Beruflich Dokumente
Kultur Dokumente
Term
(nr)
1FAT
10:1 Rule
100% inspection
3.4 ppm
4-M diagram
5S
7 Deadly Diseases that
management must cure
A8402
Aa
Ac
Acceptance Sampling
2 0f 2 lines
Acceptable quality level
(AQL)
Accuracy
Achieved Availability
Active Listening
Affinity Diagram
Agile Manufacturing
Ai
Alpha risk
Altschuler, G.
Analysis of Variance
Analytic Study
Andon board
ANOVA
ANSI Standards
ANSI/ASQ Q9000, 1, 2, 3, 4
ANSI/ASQ Z1.4
ANSI/ASQ Z1.9
Ao
AOQ
AOQL
AQL
ARIZ
Arrow Diagram
AS9100
ASQ Standards
Assignable cause
Attribute
Attribute (Binomial)
Attribute (Poisson)
Attribute Charts
Attribute Data
Audit Charter
Auditor Candidate
Audits
Audit - Vendor
Availability (equipment)
b
Balanced Scorecard
Barriers to Improvement
Bathtub curve
Bayes Theorem
Bayesian Sampling
Bell Curve
Benchmarking
Benefit-Cost Analysis
Bernoulli Principle
Beta Risk (b)
Bevel Protractor
Bias
Bimodal Distribution
Binomial distribution
Block Design
Bloom's Taxonomy
Boundaryless Organization
Bowman
Box-Behnken
Boxplots
Brainstorming
Breakthrough Achievement
Burst Point
c chart
C(n,r)
Calibration
CAR
Cause-and-Effect diagram
Cavanaugh, R.R
CCC
CCD
CCF
CCI
CE
Central Tendency
CFM
Champy, James
Change Agent
Characteristic
Chebyshev's Theorem
Checklists
Checksheets
Chi-square test
Ciampa, Dan
Classification of defects
Clearance number
CMM (Coordinate
Measuring Machine)
Cnr
Code
Code of Ethics
Coefficient of Determination
Coefficient of Variance
Combinations
Commodity Producer
Common Causes
Community
Communication Skills
Complaint Data
Components of Variance
(COV)
Compound Event
Compression Test
Concurrent Engineering
Conditional Probability
Confidence
Confidence Interval
Configuration Management
Conflict Resolution
Consequential Metrics
Consumer's Confidence
Contingency Table
Continous Flow
Manufacturing
Continious Sampling
Contract Review
Control Charts
Control Plan
Coordinate Measuring
Machine (CMM)
COPQ
COQ
Correlation Charts
Correlation Coefficient
Correlation Matrix of
Documents
Cost Model for Attribute
Plans
COV
Covariates
Cp
Cpk
Cr
Crash
Creativity Stages
Critical Characteristic
Critical Defect
Critical Path
Critical to Quality
CTQ
Culture
Cumulative Distribution
Function
Customer Data
Customer Needs
Customer Relations
Customer Satisfaction
Customer Surveys
Customer Types
CuSum
CWQC (Company-Wide
Quality Control)
Data Accuracy
Data Analysis
Data Coding
Data Collection
Data Integrity
Data Layer
Data Reporting
Data Types
Defect
Defective
Define Phase
Definitions
Deming Cycle
Dependent events
Dependent Variable
Deployment Flowchart
Descriptive Statistics
Design Input Checklist
Design of Experiment 1
Design of Experiment 2
Designer
Deviation
Deviation Permit
DF (Degrees of Freedom)
DFA
DFM
DFX
Dial Indicators
Discrete Distributions
Dispersion
Distributions
DMAIC
Dodge-Romig Tables
Drawings and Specifications
Dynamic Data
Eckes, G.
Education
Effectiveness (Equipment)
Enumerative Study /
Statistics
Equivalent Item
Error-Proofing
Evolutionary Improvement
EVOP
EWMA
Expected Value
Experimental Design
Exponential distribution
Exponential Weighted
Moving Average
External Customers
F Distribution
Facillitation Techniques
Failure Rate
Fast Follower
Fatigue Test
Fayol, Henri
Field Work
Fishbone diagram
Five Forces of Competetive
Strategy
Fixed Effect
Fixed Sampling
Flow Chart
Flinching
Fluorosopy
FMEA
Focus Group
Force Measurement
Forming
Freedman's Test
F-Test
Gage Blocks
Gage Types
Gamma Radiography
Gantt Chart
Gantt, Henry
Gaussian distribution
GD&T (Geometric
Dimensioning and
Tolerancing)
Gemba
Geometric Dimensioning
and Tolerancing (GD&T)
Goodness of Fit
Goodstein
GR&R
Grade
Hammer, Michael
Hardness Testing
Harry, Mikel
HASACC Plan
Hawthorne Effect
Hayakawa
Hazard Rate
Heteroscedasticity
Holographic Testing
Homoscedasticity
HOQ
House of Quality
Hypergeometric distribution
Hyper-Graeco-Latin Designs
Hypothesis Tests
I (Intersection)
I.D.
IEC Standards
Imai (Masaaki)
Impact Test
Inadvertent Errors
Independent events
Information Distortion
Indicator Errors
Inference Tests
Inspection Efficency
Inspection Levels
Inspection Normal
Inspection Operations
Inspection - Receiving
Inspection - Source
Inspection Types
Inspection visual
Integrated Business
Processes
Interferometry
Interrelationship Digraph
Intersection ()
Interval Data
IQL
Ishikawa diagram
ISO 10011
ISO 10012
JCAHO
Jidohka
JIT (Just-in-Time)
Joint Commission on
Accreditation of Healthcare
Organizations
Just-in-Time
Kaizen
Kaizen Blitz
kanban
Kano Analysis
Kano Model
Kaplan, R.S.
Kendall Coefficient of
Concordance
Kepner-Trego
KJ Method
Kotter, J.P.
KPIV / KPOV
Language Issues
Leading Edge
Lean
Lean Enterprise
Likert Scale
Linearity
Locational Data
Loss to Society
LPD
LTPD
Machine Capability
Maintainability
Major Defect
Malcolm Baldridge Award
Managerial Grid
Management
Communication
Management Responsibility
Management Support
MAR
Market Segmentation
Masaaki Imai
Material Control
Material Segregation
Matrix Diagram
Matrix Organization
MBTI
McGregor, Douglas
MDT
Mean
Measles Chart
Measurement Error
Measurement of
Performance
Measurement Scales
Measuring Instruments
Median
Micrometers
Microwave Testing
MIL-STD-105E
MIL-STD-414
Mind Mapping
Minor Defect
Mission Statement
Mixture DOE
MMT
Mode
Model
MSA
MTBF
MTBMA
MTTR
Muda
Multinomial distribution
Multivoting
Mystery Shopper
NCQA
nCr
Negative Skew
Nelson Funnel
Neutron Radiography
Newman, P.R.
Noise factor
Nominal Data
Non-Conforming Material
Non-Destructive Testing
Non-disclosure agreement
Normal distribution
Normal Inspection
Norming
Norton, D.P.
nPr
Null Hypothesis
OC Curve
Occurance
Odds
On Target
Operating Characteristics
Curve
Operational Availability
Optical Comparators
Optical Flats
Ordinal Data
Organizational Learning
Organizational Memory
Organizational Performance
Goals
Original Quality Tools
Orthagonal Arrays
Orthographic Projection
Ouchi, William
P(A + B)
P(A B)
P(A ~ B)
P(A B)
P(A I B)
P(E)
P(Et)
P(n,r)
p chart
Pa
Paired t-test
Pande, P.S,
Parallel System
Parameter
Pareto Chart
Part Identification
PDCA
PDPC
PDSA
Pencil Whipping
Perceptual Map
Performance Index
Performing
Performance Measurement
Permutations
Plug Gages
Pneumatic Gages
Pnr
Point Estimates
Poisson distribution
poka-yoke
Polymodal
Porter, Michael
Positive Skew
Postaudit
Power
Pp
Ppk
Pr
Pre-Award Survey
Precision
Preliminary Review
prEN 9000-1
Pressure
Primary Metrics
Prioritization Matrices
Probability
Probabiliy of Acceptance
(Pa)
Problem Statement
Procedures
Process Capability
Process Controls
Process Industries
Process Mapping
Producer's Confidence
Project Charter
Project Management
Pulse Echo
QCO
QFD
QSRMD
Qualification process
Quality Characteristic
Quality (Definition)
Quality Assurance
Quality Audits
Quality Characteristics
Quality Circle
Quality Control
Quality Council
Quality Documentation
System
Quality Engineering
Quality Evolution
Quality Function
Deployment (QFD)
Quality Information
Equipment
Quality Management
Principles
Quality Management
Standards
Quality Manual
Quality Philosophies and
Approaches
Quality Plan
Quality Policy
Quality Principles
Quality Proceedures
Quality Program
Quality Systems
Quality Teams
Quality Tools
Quartile
Radar Charts
Radiography
Random Effect
Range
Ratio Data
Rational Subgrouping
Re
Recertification
Reccurance
Receiving Inspection
Record Retention
Recording Checksheet
Reduced inspection
Reengineering
Regression Analysis
Reject
Reliability
Reliability Terms
Repair
Repeatability
Reproducibility
Resolution (for DOEs)
Resonance
Resubmitted Lot
Revision Control
Revolutionary Improvement
Rework/Repair
Ring Gages
Risk Management
(assessment)
Robustness
Rockwell Superficial
Hardness Testing
Rounding
RQL
Run Charts
S/N Ratio
Sample Integrity
Sampling
Sampling Costs
Sampling errors
Sampling Plan
Sampling Risks
Sampling Tables
Scatter Diagram
Scrap
Secondary Metrics
Segmentation
Sequential Sampling
Service Quality
Sharma, A
s-hat
Shear Test
Shewhart Cycle
Shore Scleroscope
Hardness Testing
Sholtes
SI
Sigma Levels
Signal Factor
Simple Event
Simplex/Simplex Designs
Sine Bar
Sink Point
SIPOC
Six Sigma
Slack Time
SMART
SMED
Source Inspection
Sources of variability
Special Causes
Special characteristic
Specification
Stability
Stages of Creativity
Stakeholders
Standard Deviation
Standard Work
Static Data
Statistic
Statistically Designed
Experiment
Statistical Inference
Stochastic variation
Storming
Stratified Sampling
Strategic Planning
Strength Relationship
Symbols
Stress-Strength Interference
Subgrouping (Control Chart)
Supplier Certification
Process
Supplier Communication
Requirements
Supplier performance
Supplier Surveillance
Surface Finish
Surface Plates
Suspended
SWOT
Sx-bar
System of activities to
provide quality of products
and service
Systematic Diagram
Tacit Knowledge
Taguchi (Dr. Genichi) 1924 -
Takt time
Taylor, Frederick
TC
Tchebysheff's Theorem
Team Approach
Team Facillitation
Team Leader
Team Objectives
Team Problems
Team Problem-solving
Techniques
Team Roles
Technical Assessment
Technical Drawings
Technique Errors
Tensile Test
Terrorist
Testing Types
Testing/Training Designs
TGW / TGR
Theory of Constraints
Theory Z
Thread Terminology
Three-Wire Method
Through Transmission
Tightened Inspection
TL 9000
TOC
Torque
TPM
TQ
TQM
Traceability
Trailing Edge
Training
Tree Diagram
Trend Analysis
TRIZ
Trigonometry
T-Test
Tukey
Types of Teams
Ultrasonic Tesing
Union (~)
Validation
Validity
Variability
Variable
Variable Data
Variance
Vendor Certification
Venn Diagram
Verification
Vernier Scale
Vision Statement
Visual Inspection
Visual Standards
Waiver
Weber, Max
Weibull distribution
Work Cell
Work Instructions
X-MR Chart
X-Bar R chart
X-Ray Techniques
Z-Test
Column2 Column3
Many experiments focus on 1FAT (one factor at a time) at two or three levels and try to hold everything else
CQE XI-74
constant (which is impossible to do in a complicated process). "Introduction to DOE"
Historically test accuracy ratio (TAR) of 10:1 was considered acceptable. Today 4:1 or 3:1 is more common.
With regards to minimum requirements - a 10:1 ratio of measuring instrument to part (by tolerance during
measurment) and a 10:1 ratio of standard to measuring instrument (by tolerance, during calibration) makes CQE VI-67, 68
sense. However in the case of automotive suppliers, a 10:3 in measurement might be acceptable and
ANSI/NCSL Z540-1995 sets a 4:1 minimum ratio of measurement standard resolution.
Inspection which specified characteristics of each unit of product are examined or tested to determine
CQE VI-39
conformance to requirements.
'Man-Machine-Material-Method.' See Fishbone Diagram. May also include Measurement and Environment
CQE VIII-4
(5M & E).
"Seiri" - Separate (Sort & Scrap) needed tools, parts, and instructions from unneeded materials to be
removed/scrapped later.
"Seiton" - Straighten (Store and Set in order). Neatly arrange and identify parts and tools for ease of use.
"Seiso" - Shine (Scrub & Sweep). Conduct a cleanup campaign.
CQE VIII-18
"Seiketsu" - Standardize & Spread. As a habit, beginning with self, then the workplace, be clean and tidy.
"Shitsuke" - Systemize and Sustain. Apply discipline in following procedures to ensure regular review /
cleaning / inspection of area occurs (quarterly review on calendar. Map/zones established).
Generally, steps in keeping things neat, clean, and organized.
See Deming (Dr. W. Edwards)
1) Lack of consistency of purpose to plan a marketable product and service to keep the company in business
and provide jobs
2) Emphasis on short term profits
3) Personal evaluation appraisal, by whatever name, for people in management, the effects of which are
devastating CQE II-9
4) Mobility of management; job hopping
5) Use of visible figures for management, with little or no consideration of figures that are uknown or
unknowable
6) Excessive medical costs
7) Excessive costs of warranty, fueled by lawyers that work on contingency fees
See Seven Original Quality Management tools below for more detail CQE VII-2
See Kepner-Trego
Alpha risk. See Producer's Risk
This is a type 1 error = Accuse an innocent man…. (See XI-8) CQE
V-32,40,42
XI-9
Quality Management and Quality Assurance-- Vocabulary. Defines the fundamental terms relating to quality
concepts, as they apply to all areas, for the preparation and use of quality management and assurance
standards.
The number of defectives that (if found in a random sample) allows acceptance of the balance of the lot. The
CQE V-38
defectives must be set aside.
Fixed Sampling:
Taking a fixed sample size from a lot or batch only works if the lot or batch size remains relatively constant.
This will be illustrated later with a set ofOC curves from 10% sample size plans. The only advantage is that a
fixed sample size in inspection is easy to remember. Fixed, small sample sizes are more widely used in
auditing.
Stratified Sampling:
Stratified samples are sometimes more informative than homogeneous samples. When analyzing inventory,
one would not want to put $25,000 motors in the same strata with 2¢ screws. Additionally, one might be
interested in determining the amount of pallet damage in a storage area. There might be a need to sample
more row ends, row corners or bottom pallets in preference to top pallets in the middle
Zero Defect Sampling:
There is a growing interest in zero acceptance number plans for two reasons: the advent of six sigma
CQE V-24,25
methodology and the litigious society that currently exists. Often, companies claim they are producing parts
CQE V-28
with very low ppm failure rates.This assumption is based on varieties of Cpk determinations. In fact, a
company should not make such a claim until they have inspected millions of parts. Plans with zero
acceptance numbers have existed for years. A number of these are evident in the ANSI/ASQ Z1.4-20032
tables that follow. However, many ofthese plans have relatively small sample sizes. Obviously, other plans
with nonzero acceptance numbers and larger sample sizes will better discriminate between quality levels
However, in a product liability case a person can envision themselves being questioned by a lawyer along
these lines, "Let me see if I understand. You passed a lot that had one bad unit out of fifty. As far as you
know, 2% of the entire lot was bad. Is that correct?" Among other sources, the student may wish to refer to
Squeglia (1994) for ANSI/ASQ Z1.9-2003 based plans and Schilling (1982) for TNT (tightened- normal-
tightened) based plans..
Continous Sampling:
The most widely used continuous sampling plan is the original - the Dodge CSP-1 plan. It is carried out on a
stream of product, with production units inspected in order of production. The implementation steps for the
plan are:
• Select the sampling frequency (t). This is the desired ratio between the number of product units randomly
selected and inspected at an inspection station and the number of units passing the inspection station during
the same period.
• Sample units shall be selected at sampling frequency (t) so as to give each unit of product an equal chance
of being inspected. The inspector should allow the interval between sample units to vary somewhat rather
than draw sample units according to a rigid pattern.
• Select the clearance number (i). This is the number of consecutive conforming (i.e., defect free) units
required to adjust inspection actions.
• Begin 100% inspection.
CQE V-24,25
• Inspect (i) units in succession. If no defectives have been found, begin sampling.
CQE V-28
• Randomly inspect a fraction (t) of the units. When a defective is found, revert to 100% inspection until the
clearance number has been met.
