEAN

Back 133

Takt, Flow, and Pull.

Front 134

Takt

Back 134

German word that means beat or count in music.

It is the pace of customer demand.

Determined by the customer.

Front 135

Takt Rate v. Time Time

Back 135

Reverse numerator and denominator in calculation.

Takt Time (How often do we need a (one) student to be completing registration?)

Takt Rate (X number of units -- multiple units)

Front 136

Flow

Front 137

Work Control Systems

(aka CONWIP)

Back 137

WorkControl Systems limit the amount of Work-In-Process inventory in order to control Process Lead Time.

Front 138

PCE

Back 138

Process Cycle Efficiency

(The percent of PLT that is value-added).

It is a reference number.

When we increase PCE, we compress PLT.

PCE = VAT/PLT

VAT (value-added time); this time does not change.

Front 139

Kanban

Back 139

A Japanese word, meaning “Sign”or “Signal” Kanban is considered to be a“Pull” Inventory Replenishment System.

(Ex: Replenishing pork 'n beans in a supermarket)

Front 140

One-Piece Flow

Back 140

The ultimate goal in LEAN

(Where you cannot go to one-piece flow, you kanban.)

Kanban are replenishing signals.

Front 141

Pull System

Back 141

The PullSystem isa flexible and simple method of controlling/balancing the flow of resources.Eliminates waste of handling,storage, expediting, obsolescence, repair, rework, facilities, equipment,excess inventory (work-in-process and finished)

Front 142

Components of Good Pull System

Back 142

*Production based on actualconsumption
* Small lots
*Low inventories
*Management by sight

* Better communication

Front 143

Two Types of Lead Times Related to Kanban

Back 143

Production Lead Time

Transport Lead Time

Front 144

Transport Lead Time

Back 144

The amount of time it takes totransportthe part from the time the signal is sentto thetime the part is received.

Front 145

Production Lead Time

Back 145

The amount of time it takestoproduce the part fromthe time the signal is senttothe time the part is received. Production lead time includes the transportand production time of the part.

Front 146

Push System (MRP)

Back 146

Materials Requirements Planning

--a complex computerized inventory production planningsystem with rules and procedures that supportpush, or batch manufacturing.

Front 147

Kanban Cards

Back 147

Triggers reorder.

Includes the ordering information.

Barcoding on kanban cards is GREAT.

Front 148

Gas Gauge Type Kanban Board

Back 148

Typically used for internally replenished inventory.

Front 149

Kanban Mailbox

Back 149

See-through. Should be able to see from a distance if there is a card in it.

Front 150

Safety Stock

Back 150

Covers us for variation in demand during lead time.

(Unit of measure for usage, lead time, and safety stock need to be in the same units. Ex: all in hours)

Front 151

Kanban

Back 151

Better connects the supplier to a customer.

(Should know how to calculate the Kanban quantity on the exam)

Front 152

What is SMED

Back 152

Single Minute Exchange of Design

Front 153

Benefits of SMED

Back 153

Lessadjustments means less chance for errors.•Elimination of trial processing reducesmaterial waste.•Preparation of operating conditions in advancehelps stabilize product quality.•Increased scheduling flexibility/ capacity.• Reduces the need for an on-hand inventory.• Smaller runs means less likelihood of largescale defects in inventory.•Improved service levels for our customers.

Front 154

Set-up Time (Change-over Time)

Back 154

“Set-up time is the total timebetween the lastgood piece off of a run and the first goodpieceoff of the next run.”

Front 155

Set-up Time (Classifications)

Back 155

Set-up time can be classified as:Internal Set-up:Those activities that must beperformed while themachine is idle or shut down. Example:Removing tooling, fixture, or die.External Set-up:Those activities that are performedwhile the machineis operating. Example:Preparing tooling for next set-up.

Front 156

SMED Methodology

Back 156

1. Observeand document the current set-up. 2. SeparateInternal and External elements.3. Developan Improvement Plan for each element. 4. Observeand document the new set-up process. 5. Standardizethe new set-up procedure.6. Celebrateyour success!!!!

Front 157

Seven Rules of SMED

Back 157

Rule 1 SMED begins and ends with the 5S’s.Rule 2 Convert Internal set-up to External, then improve the remaining Internaltime.Rule 3 Bolts are our enemies (if required,standardized).Rule 4 If you must use your hands, make sure yourfeet stay putRule 5 Don’t rely on special fine tuning skillsRule 6 Standards are not flexibleRule 7 Standardize all set-up procedures.

Front 158

TPM

Back 158

TPMis a common sense approach toproactively maintain equipment, eliminate unscheduled downtime, and improve thelevel of cooperation between Operations and Maintenance.

Front 159

Preventive Maintenance

Back 159

Preventive Maintenance is a timeor usage based method of maintainingequipment. Maintenance activities are performed on equipment based on definedtime and/or usage intervals to prevent equipment breakdowns from occurring.PreventiveMaintenance Schedules

Front 160

Predictive Maintenance

Back 160

Predictive Maintenance is a situation based method of maintainingequipment. Maintenance activities areperformed on equipment based on visible signals or diagnostic techniques toprevent equipment breakdowns from occurring.Vibration AnalysisLaser MeasuringUltrasoundGenerator TestingThermographyOil Analysis

Front 161

OEE

Back 161

Overall Equipment Effectiveness (OEE) isthe measure of the percent of time a piece of equipment is producing quality product at the designed rate.

Front 162

Evaluation Matrix

Back 162

A tool you could use to rate/evaluate alternative solutions.Helps you compare solutions.

Front 163

Pilot

Back 163

A pilot is a test of a proposed solution. This type of test has the following properties:–Performedon a small scale–Usedto evaluate both the solution and the implementation of the solution–Purposeis to make the full scale implementation more effective–Givesdata about expected results and exposes issues in the implementation plan.