Sequential Sampling
Sequential sampling is the most discriminating of the acceptance sampling plans.It involves making one of
three decisions as each sample item is obtained: accept the lot, reject the lot, or continue sampling. One
continues sampling until the cumulative number of defectives crosses either the lot reject limit, or the lot
accept limit. When a limit is crossed, the lot size limit is reached and a new lot begins. Sequential sampling
requires the least average sample size. Consider the following
sampling plan: n = 50, c =2, P = 0.017, at the 0.95 confidence limit using ANSI/ASQ Z1.4·20032
This is the maximum percent defective that is considered satisfactory as a process average by the producer
and the consumer. The probability of accepting an AQL lot should be high. It is also defined as the
CQE VI-33,38
maximum percentage of nonconforming units in a batch or sample that, for the purposes of acceptance
sampling, can be considered satisfactory as a process average.
CQE VI-78
data IX-12,
Same as reproducability, accuracy, on target, or bias. Measure of how close average of a sample compares
inspect V-26
to optimal target (nominal spec.). Not to be confused with repeatability.
meas VI-38
unessary VI-67
Availability: A measure of the degree to which an item or system is in the operable and committable state at
the start of the mission, when the mission is called for at an unknown point in time. The point in time is a
random variable.
Managers may spend 45% of their time listening. Ten tips for good listening:
1) stop talking, 2) put the message sender at ease, 3) show that you want to listen, 4) remove listening
CQE II-92
distractions, 5) empathize with the person, 6) be patient with your response, 7) hold your own temper, 8)
avoid argument and critism, 9) ask questions, 10) stop talking
Problem solving technique. Links ideas as with mind mapping, to form patterns of thought. See Memory
CQE VI-24, 25
Jogger (p. 12-18). Also referred to as the KJ Method in honor of its creator, Dr. Kawakita Jiro.
See Lean
Inherent availability (Ai): This is the ideal state for analyzing availability. The only
considerations are the MTBF (reliability) and the MTTR (maintainability). This
measure does not take into account the time for preventive maintenance and
assumes repair begins immediately upon failure of the system. CQE IV-77
The measure for inherent (potential) availability (Ai) is:
A = MTBF/MTBF+MTTR
Note: MTTR stands for mean time to repair.
Also called Type I risk, alpha risk, or level of significance. Only comes into play if we reject the null. May
cause me to change my process unnecessarily. The probability of rejecting a good lot. 1 - a = Producer's CQE V-32,40,42
Confidence. See also Consumer's Risk. Note: alpha may be calculated for a designed experiment as CQE XI-9
follows: Full Factorial Designed experiment alpha = (2k)1/4, where k = number of factors.
Came up with the ARIZ (algorithm to solve an inventive problem) and TRIZ (sequence of 9 action steps)
methods for solving complex problems with inventive solutions.
Analysis of Variance. Also called ANOVA. Developed by Sir Ronald Fisher to determine if the within-
treatment variation is significant in comparison to the treatment means. One way Anova means one factor.
In a single factor ATI the assumption of homogenity of variance applies to the variance of treatments within
CQE
groups. It is assumed that the column variances are equal. NOTE: ANOVA is NOT robust to non-normal
XI-50:XI-59
populations. See book for equations for normal vs. non-normal populations. Regarding GR&R, see p. VI-96.
VI-80,84:86
Also note: Calculated F value equals row mean square divided by error mean square.
For a given number of degrees of freedom, as the variablity of means increases reative to the variability
within groups the F-ratio increases as well.
A study in which action will be taken on a process to improve performance in the future (per Deming). CQE IX-35
Lighted overhead display showing current status and problems. A subset of the visual factory. CQE VIII-13
Analysis of Variance. Also called ANOVA. Developed by Sir Ronald Fisher to determine if the within-
treatment variation is significant in comparison to the treatment means. In a single factor ATI the assumption
of homogenity of variance applies to the variance of treatments within groups. It is assumed that the column
variances are equal. NOTE: ANOVA is NOT robust to non-normal populations. See book for equations for
normal vs. non-normal populations. Regarding GR&R, see p. VI-96. Also note: Calculated F value equals
row mean square divided by error mean square. CQE
For a given number of degrees of freedom, as the variablity of means increases reative to the variability XI-30, 35/38,50,53
within groups the F-ratio increases as well. XI-59
Note qualitative and quantitative factors can still be calculated even though no numerical scale can be VI-80,84:86
attached to the levels.
Note: t and Z tests are used for single sample means. Chi-square test evaluates sample variances. In an
Anova table the between group variation is compared to the within group variation. Anova utilizes the F test.
ANSI/ASQ standards for everything from QM/QA, to systems guidelines, statistics, etc. See Primer for
CQE III-30-31
extensive list.
Same as ISO 9000, 9001, 9002, 9003, 9004. See definitions in Primer. CQE III-17
Sampling plan for Attributes. Note: Pa is not fixed for various plans. AQL is based on high Pa (83% - 99%)
CQE V-45,47, 52
See MIL-STD-105E
Average Outgoing Quality. Equal to the percent defective x probability of acceptance. Max AOQ = AOQL. CQE V-38
AQL Conversion Table for variable sampling with Ansi Z1.9 is on page CQE V-58
CSSBB
See Altschuler, G.
XI-41/42
Similar to Program Evaluation and Review Techniques (PERT), Critical Path Method (CPM), Node/activity on
node diagrams (AON), Precedence diagrams (PDM), and countless other network diagrams. Used to
CQE VIII-46
identify critical paths, slack time, milestones, etc. Check it out in the Primer if, and only if, it comes up on the
exam.
This standard applies to the aerospace industry and incorporates the European standard of prEN 9000-1.
The use of this standard facilitates an organization's ability to adhere to industry specific safety and reliability
requirements.
See Primer. Basic common-sense approach to professionalism and integrity. CQE II-55
ANSI/ASQ standards for everything from QM/QA, to systems guidelines, statistics, etc. See Primer for
CQE III-30-31
extensive list.
A characteristic or property that is appraised in terms of whether it does or does not exist, with respect to a
CQE V-38
given requirement.
Regarding sampling, an item that does or does not meet inspection standards
See also, Control Charts. Summary of attribute- and variable-type control charts can be found in Memory
Jogger II, page 37. Attribute Chart equations are on page 39. Variable Chart equations on page 40. Table CQE X-29
of constants on page 42.
Counted data. Nearly always a whole number. Answers 'how many' or 'how often'. CQE IX-2,4
Completed audit records include the formal report, report response, and completed corrective actions CQE III-49
Per ISO 10011, auditor candidates must have four years of full-time practical work experience, not including
training, at least two years of which are in quality, and participating in a minimum of four audits, for a total of
at least 20 days.
Purpose/benefits (IV-2). Types of Audits, Old vs. New Methods, Philosophy (IV-3-5). Objectives (IV-6).
General Audit Matrix (IV-7). Responsibilities (IV-9). Glossary of terms (IV-13-16).
Process Audit: Audit designed to evaluate the effectiveness of a quality assurance program examining the
knowledge of, compliance to, and adequacy of current production methods.
Quality Audit: Includes Product, Process, and System audits. Note that each may be included as a subset
of the audit that follows it. Verifies that the Quality Assurance program complies with established
requirements. In general 3 parties are present - Client, Auditor and Auditee. CQE III-42/43.
Compliance Audit: Evaluation for contractual compliance of the quality system.
Assessment Audit: May be a Pre-Award Survey. Evaluation of a potential supplier to provide a product or
service. Limited in scope.
CQE
Management Audit:
III-35:49
Random Audit: Unscheduled and unannounced. Often used to assess emergency responsiveness. Need
for such audits is relatively rare.
Specific Objective / Specific Activity Audit: May be an inspection performance audit.
Product Audit: Customer-oriented sampling of finished goods. Should be oriented around customer
specification limits.
Surprise Audits: Usually occur only where health and safety and/or federal regulations are at stake.
The three broad stages of an audit include 1) Preparation, 2) Performance, and 3) Reporting (CQA-VIII-3)
The primary responsiblity for a follow-up evaluation to a corrective action usually rests with the audit group.
CQE III-42/43 and 49.
If a vendor audit were to focus on design control blueprint and attention to reliability and design capabilites CQE II-13/17, IV-7/9,
and change proceedures would typically be included V-4/7, V-14/17
From the Dodge-Romig tables. Equal to the maximum AOQ. CQE V-38
Includes four domains: Customer, Function, Physical, and Process. In each case, the previous domain
repressents the 'why' and the given domain provides information to the next domain as to 'how' to fulfull its
wants. Very similar to QFD sequence. In order, includes the four steps of 1) establish design objectives, 2)
generate ideas to create solutions, 3) analyze possible solutions for best fit, and 4) implement the selected
design.
Beta risk. See Consumer Risk. Calculatable. If goal is Beta = 0.05, then Z = 1.96 and -1.96. Use z = (x-bar
- population mean) / (sample stdev/sqrt(sample size)). Resulting values (for -1.96 and 1.96) produce the
CQE V-32,38,41
confidence interval where Beta risk is 0.05. A sample mean that falls outside the CI has 95% chance of
actually coming from a different population.
See also Norton and Kaplan. Business planning process that gives consideration to factors other than
financial ones, including 1) Financial, 2) Internal Business Process, 3) Learning and Growth, and 4)
Customer Perspectives. It provides more perspectives for stakeholder interests, and is referred to as the
balanced scorecard. Used to evaluate a company's overall business strategy.
Three general types of failures observed for complex products are described by the life-history or Bathtub
curve. Many electrical and computer based products do not follow this model.
CQE
Infant Mortaility or Burn in period (Weibull distribution is used to determine when this period is over)
IV-79
Constant failure rate or normal operating period (exponential distribution - this period is used for making
reliability predictions)
Wearout period - Component degredation and fatigue (normal and log normal distributions are often used)
P(A|B) = P(A) x P(B|A) / P(A) x P(B|A) + P(A') x P(B|A') Used to find reverse conditional probabilites.
Discovery sampling procedures sometimes used for determining appropriately small but reasonable sample
CQE
sizes during audits (during audits, time is a factor, so sample sizes are generally smaller than inspection
V-44
sampling)
See Normal Distribution. See book for formulas, graph, and applications. CQE IX-15
Decision-making tool that looks specifically at money (sum of all benefits anticipated $ / Sum of all costs
anticipated).
Daniel Bernoulli (1733) stated that "expected value equals the sum of the values of each of a number of
CQE IX-32
outcomes multiplied by the probability of each outcome relative to all the other possibilities."
Type II risk. See Consumer Risk CQE V-37, 38,41
The universal bevel protractor is a hand tool used to obtain an angular reading in degrees and minutes.
The difference between the observed average of the measurements and a reference value. The definition
CQE VI-77, 78
for accuracy almost always includes the phrase "lack of bias". See "Validity"
Big Q is big picture quality-- plant-wide/global systems. Little q is more a focus on product performance to
CQE II-17
specifications. Big Q is improvement systems, integration of quality, customer focus, etc.
When plotted data gives the appearance of two bell curves, side-by-side. Two-humped distribution. Often
jokingly referred to as the "Dolly Parton" distribution. Often caused by special cause such as two cavities in CQE VIII-24
a mold, or the same parts being run off of two different production lines. See also Memory Jogger II (p. 72).
(when sampling with replacement and proportion defective is constant and > 10%, large lot size N>50 and
small sample size relative to lot size n<1/10N) CQE IX-61,63 XI-44,45 XII-77 StatGuide 4-4
CQE IX-61-63, XI-
The outcome from a bernoulli trial has two possibilities, usually referred to as success or failure. The
44,45
probability of success or failure is independent on all Bernoulli trials. The binomial distribution is used to
model Bernoulli trials.
Note a major objection to designed experimentation is that choosing levels of factors that are far removed
from production may in fact produce scrap.
(bottom) Knowledge, Comprehension, Application, Analysis, Synthesis, Evaluation (top) CQE I-18
Adapts workload to particular needs. Allows roles and structures to change based on specific situations
rather than having rigid defintions. However, job descriptions might still exist for some roles.
Central composite design which requires three factor levels (like the CCF; unlike the CCF & CCI which
CQE XI-76
require five factor levels). See Central Composite Desisgn
Five point representation of data. Central line represents the median. Upper and lower quartiles of the data
define the ends of the box (that is, 25% of the data falls in each of the two halfs of the box). The upper and
lower quartiles of the data defijne the ends of the box. Minimum and maximum data points are drawn as
points at the end of lines (whiskers) extending from the box. Outliers are added as asterisks. Mean may be CQE IX-24
shown by a dot. If the data is skewed to the right then the mean will be to the right of the median and the
whisker on the right will be the longest. If the data is skewed to the left the mean will be to the left of the
median and the whisker on the left will be the longest. Invented by John W. Tukey in 1977.
Intentionally uninhibited technique for generating creative ideas when the best solution is not obvious. See
Primer for steps involved and details. Examples include round-robin, freestyle, Crawford Slip, and electronic CQE II-81
(the last two allowing annonymity).
Paradigm shift in design or concept (quartz watches). Best way to take market share from even a
CQE VIII-5
'revolutionary improvement' company. See primer.
See Primer. See also hardness testing method summary matrix in Primer, p. VI-59. CQE VI-59
If points on a c chart are outside of the control limits and one wishes to setup a control chart for future
CQE X-32, 34/35,
production, discard those points falling outside the control limits, which have identified assignable causes
37/43
and revise the limits. Normally twenty five data points are sufficent to establish c chart control limits.
See Combinations.
CQE IX-44 to 45
Crn = n!/r!(n-r)!
The calibration of measuring instruments is necessary to maintain accuracy (lack of bias), but does not
necessarily increase precision (repeatability). In order to improve the accuracy and precision of a
measurement process, it must have a defined test method and be statistically stable. CQE VI-69 to73
Corrective Action Request. Form started by an auditor, to be completed by the auditee, to address major
and minor nonconformances noted during an audit. The form is used to request and record corrective a tion
CQE VIII-35
root cause analysis, temporary and permanent actions, and the time for response. Any format and title will
do, as long as the form and the requireements serve the intended purpose. CARs must be formally
responded to and closed out, with documentation retained and forwarded to the auditor for approval.
Co-Author of 'The Six Sigma Way.' See Pande, P.S. for more information.
Circumscribed Central Composite (Box-Wilson). Higher order design which requires five factor levels. See
Central Composite Design
Inscribed Central Composite (Box-Wilson). Higher order design which requires five factor levels. See
Central Composite Design
CQM Handbook
See concurrent engineering
p.141,145
Includes CCC & CCI. One of several classes of response surface designs. As described by Verseput (2001),
they are popular because 1) sequential runs of linear or curvature effects can be made, 2) they provide much
information from a minimum of runs, and 3) they are quire flexible over different experimental regions. CCC
& CCI are rotatable-- that is, the variance is the same at all points that are equidistant from the design center.
The variance therefore remains unchanged when the design is rotated around the center.
If a random variable X has a mean m and a finite variance s2, as n increases, X-bar approaches a normal
distribution with mean m and variance sx-bar2. Where sx-bar2= s2/n and n is the number of observations on
which each mean is based. CQE IX-16, 17
In words, the distribution of sample averages will approach normal regardless of the shape of the parent
population (this is why x-bar & R charts work).
Measures of central tendency represent different ways of characterizing the central value of a collection of
data.
CQE IX-13-15
Mean - average value
Median - Middle value
Mode - Most often occuring value
Continous Flow Manufacturing - Material moves one piece at a time, at a rate determined by the needs of the
CQE VIII-13,21,22
customer, in a smooth and uniterrupted sequence and without WIP.
See Reengineering. Co-authored book with Michael Hammer (1990). See Primer. CQE VIII-32
Individuals who play a specific role in the planning and implementation of the change management process.
They may be insiders or outsiders. The person who takes the lead in transforming a company into a quality
organization by providing guidance during the planning phase, facilitating implementation, and supporting
those who pioneer the changes.
A property that helps to differentiate between items of a given sample or population. The difference may be
CQE V-38
either quantitative (by variables) or qualitative (by attributes).
CQE IX-21
See Tchebysheff's Theorem. See also Camp-Meidel extension.
Used for inspecting machinery or product, or when learning to operate complex or delicate equipment. A
CQE VII-10
grocery list is a good example. See also Checksheets.
Tools for organizing and collecting facts and data. Usually used to record counted, measured, and locational
data. Three kinds include Recording Checksheets, Checklists, and Measles Charts. For example locational
CQE VII-9-10
data would be recorded on a 'measels chart'. See Primer also for 'Mnemonic' or 'Memory Hook'
Checksheets. See also Memory Jogger II (p. 31-35).
See summary table on p. XI-31. Used to compare a sample variance with a known, same population
variance. Note: If the calculated Chi-square for any test is greater than the critical (table) value, the null
hypothesis is rejected. Degrees of freedom is calculated as (# rows - 1) x (# columns - 1). Contribution to CQE XI-19, 26, 38
the Chi-Square total from one cell is calculated as ((O - E)/E) 2. Required for calculations of Goodness of Fit,
Contingency Coefficient, Freedman's Test, and Mood's Median Test.
Critical, Serious, Major, Minor. See table on CQE V-20 CQE V-20
As associated with a continuous sampling plan, the number of successively inspected units of product that
must be found acceptable during 100% inspection before action to change the amount of inspection can be CQE V-38
taken
The number of distinct combinations of n distinct objects taken r at a time. This may be denoted by the
symbol Cnr, nCr, C(n,r), or (nr). Calculated as n!/r!(n-r)!. Unlike Permutations, Combinations do not take CQE IX-44-45
order into account, therefore ABC=ACB=CAB=CBA=BAC=BCA.
CQE V-45,46
Adding / subtracting a constant or multiplying / dividing by a factor to simplify data collection
CQE II-55,56
ASQ's Code of Ethics
Used in regression analysis. Measured on a scale from 0 to 1. Higher is better. CQE XI-71
standard deviation divided by the mean. Expressed as a percentage (s/x-bar x 100) CQE IX-18
The number of distinct combinations of n distinct objects taken r at a time. This may be denoted by the
symbol Cnr, nCr, C(n,r), or (nr). Calculated as n!/r!(n-r)!. Unlike Permutations, Combinations do not take order CQE IX-44,45
into account, therefore ABC=ACB=CAB=CBA=BAC=BCA.
Company strategy. See Leading Edge for strategy & product types list.
Generally represents about 80% of all causes. Usually not operator dependent, but process-dependent.
Cannot be changed without management assistance. Note that reducing Common Cause variation is one of
the primary strengths of Six Sigma methodology. Demming felt that 94% of all problems are due to common
cause variation. Special causes on the other had affect the process in unpredictable ways and result in CQE X-5
unstable outcomes. Demming felt that 6% of all problems are attributed to special causes. Management
often confuses both special and common causes of variation. The most frequent error is considering
common causes to be special causes and tampering with the process. See CQE X-6
Often used to mean Society as a whole. Often used to describe the Customer(s) for a product or service.
Vertical flow of communications, Horizontal communications, formal and informal communications
CQE II-88,89,90,91
See also communication effectiveness - CQE II-92
For both a product and a service organization complaint data should provide the degree or extent of
CQE II-96/98
customer dissatisfaction
Deals with probability. The complement of an event A is all sample points in the sample space but not in A.
The complement of A is 1-PA. CQE IX-39
May be developed for a ANOVA experiment, like GR&R. Can determine 1) the extent of contribution by each
source of variation, 2) how much of the variance is due to differences in treatment means, and 3) how much CQE XI-58
of the variance is due to experimental error.
Deals with probability. A composition of two or more events. Example the sum of two tossed dice is
CQE IX-38, 40
dependent upon the values of two simple events, thereby making it a compound event.
Working in parallel where possible. Saves time when processes are independent prior to bringing them
together in the design or assembly processes.
The conditional probability of an event A given that B has occurred is P(A I B) = P(A B) / P(B) (assuming
CQE IX-39
P(B) does not equal zero)
CI for mean, see CQE XI-32. For Variation or Proportion, see CQE XI-33.
The confidence interval is for the population mean (µ). A 95% confidence interval says that, for that
interval there is 95% confidence that this interval contains µ.
CQE XI-4 to 6
StatGuide 6-5 StatGuid 6-5
When sample size is small (n<30) and = is not known use Student's t distribution for means
If = is known then use Z to compute confidence interval for means.
For variances use F and Chi-Square
Defined by Juran and Gyrna as "the collection of activities needed to define, identify, manage, record, or
approve the hardware and software characteristics of a product."
Starting point for a configuration is a 'baseline'. The 'allocated baseline' is the general requirements for a
subsystem in the overall product. A 'product baseline' defines the detailed requirements of the system or
item.
CQE III-13,14
Consists of four main elements:
Configuration Identification
Configuration Control
Configuration Accounting
Configuration Audits
Also called Type II risk or beta risk. Only comes into play if we FAIL to reject the null. May cause me to keep
my process the same when I should change it. The probability of accepting a bad lot. 1 - b = Consumer's CQE V-32, 38, 41 XI-8
confidence. See also Producer's Risk.
But the expected (theoretical) value for a cell in a contingency table is calculated as (row total x column total) CQE XI-46, 48
/ grand total. Contingency coefficient requires a chi-square calculation. StatGuide 7-17
(CFM) Material moves one piece at a time, at a rate determined by the needs of the customer, in a smooth
CQE VIII-13,21,22
and uniterrupted sequence and without WIP.
The most widely used continuous sampling plan is the original - the Dodge CSP-1 plan. It is carried out on a
stream of product, with production units inspected in order of production. The implementation steps for the
plan are:
• Select the sampling frequency (t). This is the desired ratio between the number of product units randomly
selected and inspected at an inspection station and the number of units passing the inspection station during
the same period.
• Sample units shall be selected at sampling frequency (t) so as to give each unit of product an equal chance
of being inspected. The inspector should allow the interval between sample units to vary somewhat rather CQE V-28
than draw sample units according to a rigid pattern.
• Select the clearance number (i). This is the number of consecutive conforming (i.e., defect free) units
required to adjust inspection actions.
• Begin 100% inspection.
• Inspect (i) units in succession. If no defectives have been found, begin sampling.
• Randomly inspect a fraction (t) of the units. When a defective is found, revert to 100% inspection until the
clearance number has been met.
Reviews must ensure 1) requirements are defined and documented, 2) any requirement different than the
quote is resolved, 3) the company can meet the contract requirements, and 4) there is a defined method for CQE III-19
amendment of contracts.
Purpose is to detect assignable causes of variation in the process. Random variation is desired on a
control chart.
Method of plotting dynamic (time-dependent) data developed by Dr. Walter Shewhard. More than a dozen
different types exist, for use with either Variable or Attribute data. When to use which Chart (p. 37).
Formulas for attribute data (p. 39) and variable data (p. 40) charts. Table of constants (p. 42). Interpretation
(p. 43-47). Basic rules for out of control are 1) 1 point beyond control (3 sigma) limits, 2) trend of 8 points
in an up or down direction, 3) 8 consecutive points on one side of the center line, 4) 4 of 5 consecutive points
beyond the 1-sigma limits, or 5) 2 of 3 consecutive points beyond the 2-sigma limits.
If calculating probability that a process shift causes a point of an X-bar-R chart, calculate the UCL = X-double CQE VII-16,
bar + A2 x R-bar, Note that Sx = standard deviation for the control chart = (A2xR-bar)/3. Now compute Z X-11, 41, 43, 46, 52
value failure rate.... New mean is the process shift X-double bar value and the UCL becomes a USL. Zupper
= USL-X-double bar/Sx.
Use standard normal table to look up resultant Z value. CQE X-13/15 and CQE 58/63
See page CQE X-7 for tips on selection of control chart varaiables
If points on a c chart are outside of the control limits and one wishes to setup a control chart for future
production, discard those points falling outside the control limits, which have identified assignable causes
and revise the limits. Normally twenty five data points are sufficent to establish c chart control limits.
A written document of the activities controlling the process or product. Focused on production.
The control plan becomes a substitute for written operator instructions.
The following would be likely to find on a control plan form: Specifications, Key input variables, Key output
variables. One would not expect to find 'Potential causes for failure' on a control plan form.
Inputs for process control plans come from (Description of line items on CQE V-8/9):
Process flow diagrams
Design reviews
System FMEAs CQE V-2,8,9
Quality function deployment
Cause-and-effect analysis
Designed experiments
Special customer characteristics
Statistical applications
Historical dataMulti-vari studies
Lessons learned
Regression analysis
Team process knowledge (Breyfogle, 2003)6, (APQP, 2000)
See Primer. See also gage summary matrix on CQE VI-3,4,5 CQE VI-36
To be effective a corrective and preventitive action system must have a method for identifying
nonconformities.
Corrective action steps would include a clear definition of the problem, an identified solution to the problem
and actions necessary to prevent reccurance. It would not include department or individuals responsible as
the proper approach is to fix the system, not to assign blame.
Audit reports, customer returns, and inspection results often indicate a need for corrective action. (Calibration
CQE VIII-33-42
techniques would not..) CQE VIII-36/40
Assign an investigator, Assess the importance, Analyze the problem, then install controls -
See CQE VIII-39/41
The primary responsiblity for a follow-up evaluation to a corrective action usually rests with the audit
group. In many cases audit findings are indications of system failure, so upper management is ultimately
responsible. CQE III-42/43 and 49.
In a large company a comprehensive program might be handling CAR's from multiple sources. This
would likely make the review process rather formal and perhaps in need of prioritization by upper
management. This would like primarily involve the Management Review Team and the Corrective Action
Board as the primary administration and control of the CAR process. CQE VIII-36.
Used in regression analysis, and denoted with the letter 'r' or 'R'. As correlation decreases (scatter
increases), r2 decreases as well.
Example with negatively sloped line:
The correlation coefficient, r, is a measure of correlation. If the data is perfectly, positively correlated, r = 1; if
the data is perfectly, negatively correlated; r = -1. If there is no correlation in the data, then r = 0. The
coefficient of determination is equal to the correlation coefficient squared. In the figure above, r is
negative because as the variable on the x -axis increases, the variable on the y-axis decreases. However,
the coefficient of determination is the square of a negative number, which is positive. Thus, the coefficient of CQE XI-70, 72, 77
determination must be greater than the correlation coefficient.
Often written as 'r'. Calculated to determine the degree of relationship between two variables (see Scatter
Diagrams). r = (S(X-X-bar)(Y-Y-bar))/sqrt((S(X-X-bar)2 S(Y-Y-bar)2). r = -1.0 (strong negative); r = -0.5
(slight negative); r = 0 (no correlation); r = +0.5 (slight positive); r = +1.0 (strong positive correlation). Line of
best fit can be calculated (see Regression Analysis).
where TC = Total Cost, A = Overhead costs, B = Cost/unit of sampling, n max = Max sample CQE V-31
size, C = Cost/unit of inspecting, and n-bar = Average sample size.
No Inspection: Total Cost = NpD where N = number of items/lots, p = % defe
The best reason for implementing a quality cost system is to identify opportunities for improvement.
An improvement in quality costs is most clearly indicated when management objectives are met.
A quality cost system can be used to align quality and company goals, provide an overview of quality, and
help to prioritize company resources. Historically, categories of costs include direct costs (direct materials
and direct labor), indirect costs or overhead (indirect materials, indirect labor, fixed costs, and general &
administrative costs), and selling costs. Prevention of poor quality is far cheaper than appraisal, which in
turn is cheaper than failure (both internal and external).
CQE III-52,55,56
Internal failure costs are costs which occur prior to delivery or shipment of the product or furnishing of a
& 60/61
service to the customer. (labor and material cost documents are a usual source of this info)
External failure costs are costs which occur after delivery or shipment of the product, or during or after
furnhishing a service to the customer. (Returned material reports)
Appraisal costs are teh costs associated with measuring, evaluating, or auditing products or services to
assure conformance to quality standards and performance requirements.
Prevention costs are the costs of activities specifically designed to prevent poor quality in products or
services. (purchase orders would be a neutral or if reviewed properly may be preventitive)
COV = S/X-bar * (100%) or COV = =/ * (100%). See Coefficient of Variation CQE IX-18
Cp = (USL-LSL)/6=R. =R = R-bar/d2 and is an estimate of process capability sigma and comes from a
control chart. USL is upper specification limit, LSL is lower specification limit.
Higher Cp is better. Rule of thumb - Cp > 1.33 capable, Cp=1.00 to 1.33 Capable with tight control, Cp < CQE X-64,65
1.00 incapable. Capability Ration is inverse formula of Cp. C R < 0.75 Capable, CR = .75 to 1.00 Capable with
tight control, CR>1.00 Incapable… See Process Capability. d2 values come from Control Chart Factor table
Cpk=USL-X-bar/3=R or X-bar-LSL/3=R which ever gives the smallest answer. USL is upper specification limit,
LSL is lower specification limit.
Cpk = 1 is 3 sigma, Cpk of 1.33 is 4 sigma, Cpk of 1.66 is 5 sigma and Cpk of 2 is 6 sigma. See Process
Capability
Activity-oriented Project Management tool. Applies normal and 'crash' costs (costs associated with applying
more resoureces to complete the project in a shorter amount of time). Similar to PERT (which is event- CQE II-46, 48
oriented).
See PERT diagram - Application of more money or resources to bring a particular task in early CQE II-43,45
1) Generate: Create a list of as many ideas as possible, 2) Percolate: Allow time to think over these and
CQM Handbook
other ideas, 3) Illuminate: Return to the list and discuss what has been "discovered." Add/delete/combine,
p.120-121
etc., 4) Verify: Test out some fo the ideas that appear more feasible.
A product or process characterisic applicable to a component, material, module, subsystem, or system which
is defined by engineering as being critical to the product and having unique safety, quality, reliability, CQE V-2
durability, and regulatory criteria.
A defect classification that experience or judgment indicates is likely to cause unsafe conditions for those
who use, maintain, or depend on the product; or a defect likely to prevent performance of the end-item CQE V-18
function. Example: a defect that can cause automobile brakes not to perform.
In time sequence, this includes 1) Identify the Customers, 2) Identify the Customers needs, 3) Identify the
Customers' basic requirements, 4) Add additional requirements as needed, and 5) validate the requirements
with the Customer. Per Eckes (2001).
Defined quality as "Conformance to requirements." Known as the "Quality Executive" and proponent of
"Zero defects." Big believer in senior management taking ownership of quality (like Juran). Preaches four
absolutes of Quality Management (see Primer, page II-13). Also responsible for the 14 step approach to CQE II-4,6
quality deployment (Primer, p. II-13). Advocated a core of quality specialists within an organization (see
points 2 and 13 of his 14 points).
In time sequence, this includes 1) Identify the Customers, 2) Identify the Customers needs, 3) Identify the
Customers' basic requirements, 4) Add additional requirements as needed, and 5) validate the requirements
with the Customer. Per Eckes (2001).
Shared values, norms, and assumptions. Visible as orientation to power, risk, mistakes, and outsiders.
Shaped by policies and procedrues, structure, day-to-day management decisions, and rewards.
A graphical display of total percentage of results below a certain measurement is called the cumulative
distribution function. This is different than a probability density function or histogram which display CQE IX-29/30
probabilities over a range of values.
Gather Data: Surveys, Focus Groups, Face to Face Interviews, Satisfaction / complaint cards,
Dissastisfaction sources, competitive shopper
CQE II-98,99
Analyze Data: Statistical tests, line graphs, control charts, matrix diagrams, pareto analysis.
Juran lists these as: Stated Needs, Real Needs, Perceived Needs, Cultural Needs, Unintended Needs.
See Kano Model. For a list of pitfalls in determining customer satisfaction CQE II-95-98
See CQM Handbook for definitions including Primary Customer, Secondary Customer, Indirect Customer,
External Customer, Consumer/End User, Intermediary, False Customer, and Internal Customer.
Cumulative Sum control chart. Most sensitive chart. May use a V-mask if manually generated. CQE X-22,24
Self explanatory. Reducing time improves productivity and reduces costs. See primer for case study in
CQE VIII-15,16
Single Minute Exchange of Dies (SMED).
Adding / subtracting a constant or multiplying / dividing by a factor to simplify data collection CQE IX-11
Link across different media, such as audio, video, text data, and data bases. See Knowledge Management.
Attribute - data is discrete, can be a count of how many, or color, good or bad IX-2
Variable - continous numerical data IX-3
Locational - where a defect physcially occurred IX-3
Ordinal - data is arrainged in order, however numerical value is not signification IX-7 CQE IX-4
A unit of product that contains one or more defects, at least one of which causes the unit to fail its
CQE V-38,43 X-30
specifications
Together with Measure, should represent about half of the total project time. Define phase should identify
key processes, goals, scope, and project priority, among other items.
For a list of Quality Terms and Definitions, see Primer, II-4-11. Audit terms, Primer IV-13-16. Quality
Planning & Control Terms, Primer V-2-3. Sampling terms, Primer VI-5-6. Force/Hardness testing definition,
CQE II-4 - 11
VI-81-86. Calibration definitions (VI-89). Sampling & Measurment definitions (VI-104-106). Reliability (VII-
27-28).
"Quality control does not mean achieving perfection. It means the efficient production of quality that the
market expects."
Founder of the third wave of the industrial revolution (quality explosion). Big into Quality Philosophy.
Strongly associated with the Japanese and JUSE (Japanese Union of Scientists and Engineers).
CQE II-4,7,8,9,10
Well known for his:
Fourteen Obligations of Top Managment (CQE II-8) (Note all are required! Can't do just one!)
Seven Deadly Diseases that Management must Cure (CQE II-9)
Deming's Chain Reaction (CQE II-10).
When P(A I B) does not equal P(A). Ex. 20% of the time it rains. 30% of the time it's cloudy. 67% of the
CQE VIII-27,
time when it's cloudy, it rains. P(A I B) = 67%, but P(A) = 20%. Rain must be dependent upon clouds. See
IX-40
independent events.