Front 164

Tukey

Back 164

Tukey Pairwise Comparisons answer the question "which ones are statistically significantly different?" ANOVA shows if there is a difference, but Tukey actually shows where the difference is.

Front 165

Measurement System

Back 165

Measurementsystems are like eyeglasses, when the lenses are incorrect, the vision isblurred. A measurement system allows usto “see” the process. When a measurement system is poor, we lose the ability tomake good decisions about how to improve the process.

Front 166

Measurement System (Purpose)

Back 166

Thepurpose of a measurement system is to better understand the sources ofvariation that can influence the results produced by the process underinvestigation.

Front 167

Measurement System (Common Key Measures)

Back 167

The two most common key measuresassociated with a measurement system are accuracy and precision.

Front 168

Kappa Value

Back 168

Kappa value should be at least .7

Front 169

Accuracy

Back 169

The extent to which the average of the measurements deviate from the truevalue. In simple terms, it addresses the question, “On average, do I get the‘right’ answer?” If the answer is yes,then we say that the measurement system is accurate. If the answer is no, then we have aninaccurate measurement system.

Front 170

Bias

Back 170

the term given to the distance between the observed average measurement and thetrue value, or “right” answer

Front 171

Reputability
(Gage Precision)

Back 171

Theinherent variability of the measurement system. It is the variation that occurs when successive measurements are madeunder the same conditions

Front 172

Severity, Occurrence, Detection

Back 172

Severity–Importanceof the effect on customer requirements.•Oftencan’t do anything about this. (1 not severe; 10 very severe) •Occurrence–Frequencywith which a given cause occurs and creates failure modes. (1 not likely; 10 very likely) •Detection–Theability of the current control scheme to detect or prevent a given cause. (1 likely to detect; 10 not very likely to detect)

Front 173

5S (Second S)

Back 173

Storage -- A place for everything. Everything in its place. It is all about having the visuals in place.

Front 174

ChiSquare

Back 174

A statistical test that is used to compare observed data with the data we would expect to obtain. Were the differences between the observed and expected the result of chance or some other factor?

Front 175

Residual Analysis

Back 175

Once we run a regression analysis, we need to do a residual analysis to verify the model we used is healthy.

Front 176

Pearson Correlation Value

Back 176

Helps quantify the relationship of X to Y.
+1 = perfect, positive correlation.

0 = no correlation.

-1 = perfect negative correlation.

Front 177

R Square

Back 177

Based on data set you gather.

Front 178

Adjusted R Square

Back 178

Extrapolates from R Square; more appropriate to use than R Square.

Front 179

Predictor Interval

Back 179

Looks at one data point at a time. It is a stronger data tool to use for prediction.

Front 180

Residual Plots for Y

Back 180

The dots are the residuals -- not the data points.If all the dots follow closely to the line, there is a normal dist. of the residuals. (Versus fits) -- scattered = good; do not want to see a pattern here.

Front 181

Hypothesis Testing

Back 181

Things are not always as they appear.
We test for the null. If we reject the null, we accept the alternative hypothesis. Sample size is important. (Increase sample size to make the right call when testing hypotheses.)

Front 182

Box Plot (Interpreting)

Back 182

The cross line is the median. The plus with the circle is where the mean is.

Front 183

P Value

Back 183

If it is less than .05, we can reject the null. (Remember: If it is low, the null must go. If it is high, we keep the guy.)

Front 184

Factors & Levels (On Exam)

Back 184

With one-way ANOVA, still only looking at one factor (like teams) but may have multiple levels (the different teams). Inputs (team one and team two -- categorical) outputs (time -- continuous). Follow ANOVA with a 2-sample t-test.

Front 185

Alpha Risk

Back 185

This is also known as the PRODUCER'S RISK. It is a Type-I error. (.05). Calling something that is good bad. Example: convicting an innocent man.

Front 186

Beta Risk

Back 186

This is also known as the CONSUMER'S RISK. It is a Type-II error. (.95). Calling something that is bad good. Example: Setting a guilty man free.

Front 187

The zero plane1

Back 187

In Tukey's test. If the horizontals cross the vertical (0), there is no statistically significant difference. In the carpet sample exercise, there is a statistically significant difference between 4 and 2.

Front 188

Chi Square (Hypotheses)

Back 188

• Ho:The null hypotheses (P-Value > 0.05) means the populations have the sameproportions.

• Ha:The alternate hypotheses (P-Value <= 0.05) means the populations do NOThave the same proportions.

Front 189

Chi Square Critical Value

Back 189

3.841 (If it is lower than this, we accept the null)? Ask Kris.

Front 190

Statistics

Back 190

(LB) -- They are associated with a sample.

Front 191

Systems

Back 191

Usually made of multiple processes.

Front 192

Statistical Control

Back 192

Stable and Predictable

Front 193

Discrete Distributions

Back 193

Binomial (Binary): Data that is used to track defective items. Example: Students who fail.

Poisson: French Word. Used when looking at defects. Can have several defects related to one defective item. Example: commonalities among students who failed. Hypogeometric: Sampling w/o replacement.

Front 194

Inferential Statistics

Back 194

Inferring something from a larger sample group based on a sample.

Front 195

The Field of Quality

Back 195

Is involved with minimizing risk.

Front 196

Robust Design

Back 196

Designed so that it is almost immune to noise. Taguchi. Minimizes the impact noise has on an output. Noise factors: hold on account, financial issues, transportation issues, life issues.

Front 197

Calculated P Value

Back 197

Regardless of the test in Minitab, we will get a calculated P value. If the "actual P" is less than .05 (the reference P), then we can reject the null.

P Values are really about probability.