A factor that results in a response variable and can be equated to an independent variable.
Numerical descriptive measures create a mental picture of a set of data. These measures calculated from a
sample are numerical descriptive measures called statistics. When they describe a population, they are CQE IX-13
called parameters. The two most important measures are of central tendency and dispersion.
Consider following items:
Primer, p. IV-7
Note a major objection to designed experimentation is that choosing levels of factors that are far removed
from production may in fact produce scrap.
CQE XI-74:115
Stating that levels of a factor are fixed indicates that the levels are set at certain fixed values. Page 8-7 in
StatGuide StatGuide
Chapter 10
For factorial experiments, confounding takes place when we run a fractional part of the complete experiment.
Also for factorial experiments all levels of one variable are run at each elvel of the second variable.
Responsible for a product from initial concept to final launch CQE IV-16
The planned departure from requirements prior to the initial manufacture of an item. The departure will be
from a particular design requirement, a specific number of units or specific period or time and will be CQE V-18
identified upon shipment.
Written authorization given in advance of manufacture to deviate from specified requirements for a given
CQE V-18
number of units or for a specific period of time
See Primer. Note a Yellow Face on a dial indicator means that readings are in metric (SI) CQE VI-3,21,22
See Distributions
Binomial, Hypergeometric, Poisson Are all discrete distributions. Data only occur's at integral values
below
CQE IX-18,
See range, variance, and standard deviation
X-10,12
Binomial (when sampling with replacement and proportion defective is constant and > 10%, large lot size
N>50 and small sample size relative to lot size n<1/10N) CQE IX-61,63 XI-44,45 XII-77 StatGuide 4-4
The outcome from a bernoulli trial has two possibilities, usually referred to as success or failure. The
probability of success or failure is independent on all Bernoulli trials. The binomial distribution is used to
model Bernoulli trials.
Bivariate Normal IX-49
Chi-Square IX-56,57 XI-22, XII-9 (Regenerative, sampling distribution - used for variance comparison)
Continous IX-46
Discrete (Frequencies occur only at integral values, non-continous) CQE IX-33, 61 StatGuide 4-1
Exponential IV-73,74 IX-46,50 (Mean time between occurances - reciprocal of the poisson distribution)
F IX-57,58 XII-10,11 (used to compare equality of variances from two normal populations)
Gaussian IX-46
Hypergeometric (Lot size small AND sampling without replacement) CQE IX-61,66 StatGuide 4-1
Lognormal IX-51,52
Multinormal IX-67 CQE
Normal IX-15,46,47 X-58,62 XI-41,42 XII-2 (see references to
Poisson IX-61,64,65 XI-43,44 XII-4,6 (Defects per unit, Occurances per interval, this Distribution is the left)
reciprocal of Exponential Distribution) This distribution is an approximation to the binomial when the
probabability of occurence is equal to or less than 0.1 and the sample size is large. The smaller the fraction
defective and the larger the sample size the better the approximation. Average or expected value for this
distribution is µ. This is also µ= np-bar=c-bar. ==sqrt(c-bar) (CQE IX-64)
Probability IX-46
Rayleigh IX-53
Rectangular IX-48
Selected IX-22,23
Students t IX-59,60 XI-13,17 XII-8 ( hypothesis testing and confidence interval for means when = is
unknown). As the sampe size increases error in standard deviation gets smaller.
Time between failures IV-80
Uniform IX-48 XI-40
Weibull IV-75 IX-27,28,46, 53:55
Design, measure, analyze, improve, control. Most often employed by improvent teams. CQE VIII-8
Attribute plans for effective sampling inspection when the process average is known. If average is not
CQE V-44,53:56
known, see MIL-STD-105E. Tables are designed to minimize ATI.
To increase the probability of better quality Drawings and Specifications should be realistically set with rigid CQE IV-7/10
enforcement. CQE IV-7/10 and V-2/7. CQE V-2/7
Reports that deal with business financial status and financial ratios.
Time-Sequence data (opposite of Static Data). Examples include control charts, run charts, or trend charts. CQE X
Event being observed. Usually used in probability calculations. Example: P(E)=n e/N means 'the probability
that the Event will occur equals the number of observations of the event, divided by the Number of times the CQE IX-26
experiment is repeated. Also can be the expected value from the event.
equal to the square root of (loss when exceeding functional limits) / (loss when exceeding tolerance specs.) CQE IV-26
Leads to increased knowledge. Not to be confused with training, which leads to measurable, value-added
skills.
A study in which action will be taken on the universe (per Deming). Ex. Census information, vs an analytic
study in which action will be taken on a process to improve performance in the future.
CQE IX-35
Enumerative data is data that can be counted (classifying people by age, income, etc.). May be analyzed
using chie square, binomial, or Poisson distributions.
An item that is interchangeable and equal to or better than the item called for in a specification. CQE V-18
Designing a process/product that can't be made wrong. Evaluate where mistakes can happen. Install poke-
yokes (hard physical or electronic error-proofing). Sometimes (although not ideally) may include at-the- CQE VIII-13,44
operation inspection.
Steady improvement. Good when no competition present, but see 'Revolutionary Improvement' and
CQE VIII-5
'Breakthrough Achievement (VIII-5).'
Evolutionary Operation. Sequential experimental designs. See Primer definition. CQE XI-77,93
CQE X-26 and X-28
See Exponential Weighted Moving Average CSSBB
IX-51-60
The mean, µ of a probability distribution is the expected value, E(x), of it random variable CQE IX-33
To say that a model in an experimental design is fixed, indicates the levels used for each factor are the only
ones of interest
3 types - end users, intermediate, other impacted parties. Current customers are worth as much as 5 times
CQE II-96
new customers.
For a given number of degrees of freedom, as the variablity of means increases reative to the variability
StatGuide 8-4
within groups the F-ratio increases as well.
Brainstorming, Nominal Group Technique (NGT), Multivoting, Force Field Analysis CQE II-81:84
The number of malfunctions occuring per unit of time - can vary as a function of age.
Useful Life Period (constant failure rate, this is the exponential distribution ) StatGuide
3-7 / 3-8
Wearout Period - replacement / repair time
Company strategy. See Leading Edge for strategy & product types list.
This is a top down approach used to identify faults and their causes. See CQE IV-81 for info on FMECA,
CQE IV-86
Best suited to modeling reliability of complex equiment! CQE-IV-86/86
Developed fourteen principles of management, which included issues such as how power and authority
needed to be subdivided into reporting relationships, and the subordination of individual interests to the
common good.
Defined quality as "The total composite product and service characteristics of marketing, engineering,
manufacturing and maintenance through which the product and service in use will meet the expectations of CQE II-4;
the customer." Well known for Quality Systems and Total Quality Control (SPC, or Statistical Process II-11-12, VIII-29
Control. See Primer page II-36 for his definitionof TQC). Several famous quality quotations.
CQA
Audit term that encompasses questioning/interviewing as well as observing processes
V-21
See Primer. See also hardness testing method summary matrix in Primer, p. VI-59. CQE VI-59, 62
Also known as Ishikawa, 4-M, or Cause-and-Effect Diagram. Breaks problems into bite-sized pieces. Keeps
asking 'why' until root cause is obtained. Using Manpower, method, machine, material as typical starting
points. Laboratory use may need measure resulting in a 5-M, environment (E) may enter into this as well
(Mother nauture - ie. 5-M and E schematic). There are 3 parts in a fishbone session Brainstorming, CQE VII-4,5
prioritizing and development of an action plan. First known use was by Tomiko Hashimoto.
See also Memory Jogger II, p. 23-30. People, Policy, Procedures, and Plan are the 4-Ps often used for
transactionals.E268
See Michael Porter
Only interested in those operators (machines) that are chosen for the experiment. Test all mean squares StatGuide
against error mean squares. 8-7
Taking a fixed sample size from a lot or batch only works if the lot or batch size remains relatively constant.
This will be illustrated later with a set ofOC curves from 10% sample size plans. The only advantage is that a
CQE V-28
fixed sample size in inspection is easy to remember. Fixed, small sample sizes are more widely used in
auditing.
The flow chart can depict the sequence of product, containers, paperwork, operator actions or administrative
proceedures. A flow chart is often the starting point for process improvement.
Questions to ask while reviewing the flow chart:
Why do we do it this way?
Is there unecessary complexity?
CQE VII-6-8
Does duplication or redundancy exist?
Are there control points to prevent errors or rejects? Should there be?
Is this process being run the way it should?
Can this process be performed differently?
See Process Flow Chart
Some individuals tend to round a reading to just inside the specification May be know at either the conscouis
CQE V-61
or subconcouis level to the inspectior.
Failure Modes and Effects Analysis - Bottom up approach. Does not use life-cycle costs as a component of CQE IV-69
this assessment. See CQE IV-86 for info on FTA or Fault Tree Analysis CQE IV-81/90
Added to Max Weber's work on bureaucracy and lines of authority by adding the factor of relationships, both
within the organization and between the organization and its environment, as a consideration for
management.
Desire to understand the forces acting on a goal. Determine driving forces, determine opposing forces then
CQE II-84
add to the driving to overwhelm the restraining or remove or weaken restraining forces. Possibly do both.
Non-destructive test. See Primer. See also force measurement terms list on page VI-83-86 CQE VI-79
Stage 1 of team development (forming, storming, norming, & performing). Proud to be selected, anxious,
CQE II-74, 74
new team experience, forming team structure.
10% Gage R&R or less is excellent, 10% to 20% might be okay, 30% or more --> Timeout! (CQE V-10)
Gage R&R is a stacked ANOVA.
(10 parts, label from 1 to 10, 3 operators, 3 times 10 meas, each. We know which part is which, operator CQE VI-77-80
does not)
See Primer
Project Management tool. Chart like those used in MS Project project management software. See Memory
Jogger (p. 9). Named for Henry Gantt. Displays activities or events as a function of time (or sometimes CQE II-49, 50
cost). See Primer for list of advantages and disadvantages.
Developed the Gantt chart for project management. Proposed the concept of providing incentives to
CQE II-49, 50
employees based on output.
See Normal Distribution. See book for formulas, graph, and applications. CQE IX-46
Tolerancing, statistics associated with tolerancing, allowances and fits, material conditions, characteristics
and symbols, datum planes, blueprint definitions, orthographic projections, information included on all
blueprints.
Tolerances noted on the body of a blueprint override tolerances noted in the title block.
Check for Bonus Tolerancing in the event of a maximum or least material condition. CQE IV-54
CQE IX-49-60
A circle that contains the follow letters indicate these items: CQE IV-50, 60
M -Maximum Material Size
L-Least Material Condition
P-Projected Tolerance Zone
S-Regardless of Feature size
Japanese for "in the area" or "on the shop floor." Implies that you can't fix anything from the computer or the
CQE VIII-23
desk. You must go to the people and the process and observe/interact with it.
Focused on increasing productivity, and developed the motion study method in order to analyze work
processes.
See Reengineering. See Primer for published book (with James Champy) and article (1990).
Reengineering is the fundamental rethinking and raical redesign of buisness process to achieve dramatic CQE VIII-32
improvements in critical contemporary measures of performance, such as cost, quality, service and speed.
See Primer. See also hardness testing method summary matrix in Primer. CQE VI-59 CQE VI-59-62
Father of Six Sigma, and author of several books, including 'Six Sigma.' He promotes Six Sigma
methodology as useful for companies with a focus on cost savings, customer satisfaction, internal problems,
and even design improvements.
Hazard Analysis and Critical Controls Plan. See FMEA IV-69, 81, 85
The positive change in behavior due to being singled out for attention. First noted at the Hawthorne Plant of
Western Electric Company in the 1930s. Studies to determine the best level of lighting, length of workday,
and length of rest periods showed that performance was related more to team involvement than to the
factors that the teams were working to improve.
Stressed that in communicating, the words or symbols are not the actual thing being discussed. Clarity of
understanding requires that the receiver "gets the message" that was intended to be conveyed, rather than
receiving just the words/symbols/etc.
h(t) = lambda is equal to 1/MTBF Note: The ratio of the probability density function to the reliability
CQE IV-73
function, f(x)/R(x), is equal to the hazard function, which is the instantaneous failure rate
Two categories into which work motivation falss: satisfiers and dissatisfiers. Dissatisfiers included things
such as work conditions, salary, company policies, and relationships w/ supervisor(s). Satisfiers include
responsibility, achievement, advancement, and recognition. Dissatisfiers must be taken for granted
(minimum requirements). Satisfiers then add levels of motivation.
See also Memory Jogger. For list of rules for number of classes of data, see Memory Jogger.
"Full" Quality Function Deployment. Developed by Ishikawa for the Kobe Shipyard to pass along the
characteristics of Quality to the product development team. See Axiomatic Design Theory. Hauser suggests CQE II-100-102, IV-
the following sequence to capture customer needs in design, 1) engineering characteristics, 2) parts 17-19
characteristics, 3) key process operations, 4) production requirements.
CQE II-31
Execution tool used to deploy an existing strategic plan throughout the organization. Focus is to identify
the vital few breakthrough achievements. Divided into two parts: 'ho' means 'method' and 'shin' means
'shiny metal showing direction.' 'Kanri' means 'planning.'
See book for formulas, graph, and applications CQE IX-61, and 66
Permits the study of treatments with more than three blocking variables. A 4x4 Hyper-Graeco Latin Design
(Car vs. Carburetor type) with 4 levels & 6 factors (Driver, Tires, Days, Speeds, Car, Carburetor) would have CQE XI-96/100
46 tests if done as a full factorial experiment. (4096 tests!)
What test should I run? See variable data and attribute data.
For a t - test: Assume two machines 10 samples each. DF=n1 + n2 -2.=10+10-2=18. If H0: u1=u2 and
H1:u1<u2, then t from table = 1.734 at alpha of .05. If calculated t value is > 1.734 H0 is rejected. CQE XI- CQE XI-7, 31
32 and XII-8
Deals with probability. The intersection of two events (AB) is composed of all sample points that are in both
CQE IX-38
A and B. The same as P(A) x P(B).
ANSI/ASQ standards for everything from QM/QA, to systems guidelines, statistics, etc. See Primer for
CQE III-30-31
extensive list. International Electrotechnical Commission.
Sporadic in nature and difficult to avoid. Rigidly enforced procedures, automated inspection or error-proofing
CQE V-62
may help.
When P(A I B) equals P(A). Ex. 50% of the time the coin lands on heads. 16.7% of the time the die lands on
the number 6. 16.7% of the time die lands on the numbe 6 when the coin lands on heads. P(A I B) = 16.7% CQE IX-40
and P(A) = 16.7%. The die roll is independent of the coin flip (see Dependent events.
Often management receives organizational information that is distorted. This may be minimized by
CQE II-88/89
practicing management by walking around, establishing an open door policy, and not killing the messenger!
See Primer. See also gage summary matrix CQE VI-3,4,5 CQE VI-23
Quality level somewhere between the AQL and RQL. It is normally defined as the quality level having
probability of acceptance of 0.50. IQL is rarely used, so they'll probably ask one question on the exam just to CQE V-33
annoy you.
General Inspection levels are Level I, Level II, and Level III. Unless specified, Level II should be used. Level
CQE V-48
I may be specified when less discrimination is required, Level III when greater discrimination is required.
Inspection, under a sampling plan, which is used when there is no evidence that the quality of the product
CQE V-39
being submitted is better or poorer than the specified quality level. This is the usual inspection starting point
Logical reasons for providing source inspection are key inspection points are hidden after assembly at the
suppliers plant or the criticality of the product warrant's it. Simply having shown consistently poor quality CQE V-24/26
would be a reason to select another supplier.
Normal --> Tightened: When 2 out of 5 consecutive lots or batches have been rejected on original inspection
Tightened --> Normal: When 5 consecutive lots or batches have been considered acceptable on original
inspection.
Only 80-90% of total defects will be detected via normal methods and conditions CQE V-18,20,62
p.287
See Primer. See also gage summary matrix CQE VI-3,4,5 CQE VI-29
Similar to Affinity Diagram. Technique created for the more complex problems or issues that managment
CQE VII-32, 33
may face, where direct, exact relationships may be difficult to determine.
Deals with probability. The intersection of two events (A B) is composed of all sample points that are in
CQE IX-38
both A and B. The same as P(A) x P(B).
Data is arranged in order and differences can be found. However, there is no inherent starting point and
ratios are meaningless (ex. Temperatures of steel rods (degs. F) after one hour of cooling)
CQE IX-7, 8
Central Location: Arithmetic Mean. Dispersion: Standard or Average Deviation. Significance Test: t-test,
F-test, or correlation analysis.
Founder of company-wide Quality Control (CWQC. See Primer p. II-36 for definition). Father of Japanese
Quality Control and Total Quality Control (TQC- first coined by Feigenbaum). In the 1950s, instigated "next
CQE II-4 , 13, 14
operation as customer," "training of workers," "empowerment," "customer satisfaction," "elimination of
sectionalism," "management of workers," and other now-vogue ideas.
Also known as Fishbone, 4-M, or Cause-and-Effect Diagram. Breaks problems into bite-sized pieces.
Keeps asking 'why' until root cause is obtained. Using Manpower, method, machine, material as typical
starting points. Laboratory use may need measure resulting in a 5-M, environment (E) may enter into this as
well (Mother nauture - ie. 5-M and E schematic). There are 3 parts in a fishbone session Brainstorming,
CQE VII-4, 5
prioritizing and development of an action plan. First known use was by Tomiko Hashimoto.
See also Memory Jogger II, p. 23-30. People, Policy, Procedures, and Plan are the 4-Ps often used for
transactionals.
See 'Fishbone diagram' or Cause-and-Effect Diagram.
Quality Assurance requirements for measuring equipment and metrological confirmation system for
CQE VI-
measuring equipment. See Primer for details.
CQE III-18
See table on III-18
CQE III-18
See definitions in Primer or Handbook.
ANSI/ASQ standards for everything from QM/QA, to systems guidelines, statistics, etc. See Primer for
CQE III-30-31
extensive list.
The recognition that sequential experimentation will often yield more satisfactory results than 'one big
CQE XI-88
experiment.'
This is defined as a device that stops a machine whenever a defective product is produced. It is
autonomation - a form of automation with human elements attached. When an equipment malfunction CQE VIII-23
occurs a light turns red or a signal comes on to indicate a problem.
Developer of the Affinity Diagram (also referred to as the KJ Method, in his honor).
Producing and delivering the right items at the right time and in the right amounts. CQE II-110, VIII-13
Defined quality as "Fitness for use." Developed the 80/20 Pareto concept. Advocate of the "revolutionary
rate of improvement" concept. Preached top-management involvement (like Crosby). The 'Juran Trilogy' CQE II-4, 15, 17
includes Quality Planning, Quality Control, and Quality Improvement.
From Japanese 'kai' meaning 'change' and 'zen' meaning good.' Connotes 'gradual and unending
improvement.' Originates from Masaaki Imai (yes, that's his name). Applies to doijng little things better and CQE VIII-11
setting and achieving increasingly higher standards. Focus is on small, gradual evolutionary improvements.
tool/technique system of signals and control (1-for-1 replacement) (Provides material control for the factory
CQE VIII-23,24
floor. Cards control the flow of production and inventory)
Also called the Kano Model. Used to analyze Customer requirements, often as part of the Define Phase.
Requirements include basic, variable, delighter (latent), and satisfiers.
Model for customer satisfaction, including satisfaction/dissatisfaction vs service dysfunctional/ service fully
functional. Evaluates what must be present vs. exciters/delighters. One dimensional factors drive
satisfaction in direct relationship to their presence.
See also Norton and Balanced Scorecard. Advocated a business planning process that gives consideration
to factors other than financial ones, including 1) Financial, 2) Internal Business Process, 3) Learning and
Growth, and 4) Customer Perspectives. It provides more perspectives for stakeholder interests, and is
referred to as the balanced scorecard.
Non-parametric test which makes a ranking evaluation by comparison with a critical value of Chi-Square.
This definition is true of Mood's Median Test, The Kendall Coefficient of Concordance, and the Kruskal -
Wallis test.
Same as 8-D (8 disciplines) problem-solving process. Technique that involves 1) Name the team, 2)
identify/quantify the problem, 3) implement containment, 4) identify possible causes, 5) develop and verify
possible corrective actions, 6) implement corrective actions, 7) leverage corrective actions/learning points
across all similar processes, 8) congratulate team.
Named after Dr. Kawakita Jiro, founder of Kawayoshida Research Center. See Affinity Diagram. CQE VII-24
See Primer. See also hardness testing method summary matrix in Primer. CQE VI-59 CQE VI-59, 61
The set of active processes that supports organizational learning. Includes linkage of people to data (user
interface or 'processing layer'), data storage devices ('data depository'), and data layer (link across different
media, such as audio, video, text, and database).
Recognized authority on organizational change. Surprisingly, recommends all five of the following responses
to overcome resistance to change (in order), 1) educate and communicate the necessary change, 2) enlist
employee participation in the project, 3) provide training or counseling support, 4) use manipulation to obtain
support, and 5) use threats or direct force (really!).
Key process input variables / Key process output variables CQE X-7
Non-parametric test which makes a ranking evaluation by comparison with a critical value of Chi-Square.
This definition is true of Mood's Median Test, The Kendall Coefficient of Concordance, and the Kruskal -
Wallis test.
Avoid words with multiple meanings, maintain consistency of terminology, stick to a logical sequence of
events, do not use complex compound sentences, do use simple direct sentences, avoid most abbreviations,
CQE II-93
acronyms, and contractions, avoid puns, slang and idiomatic expressions, be careful with contract language,
avoid latin abbreviations (etc., e.g., ie.)
See Primer. See also gage summary matrix CQE VI-3,4,5 CQE VI-28
Used with Design of experiments. Number of levels must be the same for each factor.
Latin Square design = Can also analyze position effect. NOT to be used when interaction effects exist. StatGuid 10-6
Graeco - Latin Square = Another factor at 4 levels can be embedded into the experiment.
Company strategy used by companies that want to be first to market. Strategy - Product mixes include:
Leading Edge - First to market; Fast follower - high quality, good price; Commodity Producer - High volume,
Competetive; Trailing Edge - Customer needs the product.
Only produced what is needed, when it is needed. Relates to reducing cycle time and improving agility /
flexibility. Lean thinking includes value stream mapping, spaghetti diagrams, visual control, work balance CQE VIII-12 to 14
analysis, and monument identification.
In summary, it is an entire enterprise involved in improvement. As Womack puts it, "make value flow."
Level 1: Quality Manual. Level 2: Procedures. Level 3: Work Instructions. Level 4: Records.
Ranking scale commonly used for Customer satisfaction (as in, on a scale from 1 to 5, how satisfied are you
with...).
The difference in bias (offset) values throughout the expected operating ranges of a gage or measurement
CQE VI-77
system
Best to identify things such as paint blemishes. Typically recorded on a check sheet called a measles
CQE IX-3
chart. CQE VII-10
See also Taguchi Loss Function. L(y) = k(y-m)2 where k = cost of a defective product divided by the square
of the full tolerance (ex. +/- 3 tolerance = 6 2 in denominator). y = desired target. m = actual units. L(y) = CQE III-65
Loss to society.
This percentage is estimated by dividing the number of defectives by the sample size and then mulitplying by
CQE V-40
100 (ex. D / N x 100)
A collection of units of similar product from which a sample is drawn and inspected CQE V-40
Lot Tolerance Percent Defective. From the Dodge-Romig tables. See Rejectable Quality Level CQE V-53, 54
Lot Tolerance Percent Defective. From the Dodge-Romig tables. See Rejectable Quality Level CQE V-53, 54
Best description of machine capability is "The inherent variation of the machine". The variation in a short run
CQE X-55
of consecutively produced parts is close but not as good a descriptor as inherent variation.
The probability that a device will be restored to operational effectiveness within a given period of time when
the maintenance action is performed in accordance with the prescribed proceedures.
CQE IV-64, 69, 77
See Primer definition.
A serious defect, but less severe than a critical defect, that is likely to result in failure or substantially reduce
the usability of an item for its intended purpose
CQE V-18, 20
This is a defect other than critical that may cause the product to fail, cause poor performance, shortened life,
or prevent interchangeability. Ex. A defect in the motor that stops the car along the highway.
Named after former Secretary of Commerce. Instigated through Public Law 100-107 in 1987. Purpose is to
promote TQM (Total Quality Management). Emphasis is on business results. Categories include 1)
Leadership, 2) Strategic Planning, 3) Customer and Market Focus, 4) Information and Analysis, 5) Human CQE III-31, 34
Resource Focus, 6) Process Managmenet, and 7) Business Results. See Primer for details or handbook for
details.
Two-axis perspective of managerial attention, including task-focus on the X-axis and people-focus on the Y-
axis. X-only focus shows authority focus. Y-only focus shows "country club" mentality. Strong focus on both
is best. Focusing on neither means the manager is pretty much useless.
Often management receives organizational information that is distorted. This may be minimized by
CQE II-88/89
practicing management by walking around, establishing an open door policy, and not killing the messenger!
Ensure a consistency of purpose, reinforce positive results, sharing business results, giving people a sense
CQE II-63
of mission, developing a realistic and integrated plan, provide direction and support
See Segmentation
Guy credited with inventing kaizen process. See kaizen. CQE VIII-11
5 stages of needs. 1) Physiological (eat, sleep, shelter), 2) Safety (economic & physical security), 3)
Belonging (acceptance by family & friends), 4) Esteem (held in high regard/status, 5) Self-actualization (to CQE II-96
achieve one's best)
ISO 9001 :200011 and ISOITS 16949:200212 both require material identification by a suitable means from
receipt through all stages of production, delivery, and installation. In order for the next operation to be
CQE V-12
successful, there must be assurances that the right inputs are used. This is accomplished by matching the
material identification with the requirements on the process control documents
A formal multidisciplinary panel of peers established to perform material review. This panel reviews, valuates,
and either fixes, or disposes of, specific nonconforming materials or services. Important supportive
responsibilities include the initiation and achievement of corrective action which precludes recurrence. See
detailed responsibilites on CQE V-21
Cross-functional team, often including internal and Customer experts to review the appropriateness of use CQE V-18, 21, 23
for non-conforming parts.
One item which is often neglected by Material Review boards is the initiation of corrective action.
However the MRB itself is not responsible for developing the corrective action plan.
Principal purpose is disposition of non-conforming material
• MRR number
• Discrepant part name and number
• The contract, project or work order identification
• The department where the discrepancy was detected
• The PO number (customer supplied product)
CQE V-22
• The quantity of units rejected
• The location of the held items
• Any pertinent part drawing information
• A description of the nonconformance
• An explanation of the probable cause
When nonconforming items or lots exist, these products should immediately be identified and segregated
from good material. Often the nonconforming material is identified with a "hold" status. All nonconforming
items should be subjected for review by the designated parties (MRB) to determine if the product can be CQE V-14
used "as is", repaired, reworked, or scrapped. Disposition of nonconforming material should
be made as soon as practicable
• Material identification
• Which process(es) the material has completed CQE V-12
• Which testes) or inspection(s) the material has completed
• Where the material is going next
• Additionally, an auditor would also determine if it is in the right place
Used to show the relationship between objectives and methods, results and causes, tasks and people, etc.
The objective is to determine the strength of relationship between a grid of rows and columns. The
intersection of the grid will clarify the problem strength. May also be referenced as Matrix Diagrams.
Several basic types of matricies:
L-type: Elements on y-axis and elements on x-axis
CQE VII-30, 31
T-type: 2 sets of elements on the y-axis, split by a set of elements on the x-axis
X-type: 2 sets of elements on both y-axis and x-axis
Y-type: 2 L-type matricies joined at the y-axis to produce a matrix design in 3 planes
C-type: 3D Matrix, 2 L-type matricies joined at the y-axis, but with only 1 set of relationships indicated in 3
dimensional space. (needs a comupter software package to effectively generate.)
Used to show the relationship between objectives and methods, results and causes, tasks and people, etc.
The objective is to determine the strength of relationship between a grid of rows and columns. The
intersection of the grid will clarify the problem strength. May also be referenced as Matrix Diagrams.
Several basic types of matricies:
L-type: Elements on y-axis and elements on x-axis
CQE VII-30, 31
T-type: 2 sets of elements on the y-axis, split by a set of elements on the x-axis
X-type: 2 sets of elements on both y-axis and x-axis
Y-type: 2 L-type matricies joined at the y-axis to produce a matrix design in 3 planes
C-type: 3D Matrix, 2 L-type matricies joined at the y-axis, but with only 1 set of relationships indicated in 3
dimensional space. (needs a comupter software package to effectively generate.)
An attempt to create a hybrid product/project approach. It shares functional specialists, and can lower costs,
but can also create competing priorities.
Myers-Briggs Type Indicator. Test that breaks out personality into four letters: (E)xtrovert vs. (I)ntrover,
(S)ensing vs. I(N)tuitive, (T)hinking vs. (F)eeling, and (J)udging vs. (P)erceiving. Useful for team building as
well as for ensuring that all personalities and perspectives are equally represented. Answers the questions
1) Where do you prefer to focus your attention, 2) How do you acquire information, 3) How do you make
decisions, and 4) How do you orient toward the outer world?
Formulated the theory X vs. theory Y model. Theory X takes a negative view of human nature, assuming
that most employees do not want to work and will require continual prodding. Theory Y takes a positive view,
believing that employees want to work, and will seek responsibilities that can offer solutions to organizational
problems and their own personal growth. Not directly related to Theory Z (see William Ouchi).
x-bar. The sum total of all data values divided by the number of data points. Most useful when the data is
CQE IX-13
fairly symetric and shows just the one trend.
Used to collect locational data. Also known as a Defect Location Checksheet. A picture of the inspected
item, on which an inspector marks the location of defects using a simple check, X, or dot. (or perhaps a CQE VII-10
drawing of a human body where a mark indicates an injury location)
Error of a measuring instrument is the indication of a measuring instrument minues the true value. See
CQE VI-39, 75, 76
Primer for a list of sources of measurement error (VI-88).
Metrics of quality performance must be meaningful for proper goal attainment CQE II-32
Middle value when the data is arranged in ascending or descending order. For an even set of data, the
median is he average of the two middle values. Most useful when the data is skewed due to outliers.
CQE IX-14 and 15
Outliers 'pull' the mean in the direction of the outliers. (Note: median position is n+1/2. If this falls halfway
between two numbers take their average)
See Primer. See also gage summary matrix. CQE VI-4, VI-12-14
Technique for creating a visual representation (use of symbols) of a multitude of issues or concerns by
forming a map of the interrelated ideas, to get a more complete and common understanding of a situation.
A defect that is not likely to reduce materially the usability of the unit. Ex. A scratch on the side of a car. CQE V-18
Department Mission Statements should concisely state how the strategic quality goals and need of the
organization will be implemented. Should include specific quantitative goals. The Company Mission
CQE II-24
Statement should address how the company will realize its vision and strategic goals. Should also include
concise statements of objectives to be achieved. See also Vision Statement.
Special case DOE designed around mixed ingredients. The major design assumptions are similar to those in
a factorial design (assumptions of independence, normalcy, & homoscedasticity). Two ingredients constitute
the entire sample space. Mixture components, as a proportion, must sum to unity. The measured response CQE XI-79
is dependent upon ingredient proportions, and not on measured amounts. See also Simplex-Lattice Mixture
Designs
Most frequently occurring number in a data set. Most useful when used with categorical data or when the
CQE IX-14
data has very few modes.
With reference to experimental design, if we say that a model in an experimental design is fixed, this
indicates the levels used for each factor are the only ones of interest
See Primer. See also hardness testing method summary matrix in Primer, CQE VI-59 CQE VI-59,VI-61
Modeling Technique to predict the behavior of a system from the known random behaviors and interactions
CQE VII-31
of the system's component parts
Non-parametric test which makes a ranking evaluation by comparison with a critical value of Chi-Square.
This definition is true of Mood's Median Test, The Kendall Coefficient of Concordance, and the Kruskal -
Wallis test. Requires a chi-square calculation.
A formal multidisciplinary panel of peers established to perform material review. This panel reviews, valuates,
and either fixes, or disposes of, specific nonconforming materials or services. Important supportive
responsibilities include the initiation and achievement of corrective action which precludes recurrence. See
detailed responsibilites on CQE V-21
Cross-functional team, often including internal and Customer experts to review the appropriateness of use CQE V-18, V-21-23
for non-conforming parts.
One item which is often neglected by Material Review boards is the initiation of corrective action.
However the MRB itself is not responsible for developing the corrective action plan.
Principal purpose is disposition of non-conforming material
Measurement System Analysis (also, rarely, Manufacturing Systems Audit). See Gage Repeatability and
CQE
Reproducibility (GR&R). Measurement System Analysis includes an overall evaluation of
VI-74-75
consistency/accuracy of a measurement system. GR&R is a subset of MSA.
Mean Time Between Maintenance Actions (1/maintenance action rate) CQE IV-78
Japanese for 'waste.' Includes all non-value-added activity. Specifically, 1) overproduction, 2) inventory, 3)
CQE VIII-13, 19, 20
repair/reject, 4) motion 5) processing, 6) waiting, and 7) all forms of transportation.
This is a popular way to select the most popular or potentially most important items from a previously
CQE II-83
generated list.
If event A contains no sample points in common with event B, then they are said to be mutually exclusive. CQE IX-40, 41
See MBTI
A hired "customer" who is paid by the organization to test the system (eats in your restaurant and reports
whether the food was good/timely/server smiling & helpful, etc.).
When the 'tail' of a distribution points to the left (toward the 'negative' numbers on the number line). CQE IX-15
A moveable funnel is placed over a grid and a ball is dropped through the funnel creating a mark. CQE X-5
See 'Seven New Quality Management Tools.' Relations Diagrams, Affinity Diagrams, Systematic Diagrams,
CQE VII-23
Matrix Diagrams, Matrix Data Analysis, Process Decision Program Charts, and Arrow Diagrams.
Co-Author of 'The Six Sigma Way.' See Pande, P.S. for more information. CQE VIII-47
Developed at University of Wisconsin in 1968 by Andre Delbecq and Andrew Van deVen. Limits team
CQE II-82
interactions to reduce peer pressure. See Primer for details (not often used).
Factor which cannot be controlled by the designer and which can cause unknown variability, or an error in
CQE IV-21
the response factor.
An organization shall deal with nonconforming product in one or more of the following ways:
• Taking action to eliminate the defective nonconformity
• Authorizing its use, release, or acceptance, (including re-work or repair) under concession by a relevant
authority (or customer)
• Taking action to preclude its intended use CQE V-14:20
Non-Conforming Material Definitions - see page CQE V-18
Final disposition of Non-Conforming material can include: Completing missed operations, reworking, using
material "as is", and / or scrapping. CQE V-15
Tests that do not impair future use of a sample. Choosing most suitable method. Basic Techniques.
CQE VI-46-48
Comparison of techniques.
The non-disclosure agreement is a legal document often used between organizations to indicate htat
confidentiality will be maintained.
Receiving Inspection (On the other hand things like design reviews, vendor assessments, and inventory
CQE VIII-19/20
reductions can add value)
Bell curve. Distribution obtained when all special causes of variation are eliminated from a process. See CQE IX-15, IX-46, 47,
book for formulas, graph, and applications X-58-62, XI-41, 42
The usual starting point within each inspection level, unless otherwise directed by the responsible authority.
Reduced inspection - the plans allow for a smaller sample to be taken. Reduced inspection may be
implemented when it is evident that quality is runinng extremely well. Tightened inspection the plan requires
more stringent acceptance criteria. This is used when it is evident that quality is deteriorating. CQE V-48
Shifting from normal to reduced or tightened inspection depends on the quality levels observed within the
selected inspection level. See V-49 for switching rules.
Stage 3 of team development (forming, storming, norming, & performing). Team coalesces, cooperation,
CQE II-73, 74
willing to dialogue, conflicts reduced, focus on team objective.
See also Kaplan and Balanced Scorecard. Advocated a business planning process that gives consideration
to factors other than financial ones, including 1) Financial, 2) Internal Business Process, 3) Learning and
Growth, and 4) Customer Perspectives. It provides more perspectives for stakeholder interests, and is
referred to as the balanced scorecard.
The null hypothesis is a statement about the value of a population PARAMETER such as the mean, and
must contain the condition of equality. The null hypothesis is NEVER based on a statistic. The alternative
CQE XI-2,7
hypothesis is a statement that must be true if the null hypothesis is false. A null hypothesis can only be
rejected or fail to be rejected, it cannot be accepted because of a lack of evidence to reject it.
The first time a nonconformance is detected, attributed to the same cause and identified before a specified
CQE V-19
commitment for effective corrective action.
Chances favoring an event divided by the chances not favoring an event. Odds have no part in quality
CQE IX-6
engineering analysis or applications.
See validity
Quantifies Alpha and Beta risk. OC Curve is a graph of the percent defective in a batch versus the
probability that the sampling plan will accept that batch. The 'ideal' OC curve has zero alpha or beta risk . CQE V-32
(More protection for both the producer and the consumer).
See Primer. See also gage summary matrix CQE VI-3,4,5 CQE VI-4, 32, 33
See Primer. See also gage summary matrix CQE VI-3,4,5 CQE VI-4, 34
Data is arranged in some order but differences between values cannot be determined or are meaningless
(ex. Number of low, middle, and high income consumers using a product) CQE IX-7, 8
Central Location: Median. Dispersion: Percentages. Significance Test: Sign or Run Test.
Short or long term emphasis, profit, cycle times, market place response, resources CQE II-23
See 'Seven Original Quality Management Tools.' Cause-and effect diagrams, flow charts, checksheets,
histograms, control charts, paretor charts, and scatter diagrams (see problem solving chart in Primer).
These focus on quantititative and qualitative analysis. See Problem Solving Steps in Primer.
Fractional factorial designs that cannot take into account interactions. Developed by Taguchi. CQE XI-110
Shows an object from different views. There may be as few as one view, however there may be up to six
CQE IV-37/38
views possible. Top, Left, Front, Right, Rear, Bottom
Formulated the theory Z model, which refers to a Japanese style of management characterized by long-term
employement, slow promotions, considerable job rotation, consensus-style decision making, etc. Not directly
related to Theory X vs. Theory Y (see McGregor, Douglas).
P(A B) = P(A I B) x P(B) . Multiplicative law applies. If A and B are exclusive then P(A B) = 0 CQE IX-42
P(A ~ B) = P(A) + P(B) - P(A B) = Probability of A or B. Additive law applies. If A and B are exhaustive,
CQE IX-41
then P(A ~ B) = 1
Conditional probability of event A given that B has occurred. P(A I B) = P(A B) / P(B). Note that if P(A I B) =
P (A) then the two events are independent (ex. Die roll and coin flip). If not equal, the two events are CQE IX-39
dependent (ex. Rain and cloudiness).
Probability that an event will be observed. P(E) =n e/N (that is, the probability that the Event will occur equals
CQE IX-37
the number of observations of the event, divided by the Number of times the experiment is repeated).
Attribute type chart. One of the better attribute charts. To effect a change and reduce the fraction defective CQE X-29
would likely require a basic change to design or production process. StatGuid 2-16
If all items are operating at the outset - then this is an active parallel system. If some are not operating until
needed this would be a passive parallel system. CQE IV-72
Formula for caclulating parallel system reliability is P (event) = 1-(UA x UB x UC)
A population value (as opposed to a Statistic, which is a sample value). CQE IX-33
Specialized forms of column graphs. Used to prioritize problems so that the major problems can be
identified. Named after Vilfredo Pareto (1848-1923). Graphic display of the principal of the 'vital few' and the
CQE VIII-17, 18, IX-
'trivial many' (also called the 80-20 rule). These charts display the significant few categories, help to
23
eliminate insignificant categories and focus attention in priority order. They do NOT demonstrate
relationships between variables.
3 Major Areas:
Control Marking
By the supplier, manufacturing department, or quality assurance as required by drawing or inspection
routing specifications. These stamps and symbols may be used to verify the performance and/or
acceptance of special inspections, fabrication operations, or tests. These stamps mayor may not be
given a serial number or identification to link the user to the symbol applied.
Plan-Do-Check-Act. See Shewhart Cycle. While it can be widely used by teams, PDCA can be used
effectively by individuals CQE VIII-3/12, and 32.
CQE VIII-3
Modifiying or redesigning a product would likely occur during the 'do' and 'act' stages. CQE VIII-18
A specialized matrix diagram that captures the perceptions of the customer. An alternative to QFD (uses a CSSBB
Likert scale, if anyone cares). III-40
Stage 4 of team development (forming, storming, norming, & performing). Team has matured, smooth
CQE II-74
cohesive unit, focus on process, feelings of satisfaction, achieving goals.
Goals and corresponding measures - Profit / cycle time / market place response / resources CQE II-36
An ordered arrangement of n distinct objects is called a permutation. The number of ways of ordering n
distinct objects taken r at at ime is designated by the symbol nPr or P(n,r) or P nr. Calculated as n!/(n-r)!.
CQE IX-43, 44
Permutations take order into account, therefore the items ABC, ACB, CAB, CBA, BAC, and BCA are all
counted as different.
Accronym for Program Evaluation and Review Technique. Event-oriented project management tool. Uses
Critical Path formula. Requires time estimates for each activity in the network. Activity Network Diagram
(AND) is a simpler version of the CPM, which in turn is simpler than the PERT method. AND plus time
equals CPM. Add pessimistic, expected, and optimistic times for each activity, and you get a PERT. Critical
Path is the longest amount of time from 1st to last task. CQE II-43-45
Slack time 'S' is available time for steps not on the critical path. (Latest data an event can occur or can be II-49/50
finished without extending the project S=Tlate - Tearly. For events on the critical path Tlate = Tearly and S=0). Sink
point is where more than one activity comes into an activity. Burst point is more than one path leaving an
activity. Crashing is throwing extra resources at an event to reduce time. The expected time between events
T =(To+4Tm+Tp)/6. [(optimistic time + (4 x most likely time) + pessimistic time) / 6]
See Primer. See also gage summary matrix CQE VI-3,4,5 CQE VI-4, 18
See Primer. See also gage summary matrix CQE VI-3,4,5 CQE VI-28
A point estimate of the population mean , is the sample mean, X-bar. The sample variance s 2 is the best
point estimate of the population variance =2. The sample standard deviation s, is the best point estimate of CQE XI-3
the population standard deviation =.
Same as a bimodal distribution, except with more than two humps. CQE VIII-24
Developed the Five Forces of Competetive Strategy
When the 'tail' of a distribution points to the right (toward the 'positive' numbers on the number line). CQE IX-15
Full review of a project things gone right and things gone wrong in order to learn and apply project
experiences to the future (no reinventing the wheel, and no falling into the same pits twice).
May be an assessement type audit for suppliers. May be less formal than other types of an audit. CQE III-38/39
The review and disposition of nonconforming material, by quality assurance supervisors, as it is initially
discovered at the applicable inspection area priorto referral to the material review board. Typical preliminary CQE V-19
review disposition types are rework, scrap, or referral to the MRB
This standard applies to the aerospace industry and incorporates the European standard of prEN 9000-1.
The use of this standard facilitates an organization's ability to adhere to industry specific safety and reliability
requirements.
With regards to sampling errors: An individual may yeild to the need for delivery performance CQE V-61
Define Phase. Include such items as conformance quality, color range, and average age of receivables.
See also Secondary or Consequential Metrics.
The mechanism by which we take statistics and begin to make inferences about a population.
Mathematically, probability is the ratio of the chances favoring anevent to the total number of chances both
for and against the event. Probability is always expressed as a ratio.
CQE IX-37
BTW: The ratio of the probability density function to the reliability function, f(x)/R(x), is equal to the
hazard function, which is the instantaneous failure rate
The probability that a lot will be accepted under a given sampling plan CQE V-40
Displays probabilites over a range of values. The area under the probability density function must equal 1.
CQE IX-29/30
The probability density function must be positive for all values of x, since negative probabilites don't exist.
ISO 4.14.1.1 states that 'When an internal or external nonconformance occurs, a supplier shall use a
disciplined problem solving methodology. In the case of an external nonconformance, the supplier shall
further respond in a method prescribed by the customer.' 7 steps include, 1) Identify problem, 2) List the
possible root causes, 3) Search most likely root causes, 4) ID potential solutions, 5) Select and implement a CQE VII-2, 3
solution, 6) follow up to evaluate the effect, 7) standardize (CQM Handbook p. 116-117). 1-4 = Plan, 4 = Do,
6 = Check, and 7 = Act. 1 = Recognize, Define, and Measure, 2 - 3 = Analyze, 4 - 5 = Improve, 6 - 7 =
Control & Validate.
Define Phase. Should detail the issue that the team wants to improve, as well as a baseline metric and the
CQE VIII-2
anticipated improvement in that metric. Measureable. Frequently includes a target timetable.
Level 1: Quality Manual. Level 2: Procedures. Level 3: Work Instructions. Level 4: Records.
Basic content of proceedures should include: Pupose of the document, Basis of the document and scope of
CQE III-12, V-4
the document
Process capability is a measure of the inherent uniformity of the process and the ability to direct the process
to a defined target. It is often necessary to compare the process variation with the specification tolerances to
judge the suitability of the process. CQE V-3
Cp, Cpk, Pp, and Ppk. Cp = (USL-LSL)/6s-hat (where s-hat = estimated standard deviation (R-bar/d 2)). Cpk CQE X-3, X-53
= the lesser of Cpupper and Cplower (Cpl = (X-bar-LSL)/3s-hat, and Cpu = (USL-X-bar) / (3s-hat). Pp and Ppk
use identical formulas to Cp and Cpk except with actual, calculated s instead of estimated s-hat. See
Memory Jogger II (p. 132-136).
Production operations, which directly affect quality, are identified and planned to ensure that they are carried
out under controlled conditions. Controlled conditions include the following:
Used to chart the course of events that will take us from a start point to our final complex goal. Various
events are charted and contingencies are planned for based on uncertainty of obtaining intermediate goals.
CQE VII-28, 29
Similar to contingency planning or risk management. May be developed by using a Gantt chart along with
the question 'what if' associated with each step. Primarily used for new, unknown processes.
Process industries such as chemicals, paints, drugs at issue. Quality control has some responsibility for
CQE V-2/7
process selection and process tolerance and process development issues process specifications.
CQE VII-6/8
See Process Flow Chart CQE VII-6/8
CQE VIII-13
Also called Type I risk, alpha risk, or level of significance. Only comes into play if we reject the null. May
cause me to change my process unnecessarily. The probability of rejecting a good lot. 1 - a = Producer's CQE V-32, V-40, 41,
Confidence. See also Consumer's Risk. Note: alpha may be calculated for a designed experiment as XI-8
follows: Full Factorial Designed experiment alpha = (2 k)1/4, where k = number of factors.
Determining product and process control methods is often called the development
of a quality plan.
The first necessity is to identify all the key internal and external customer requirements. One should
remember to include all of the critical product and process characteristics uncovered throughout the
complete design process.
CQE V-5
The second step is to identify the manufacturing process flow and the manufacturing support processes.
The third step is to identify the quality tools that a company will use to control the processes.
This is the long term variation in a product. Key objective is to reduce lot to lot variability. CQE X-9, 43
Define Phase. Designed to help the team focus on the correct problem, align the solution toward
CQE II-63
organizational goals, and help ensure Champion support.
Includes planning, scheduling, and controlling a series of activities and tasks with a specified objective,
CQE II-40-50
starting and ending dates, and resources.
The push concept is associated with management maximizing production output CQE VIII-13
Quick Change-Over. Requires 5-S, standardized work, separating internal and external processes (those
that can be done while the machine is still running vs. what must be done when machine is shut down). See
SMED (Single Minute Exchange of Dies). Move as much internal time as possible to external time. Then
minimze internal time.
"Full" Quality Function Deployment. Developed by Ishikawa for the Kobe Shipyard to pass along the
characteristics of Quality to the product development team. See Axiomatic Design Theory. Hauser suggests CQE II-100-102, IV-
the following sequence to capture customer needs in design, 1) engineering characteristics, 2) parts 17-19
characteristics, 3) key process operations, 4) production requirements.
Tool that is sometimes referred to as "the voice of the Customer" or as the "House of Quality." The process
to ensure that the customer's wants and needs are heard and translated into technical characteristics.
CQE II-100-102, IV-
Includes a cross-functional team of sales, marketing, design engineering, manufacturing engineering, and
17-19
operations. First applied in Kobe shipyards in 1972 by Yoyi Akao and Associates. See Primer for more
details. Its real strength is in bringing people together with the information they need.
Organized method of collecting, storing, analyzing and reporting information on quality to assist decision
CQE II-51:54
makers at all levels. The purpose of an effective QIS is to achieve timely corrective acion.
This standard applies to the automotive industry, and was developed by the "Big Three" automotive
manufacturers (Chrysler, Ford, and General Motors). Incorporates teh requirements of ISO 9001:1994 and
additional requirements for quality management system attributes of business planning, advanced product
quality planning (APQP), satisfaction, and continuous improvement. An international standard (ISO/TS
16949) for the auto industry has also been developed,and incorporates the QS-9000 requirements plus a
few additions (ex. the need for measuring employee satisfaction).
A process of demonstrating that a product or process is capable of fulfilling a stated specification CQE V-2
A characteristic of a process or product which defines the quality of the process or product CQE V-2
ASQ defines Quality as "the totality of features and characteristics of a product that bear on its ability to
satisfy a given need." See also definitions by Juran (Fitness for use), Deming, Crosby (Conformance to
requirements), and Feigenbaum (The total composite product and service characteristics of marketing,
engineering, manufacturing and maintenance through which the product and service in use will meet the
expectations of the customer). Per TS-16949, "the degree to which a set of inherent characteristics fulfils CQE III-2
requrirements")
A thorough review of the works of quality guru's would indicate one of the most effective ways to
create quality is through continous improvement.
Quality assurance is planned and systematic actions that provide confidence that a product or service will
CQE III-6
satisfy given requirements for quality. See page III-6 for typical activities
Scope may consist of criteria for the Demming prize, or Malcom Baldridge. May be conducted by internal or CQE II-32
external teams. Types of audits: System Audit, Process Audit, Product Audit. CQE III-36
Definitions of terms, see Primer p. V-2-3. List of Product and Service Characteristics. Product and Process
CQE IV-2, 3, V-2
Characteristics.
Quality control operations are those techniques and activities that monitor and control quality. See text for
CQE III-6
typical techniques and activities. CQE III-6
Emphasized by Feigenbuam (USA) and Ishikawa (Japan) in the early 1980s. CQE II-4,11,12
Team for guiding continuous quality improvement. Should be cross-functional, but may take place on any
level of the organization (although most commonly focused on upper management).CEO must serve on CQE II-29
this team.
Quality documentation heirarchy includes Level I through Level IV documents. See Primer for definitions. CQE III-14
Crafsmanship
Standardization of parts (1798 - Eli Whitney)
Definition of a System (1911 Fredrick Taylor)
Quality Control (1950s - 1960s)
CQE II-2,3
Quality Assurance (1970s - 1980s items like SPC and quality cost measurement)
Total Quality Management (1980s-1990s added key ingredient of management direction)
Continous Quality Improvement (1980s - 2000s)
Six Sigma (lean six sigma)
CQE II-100:102
See QFD. See also House of Quality.
CQE IV-17:19
Can be used to measure both products and processes and feed the data back for decision making CQE II-51/54
Organized method of collecting, storing, analyzing and reporting information on quality to assist decision
CQE II-51
makers at all levels. The purpose of an effective QIS is to achieve timely corrective acion.
Per TS-16949 and ISO-9000:2000, the eight quality management principles include: Customer focus,
Leadership, Involvement of people (empowerment, motivation, etc.), Process approach, System approach to
management, Continual improvement, Factual approach to decision making, and Mutually beneficial supplier
TS-16949
relationship.
To increase the probability of better quality Drawings and Specifications should be realistically set
with rigid enforcement. CQE IV-7/10 and V-2/7.
ANSI/ASQ standards for everything from QM/QA, to systems guidelines, statistics, etc. CQE III-19
See 'Original Quality Tools' and 'New Quality Tools.' CQE VII-23
Provides, in writing, a description of the organization and the policies of the qualtiy assurance system.
Policy Statement, General description of the policy implementation, Correlation of policy and implementation
to applicable standards.
ISO 9001:2000 in element 4.2.2 states "The organization must establish and maintain a quality manual that CQE III-12
includes the scope of the QMS, documented proceedures or references to them and a description of the
interactions of the QMS processes.
Crosby - contribution: Senior manager involvement, 4 absolutes of quality management, quality cost
measurements.
Demming - PDSA, top mgt involvement, concentration on system improvement, consistency of purpose
Juran - Top mgt involvement/quality trilogy (project improvement), quality cost measurement, pareto analysis
Shewart - Assignable cause vs. chance cause, control charts, PDCA, use of statistics for improvement
Taguci - Loss function concepts, signal to noise, experimental design methods, concept of design
robustness
CQE II-29:32
See Primer
CQE V-2
Top management's formally stated intentions and direction for the organization pertaining to quality. Per the
CQA Primer, Section III, the principal perpose of a Quality Policy Statement is to define the company's CQE II-25, III-5
interest in quality of products and services. Defined in ISO9001:2000 Element 5.3
Customer satisfaction is key, defects must be prevented, manufacturing assuems responsibility for quality
assurance, process must be controlled, everyone participates in quality, quality is designed into the product, CQE II-25
TQ is a group activity, respect for humanity, adopt a revolutionary rate of quality improvement
A system is defined as a series of actions, activities, elements, components, departments, or processes that
CQE III-4
work together for a definate purpose.
The Food and Drug Administration (FDA) updated the Good Manufacturing Practices requirements 21 CFR
Part 820 used to regulate manufacturers of medical devices to include the ISO 9001 standard requirements.
Two quality models used in healthcare include JCAHO and NCQA
Most company training programs should focus on real problems not on minutia or trifles CQE III-66/70
See Primer. Matrix explains relationships of doing right or wrong, the right or wrong way. CQE III-11
Used to calculate a mini-range to limit the damage that an outlier does to the range value. To calculate the
lower quartile, use n/4. If the value is an integer the lower quartile is halfway between this position and the
next one. If not an integer then round up to find the position. To calculate the upper quartile use 3n/4. If an
integer then upper quartile is positioned halfway between this position and the next. If not an integer round
up. (Note: median position is n+1/2. If this falls halfway between two numbers take their average)
Interested in estimating effect due to operators and machines. For interaction test mean square of StatGuide
interaction against error mean square. For operators or machines test against interaction mean square. 8-7
CQE IX-18
The difference between the smallest and largest values in a set of data. CQE X-12,37 CQE V-
61, VI-80
An extension of the interval level that includes an inherent zero starting point. Both differences and ratios
are meaningful
CQE IX-6,8
(ex. Product A costs $300 and product B costs $600 (B is 2x the cost of A)).
Central Location: Geometric or Harmonic Mean. Dispersion: Percent Variation. Significance Test: t-test, F-
test, Correlation Analysis.
The underlying subgrouping scheme for control charting consists of products produced as nearly as possible
at one time (products produced in a tight time sequence), alternatively include product representative of all
production over a given period of time. Subgroups should be be homogeneous NOT heterogeneous.
Considered a non-value added activity. The vendor should be shipping quality product. CQE VIII-19/20
Per ISO 9001:1994, Section 4.16, "Retention times of quality records shall be established and recorded." No
CQE III-49
specific durations are required.
Used to collect measured or counted data. See Checksheets, or Primer example. CQE VII-9,10
Inspection under a sampling plan using the same quality level as for normal inspection, but requiring a
smaller sample.
CQE V-41,48
Switching to reduced inspection is usually done when quality levels are observed to be getting better. See V-
49 for switching rules.
Used to obtain a predictive model and determine a relationship or correlation of variables. The least squares
method is used to create the model. In multiple regression 3 or more variables may be used. (If this is the
case then there is one dependant variable and two or more independant variables.) The Correlation
coeffecient's (r) may range from -1 to 1. The closer to -1 or 1 the better. Coefficent of determination ( r2) StatGuide 9-4
ranges from 0 to 1. Larger is better. As the scatter of points about a regression line becomes greater, the CQE XI-60/66,71
Coeficent of Determination becomes smaller. Standard error of estimate - smaller is better.
A defective or nonconforming item which is unsuitable for use as offered. A product or service which is not
CQE V-19
accepted because it fails to meet the requirement criteria
The number of defectives that (if found in a random sample) dictates rejection of the entire lot and returns it
CQE V-41
for screening (or 100 percent inspection)
This defines unsatisfactory quality. In the Dodge-Romig plans, the term 'lot tolerance percent defective
(LTPD) is used instead of RQL. The probability of accepting a RQL lot should be low. May also be known as CQE V-33
consumer's risk in some tables (standardized at 0.1)
Shigeru, 1988. See Interrelationship Digraph. CQE VII-32
A function to determine the probability of a product performing its intended functions for a specified time
interval under stated conditions.
Useful Life Period (constant failure rate, this is the exponential distribution )
t -t/m R(t)
CQE III-3, IV-64,69,71
------------------------------------------------
m/1000 -.001 .999
StatGuide
m/100 -.01 .99
3-7
m/10 -.1 .90
m -1.0 .37
2m -2.0 .135
10m -10.0 .0005
Same as repeatability, precision, spread. Measure of how closely grouped a sample is (ideal process
capability, or Cp). Not to be confused with reproducibility.
See MTBF
Note: The ratio of the probability density function to the reliability function, f(x)/R(x), is equal to the
hazard function, which is the instantaneous failure rate
For Series Systems see StatGuide 3-11 / For Parallel Systems see StatGuide 3-13
The subjection of nonconforming material to an approved process designed to reduce but not completely
eliminate the nonconformance. The purpose of repair is to bring nonconforming material into an acceptable CQE V-19
condition
The variation in measurements obtained with one measurement instrument, by the same operator,
measuring the same characteristic on the same part at or near the same time. The definition is nearly the CQE VI-77,79,80,82
same as precision
The 'reliability' of the gage system or similar gage systems to reproduce measurements. The reproducibility
of a single gage is customarily checked by comparing the results of different operators taken at different CQE VI-77,79,80,82
times. Gage reproducibility affects both accuracy and precision.
Resolution I = Experiment where tests are conducted, adj one factor at a time. Not statistically sound.
Resolution II = confounding of the main effects (Factors).
Resolution III = Confounding between Factors and two-factor interactions.
Resolution IV = Confounding between Factors and three-factor interactions, plus confounding between 2 f.i.
CQE XI-81
Resolution V = Confounding between Factors and 4 f.i., plus 2 f.i. with 3 f.i.
Resolution VI = Also called Resolution V+. This is at least a full-factorial experiment with no confounding.
Can also mean two blocks of 16 runs.
Resolution VII = Can refer to eight blocks of 8 runs.E217
DOE Terminology. Graph of system response plotted against one or more system factors. Employs
experimental design to discover the "shape" of the response surface and then uses geometric concepts to CQE XI-82, 85
take adavantage of the relationships discovered.
A lot that has previously been designated as unacceptable and that is submitted again for acceptance
CQE V-41
inspection after having been sorted or reprocessed.
Steep rate of improvement. Often necessary in highly competetive markets (technology). Better than
CQE VIII-5
'Evolutionary Improvement.' However, see 'Breakthrough Achievement.'
The action taken on non-conforming items so that it will fulfill the originally specified requirements. This CQE V-2
requires customer involvement. CQE V-12/23 CQE V-19
Handbook definitions include 1) The formal process by which risk factors are systematically identified,
assessed, and provided for, 2) A formal, systematic method of managing that concentrates in identifying and
controlling areas or events that have a potential of causing unwanted change, and 3) In the project context, CQE IV-82 to 85
is the art and science of identifying, analyzing, and responding to risk factors throughout the life of a project
and in the best interest of its objectives.
Insensetivity to variation. When statistics from a procedure are not affected by moderate deviations from
CQE IX-35, IX-36
theoretical expectations.
See Primer. See also hardness testing method summary matrix in Primer, p. VI-59 CQE VI-59, 60
Noise factors are not controllable by the designer. Signal (mean) may be desired to be large, small, or close
to some normal value. A smaller S/N ratio is desirable when shrinkage, wear, or deterioration concerns are
CQE IV-24 and 25
at issue. A high S/N ratio is desirable when life, strength, or fuel efficiency are concerns. An average S/N
ratio is desirable with clearness or weights.
A technique, of possible recurrent use, for repairing a nonconformance when it has been demonstrated that
the technique, when properly applied, will result in an adequate and cost effective method for dispositioning CQE V-19
the nonconformance
Discussion of factors that could influence the integrity of the sample or sample data. See CQE V-61,62 CQE V-61,62
Number of sample defectives divided by the sample size and then mulitplied by 100 (d/N x 100) CQE V-41
One or more units selected at random from a lot without regard for their quality CQE V-41
Single - Sampling inspection in which, after each unit is inspected, the decision is made to accept the lot or
reject it.
CQE V-38-42
Double - Sampling inspection in which the inspection of the first sample of size n1 leads to a decision to
accept a lot, not to accept it, or to take a second sample of size n2
Multiple - Sampling inspection in which, after each sample is inspected, the decision is made to accept a lot;
not to accept it, or to take another sample to reach the decision. There may be a prescribed maximum
number of samples (7), after which a decision to accept or not to accept the lot must be reached.
TC = A + Bnmax + Cn-bar where TC = Total Cost, A = Overhead costs, B = Cost/unit of sampling, n max = Max
sample size, C = Cost/unit of inspecting, and n-bar = Average sample size.
No Inspection: Total Cost = NpD where N = number of items/lots, p = % defective in lot, and D = Cost if a
defective passes.
100% Inspection: Total Cost = NC where N = number of items/lots, and C = Inspection cost/item.
Sampling: Total Cost = nC + (N-n)pDPa + (N-n)(1-Pa)C, where N = Number of items/lots, n = Number of
CQE V-31
items/sample, C = Inspection cost/item, p=% defective in lot, D=Cost if a defective passes, and
Pa=Probability of acceptance.
Sample Size Break-Even Point: D/C where D = Cost if a defective passes, and C = Inspection cost/item.
One way to lower sampling costs for attribute sampling once a quality level has been established
with relatively small risks is to reduce the number of samples. Such as going from single to double
sampling.
In sampling we never know whether the lot is good or bad. We only know what the sampels tell us. See the
Lot Quality Decision Matrix (alpha and beta risk matrix)
CQE V-41
CQE V-62
Sources of errors may be Rounding, Pencil Whipping, Pressure, Flinching, Inadvertent Errors, Technique
Errors (each of these is listed in this index or on page CQE V-62).
Divide the number of sample defectives by sample size and multiply by 100. d/n x 100. CQE V-41
For attribute data, consists of Sample Size (n), the Accept number (A c), and the Reject number (Re)
CQE V-42 to 44
Changing the lot size has the smallest impact on an attribute sampling plan. CQE V-35/36
See MIL-STD-105E (attribute sampling plan which includes single sampling, double sampling, and multiple
sampling (VI-6). Also, MIL-STD-414 (ANSI/ASQ Z1.9) sampling plans for variables. Dodge-Romig tables
are attribute plans for effective sampling inspection when the process average is known. Sequential CQE V-47, 51, 52
Sampling (VI-33). For a matrix of attribute and variable sampling plans, their sampling types, applications,
and key features, see Primer, p. VI-35.
Graphic display of many data points which represent the relationship between two different variables. See
CQE VII-21, 22
Correlation Coefficient. See also Memory Jogger (p. 145-149).
Nonconforming material that is not usable for its intended purpose or cannot be economically reworked or
CQE V-19
repaired.
Nominal size, Threads per inch, Thread Series, Thread Class Symbol, Double, RH, LH, MOD CQE IV-46
Statistically Designed Experiments. Most widely pushed by Taguchi. Assumes independence, normalcy, &
homoscedasticity. Methodology includes 1) Set objectives, 2) select process variables, 3) select an
experimental design, 4) execute the design, 5) verify the data, 6) analyze and interpret results, and 7)
present the results. It is vital to have the right team develop the experiment, or the following errors could be
made: 1) unwarranted assumptions made about the process, 2) undesireable combinations of factors CQE XI-74-115
selected, 3) violation of the known laws of physics, 4) too large or small design sizes, 5) inappropriate
confounding, 6) imprecise measurement, 7) unacceptable prediction error, or 8) undesireable run order.
Objectives may include 1) response surface determination, 2) screening objectives, 3) comparative
objectives, or 4) optimal mixture proportions (among others).
Define Phase. Same as Consequential Metrics. Include such items as Defects per Unit and Technical
Support.
See CQM Handbook for definitions and concepts, including no segmentation, complete segmentation,
various single segmentations, segmentation by multiple criteria, patterns, homogenous preferences, diffused
preferences, clustered preferences, etc..
Sequential sampling is the most discriminating of the acceptance sampling plans.It involves making one of
three decisions as each sample item is obtained: accept the lot, reject the lot, or continue sampling. One
continues sampling until the cumulative number of defectives crosses either the lot reject limit, or the lot
accept limit. When a limit is crossed, the lot size limit is reached and a new lot begins. Sequential sampling
requires the least average sample size. Consider the following
sampling plan: n = 50, c =2, P = 0.017, at the 0.95 confidence limit using ANSI/ASQ Z1.4·20032 CQE V-30
Sequential plans are often applied where sample economics are critical, and a
minimum sample size is required. A sequential plan is more complex and more
difficult to administer than other plans. Samples must be obtained one item at a
time, and operators require more training
Relations Diagrams, Affinity Diagrams, Systematic Diagrams, Matrix Diagrams, Matrix Data Analysis,
CQE VII-23
Process Decision Program Charts, and Arrow Diagrams.
Cause-and effect diagrams, flow charts, checksheets, histograms, control charts, pareto charts, and scatter
diagrams (see problem solving chart in Primer).
These focus on quantititative and qualitative analysis. See Problem Solving Steps in Primer.
Along with Moody, co-authored the book, "The Perfect Engine: How to win in the new economy by building to
order with fewer resources' (just rolls of the tongue, doesn't it?). See definition of Standard Work
Estimated s value (as opposed to calculated s). See also Memory Jogger II (p. 136). CQE XI-67
Plan-Do-Check-Act. Continuous improvement technique. Same as Deming Cycle, PDSA Cycle (Plan-Do- CQE II-19
Study-Act), or PDCA Cycle. Invented by Walter Shewhart, but popularized by W. Edwards Deming in Japan. CQE VIII-3
Best known for his development of Control Charts and the Plan-Do-Check-Act cycle. CQE II-18,19
See Primer. See also hardness testing method summary matrix in Primer, p. VI-81. CQE VI-59, 60
System International D'unites (VI-64). For a list of SI system units, see Primer, CQE VI-65 CQE VI-64,65
CQE V-32,40,41
Also called Alpha risk. See Producer's Risk. The probability of rejecting the null hypothesis when it is true.
CQE XI-9
Deals with probability. A simple event cannot be decomposed. Example: A coin must be either Heads or
CQE IX-37
Tails. I die rolled will be either 1, 2, 3, 4, 5, or 6. No compounding is occuring.
Designed experiments characterized by regression functions that are referred to as cononical polynomials.
Like any Mixture DOE, proportions must total 1.0. There is no requirement that the same replicate number of
CQE XI-83
tests results exist for all test conditions. The number of design points (Y) is calculated as (q + m -1)! / m! (q-
1)!
Machined steel bar that has two cylinders spaced at known dimensions on the bar. Considered one of the
most accurate methods of obtaining angular readings of a surface plane.
Accronym for Supplier, Input, Process, Output, Customer. A business model that helps everyone within the
company see the company from an overall process perspective by 1) providing a framework applicable to
CQE II-33
processes of all sizes, 2) displaying cross-functional activities in simple terms, and 3) helping to maintain a
big business picture.
Emphasizes the reduction of variation, consideration of internal processes, concentration on the bottom line,
utililization of advanced technical tools, use of a formalized problem solving approach (DMAIC), the
elimination of internal wastes, and the need for key management leadership and support.
CQE II-3,
Sometimes called TQM on Steriods due to the inclusion of statistical items such as DOE, DFSS and
VIII-7/12, XII-3
statistical analysis and certain lean manufacturing tools. An average black belt program will save about
$175,000.00. A ratio of one black belt to 100 employees can provide a 6% cost reduction per year. For
larger companies there is usually one master black belt for every 100 black belts.
Acronym for a tool Specific, Measurable, Achievable, Relevant, and Timely. Often used for problem solving,
goal setting, etc.
Concept of performing a change-over in a 'single digit' of time. That is, 9 minutes or less. 'single minute' is
CQE VIII-14,26
somewhat of a misnomer. Single Minute Exchange of Dies (see Primer for case study)
CQE VI-59,61
Logical reasons for providing source inspection are key inspection points are hidden after assembly at the
suppliers plant or the criticality of the product warrant's it. Simply having shown consistently poor quality CQE V-24/26
would be a reason to select another supplier.
See CQE X-9 for some of the potential sources of variability CQE X-9, 43
Special causes affect the process in a unpredictable ways & result in unstable outcomes. Demming felt that
6% of all problems are attributable to special causes. Only a portion of these are operator controllable. See CQE X-6
above "Common Causes" for additonal detail or CQE X-5,6
A customer identified product or process characteristic that is applicable to a component, material, module,
subsystem or system which is defined by engineering as being critical to the product and having unique
safety, quality, reliability, durability, and regulatory criteria.
CQE V-2
In the automotive industry, these characteristics must be addressed in the contol plan during manufacturing.
Automotive special characteristics are grouped into two categories. Those with safety and legal
considerations versus those with non-legal and non-safety considerations.
Referring to Table 5.20, four special inspection levels 8-1, 8-2, 8-3, and 8-4 are provided. They are used
where relatively small sample sizes are necessary and large sampling risks can or must be tolerated.
CQE V-48
In the designation of inspection levels 8-1 to 8-4, care must be exercised to avoid AQLs inconsistent with
these inspection levels.
Documented, detailed requirements with which a product or process must conform. CQE V-3
The drift or change in bias obtained with a measurement system on the same measurement characteristic
CQE VI-77,78
over an extended time
A common problem in the stage gate process is a failure to kill off enough projects. Too many marginal
CQE IV-13
projects tend to exist overloading the team.
Stakeholders in a company project include 1) stockholders and owners, 2) management and employees, 3)
suppliers, 4) customers, and 5) society. See ISO-9000 -1, Section 4-2 (Shareholders and their
expectations). CQE II-23-36
s for population, S for a sample. The square root of variance. Additive law of variences says that:
CQE IX-18-20
=2measured = =2True + =2Error and =2Error = =2Measured - =2Error. See CQE VI-75
According to Sharma (2001), standard work is 'the best combination of machines and people working
together to produce a product or service at a particular point in time.' Standard work is the documentation of
CQE VIII-14,25
each action required to complete a specified task. The elements that compose standard work are: Cycle
time, Work Sequence, Standard Inventory.
Summary data (not in time sequence). Opposite of Dynamic Data. CQE VIII-6
A Sample value (as opposed to a Parameter, which is a population value). CQE IX-33
Statistically Designed Experiments. Most widely pushed by Taguchi. Assumes independence, normalcy, &
homoscedasticity. Methodology includes 1) Set objectives, 2) select process variables, 3) select an
experimental design, 4) execute the design, 5) verify the data, 6) analyze and interpret results, and 7)
present the results. It is vital to have the right team develop the experiment, or the following errors could be
made: 1) unwarranted assumptions made about the process, 2) undesireable combinations of factors CQE XI-74-115
selected, 3) violation of the known laws of physics, 4) too large or small design sizes, 5) inappropriate
confounding, 6) imprecise measurement, 7) unacceptable prediction error, or 8) undesireable run order.
Objectives may include 1) response surface determination, 2) screening objectives, 3) comparative
objectives, or 4) optimal mixture proportions (among others).
Steps:
Precisely define the problem objective
Formulate a null hypothesis and an alternative hypothesis
Decide if the problem will be evaluated by a one-tail or two-tail test
CQE XI-2
Select a test distribution and a critical value for the test statistic
Calculate a test statistic value from the sample information
Make a inference by comparing the calculated and the critical values
Report the findings
Effectively a histogram made up of the actual values, rather than simple bars. Has the same benefits as a
histogram, but without losing the actual data. See CQE Primer for example. See also Memory Jogger II (p. CQE IX-25,26
74).
Random variation.
Stage 2 of team development (forming, storming, norming, & performing). Individualistic thinking, tug of
CQE II-73,74
external loyalties, fluctuating attitudes, confrontations, team's task sinks in
Stratified samples are sometimes more informative than homogeneous samples. When analyzing inventory,
one would not want to put $25,000 motors in the same strata with 2¢ screws. Additionally, one might be
CQE V-28
interested in determining the amount of pallet damage in a storage area. There might be a need to sample
more row ends, row corners or bottom pallets in preference to top pallets in the middle
Vision and statement of purpose, gather data, assess corp strength and weakness, make assumptions about
factors outside control, develop steps (strategic and tactical) for implementation, evaluate performance to CQE II-22
goals, re-evaluate prev steps for perpetual use
Strategic goals are long term. Tactical are short term. These include … Vision / misson statement /
CQE II-26
customer complaints / returns / customer surveys, etc…
1) Develop certification process (selection, evaluating, rating, communication, and reaction processes)
2) Identification of supplier categories for which certification would be advantageous.
3) Evaluation of selected suppliers against defined criteria
4) Reporting of evaluation results, awarding of certification for those who qualify, and clarification of future CQE II-109
interrelated processs (ex. supplier reporting requirements).
5) Ongoing monitoring of supplier performance, with actions taken in accordance with the defined
certification process.
See CQM Handbook list for quote requirements, product technical req., process req., product verification and
CQE II-104,105
traceability req., quality system req., delivery, order, corrective action requirements, etc.
May be measured using Past Performance Index (PPI), Supplier performance index (SPI), Commodity
Performance Index (CPI), Quality Performance Index (QPI), and/or Delivery Performance Index (DPI). See CQE II-105
also Performance Index (PI).
No Rating - The rationale is that purchasing and quality know which companies are good or bad.
Quality Rating only - A rating based on incoming inspection statistics
Quality and delivery rating - Graphic method. A quality rating charted against the delivery rating
Quality and delivery rating - Cost index method. Use of a rating sytem based on a fixed dollar penalty of
nonconformances
Comprehensive Method - The measuring and rating of aggreed on variables such as: quality, cost, service,
delivery, etc. CQE II-106
Production percent nonconformance: The ration of defective items received to the total number of items
received
Disposition of lots: A weighted analysis based on nonconforming lots
Economic analysis: A comparison of suppliers based on total dollars
Composite Plan: This includes categories other than quality. (schedule, cost or delivery)
Auditing or in-process surveillance which consists of monitoring the manufacturing process of the supplier
and can involve: witnessing key events, such as operations, inspections, tests; Critical characteristic CQE II-108
inspection by withnessing or performing; joint troubleshooting of mutual quality related issues.
This is of greater value to the company than to the supplier. CQE II-103/104
A surface profiler or profilometer is the most common method of measuring surface roughness, although
other methods (ultasonic, eddy current probe and 3D laser scanning) are possible.
Roughness represents the size of the finely distributed surface pattern deviations CQE VI-29
Lay represents the dominant direction of the surface pattern (such as grinding scores)
Waviness respresents the deviations which are relatively far apart. (most difficult to detect with the naked
eye.) CQE VI-30
Roughness represents the size of the finely distributed surface pattern, Lay represents the dominant
CQE VI-30,31
direction of the surface pattern (ie from grinding), Waviness represents deviations that are relatively far apart
Only moderately useful (often subjective, skewed toward the negative, skewed by regional personalities, and
minimally useful with "thrilling" information). Various methods of surveys, including written/mail surveys,
telephone surveys, in-person interviews, focus groups, customer councils, joint planning meetings, panels, CQE II-97,98
comment cards/suggestion boxes, observation, mystery shoppers, and listening posts. See CQM Handbook
for details.
Normal --> Tightened: When 2 out of 5 consecutive lots or batches have been rejected on original inspection
Tightened --> Normal: When 5 consecutive lots or batches have been considered acceptable on original
inspection.
Acronym for Strengths (internal, good), Weaknesses (internal, bad), Opportunities (external, good), and
Threats (external, bad). Method used by companies to evaluate themselves and plan a strategy. Primary
weakness is objectivity when the SWOT is being performed (voiced by Goodstein, Bowman, & Porter)
Systematic method to outline all the details needed to complete a given objective. It is an orderly structure
similar to a family tree chart or an organization chart. Organization is by levels of importance. Suspiciously CQE VIII-36
similar to the cause-and-effect diagram process, except starts with the 'whys' and moves toward the 'hows.'
Knowledge that resides in the minds of employees, but not in a database (also less formally referred to as
'tribal knowledge.'
Best known for the Taguchi Loss Function (financial loss to society after an article is shipped), and Designed
Experiments (DOE). His system began with optimizing the design of products and processes. His tools tell
CQE II-4,20,21
us how to make something happen (proactive) versus responding to things that have already happened.
IV-21:27
Also performed work on signal-to-noise ratio. Robust products. Approx 50% of engineers in Japan are
familiar with Taguchi methods.
Theoretical quadratic (or parabolic) relationship that calculates loss to society due to poor quality products.
CQE III-65
See Primer for equation and example. See also Loss to Society for equation.
The available production time divided by the rate of customer demand. For example if a customer wants 480
widgets a day and the factory operates 960 minutes per day, the takt time is two minutes. Takt time becomes CQE VIII-14,22
the heartbeat of any lean organization
One of the first contributors to the scientific theory of management, which had the aim of finding the one best
way to perform a task so as to increase productivity. He emphasized efficiency, not satisfaction, of workers.
CQM Handbook p.274
He did this by breaking a job down into small-task components that could be studied to find the more efficient
way of doing it.
Given a number K greater or equal to 1 and, for any set of n measurements, at least (1-1/K 2) of the
measurements will lie within K standard deviations of their mean. This applies to any set of measurements,
regardless of the shape of the distribution. See also Camp-Meidel extension. CQE IX-21
Organizes and leads meetings. Updates sponsors/steering committees. Supports factors for effective
CQE II-67
teamwork.
Waning management support, lack of documentation of meeting results, inadequate time or training, lack of
CQE II-77
people with facilitation skills, etc… See primer for more details
Forming: Stage 1 of team development (build phase) - Proud to be selected, anxious, new team experience,
forming team structure.
Storming: Stage 2 of team development (build phase) - Individualistic thinking, tug of external loyalties,
fluctuating attitudes, confrontations, team's task sinks in
Norming: Stage 3 of team development (develop phase) - Team coalesces, cooperation, willing to dialogue,
conflicts reduced, focus on team objective. CQE II-73,74,75
Performing: Stage 4 of team development (optimize phase) - Team has matured, smooth cohesive unit,
focus on process, feelings of satisfaction, achieving goals.
See also Team Opposition CQE II-77 and Additional Problem Areas CQE II-77
With regards to sampling errors: Errors that are made consistently by some individuals and may indicate a
lack of training, lack of skill, or a lack of capacity. Remedies include: additional training, product CQE V-62
magnification, and/or individual replacement.
Term used for a former customer who turns on you after a negative experience. CQM Primer V-37
Nominal size, Threads per inch, Thread Series, Thread Class Symbol, Double, RH, LH, MOD CQE IV-46
Switching to tightened inspection is usually done when quality levels are observed to be getting worse. CQE V-48
Telecom Standard
Emphasized by Feigenbuam (USA) and Ishikawa (Japan) in the early 1980s. CQE II-11,12
A management system for a customer focused organization that involves all employees in a continuous
improvemt of all aspects of the organization. It's an integrative system that uses strategy, data, and effective
communications to integrate the quality discipline into the culture and activities of the organization. 1)
CQE II-3, VIII-29,30
Customer focused, 2) Total Employee Involvement, 3) Process Centered, 4) Integrated System, 5) Strategic
and Systematic Approach, 6) Continual Improvement, 7) Fact-Based Decision Making, 8) Communications
(same as ISO-9000 items, except #8 for ISO is Supplier-focus).
Accronym. (Total Productive Maintenance) Involves equipment availability. See Availability, also
Effectiveness. The six big losses to equipment effectiveness include 1) equipment failure, 2) set-up and VIII-22
adjustment, 3) idling and minor stoppages, 4) reduced speed, 5) process defects, and 6) reduced yield.
A management system for a customer focused organization that involves all employees in a continuous
improvemt of all aspects of the organization. It's an integrative system that uses strategy, data, and effective
communications to integrate the quality discipline into the culture and activities of the organization. 1)
CQE II-3, VIII-29,30
Customer focused, 2) Total Employee Involvement, 3) Process Centered, 4) Integrated System, 5) Strategic
and Systematic Approach, 6) Continual Improvement, 7) Fact-Based Decision Making, 8) Communications
(same as ISO-9000 items, except #8 for ISO is Supplier-focus).
The ability to trace the history, application, or location of a product, and in some cases, service, by means of
CQE V-3
recorded identifications
Company strategy. See Leading Edge for strategy & product types list.
Leads to measurable, value-added skills. Not to be confused with education, which leads simply to
increased knowledge.
CQE III-66,68,69,
70,71,72
Training only a partial crew (for instance only 1 shift) could be to address a specific issue identified with that
shift. (This is not a cost saving measure, it is a corrective action)
See Systematic Diagram (used to outline all of the details needed to complete a given objective)
CQE VII-26, 27
CQE VII-17/20
Time sequence (dynamic) data analysis. Identifies change over time, patterns, cycles, process shifts,
increasing or decreasing variability, etc. Quality Improvement tool. Examples include Control Charts, Run CQE X-41
Charts, and Trend Charts.
See Altschuler, G. Incorporates 'tricks,' 'physical effects,' and combinations of the two. Theory of inventive
problem-solving.
SOH-CAH-TOA
Sine = O/H
Cosine = A/H
Tangent = O/A
Test to compare the means (averages) of two samples in an effort to determine the likelihood that both
CQE XI-13, 32
samples originated from the same population. Often used in conjunction with the F-test (see ANOVA)
John Tukey. Inventor of the Boxplot (1977) and stem-and-leaf plots (1977). CQE IX-24
Type 1 = , (alpha) results when null hypothesis is rejected when it is in fact true also known as producers
risk, Type 2 = - (beta) results when null hypothesis is not rejected when it should be rejected, also known as
CQE XI - 8
consumers risk. A small increase in alpha increases the beta risk. Increasing sample size can reduce both
alpha and beta risks.
Six Sigma, Improvement Teams, Project Teams/Task Force/Ad Hoc, Cross functional, Self Directed, Quality
CQE II-64,65,66
Circles, Quality Teams, Natural Work Teams
Union. Deals with probability. A U B means the Union of event A and event B, and contain all sample points
CQE IX-38
in event A, or B, or both.
Non-destructive test. See Primer. Ultrasonic testing equipment is portable, has good penetration capability
and results can be recorded. It requires skilled operators, and good test standards. Coupling materials such CQE VI-49
as water, glycerine or petroleum jelly must be used and complex part geometries present difficulties.
Union. Deals with probability. (A ~ B) means the Union of event A and event B, and contain all sample
CQE IX-38
points in event A, or B, or both.
Confirmation by examination and provision of objective evidence that the particular requirements for a
CQE V-3
specific intended use are met
Same as reproducability, accuracy, on target, or bias. Measure of how close average of a sample compares See reference to
to optimal target (nominal spec.). Not to be confused with repeatability. BIAS on this page
A graphical flow charting technique that shows material and information flows CQE VIII-15,16,17
Total Product Variability is the total variability in a product, which includes the variability of the measurement
CQE X-9, 43
process
Regarding sampling, not only is an item determined to be good or bad, buta also how good or how bad. CQE IX-6
Measured data. Nearly always can extend to multiple decimal points. Answers 'how long,' 'what volume,'
CQE IX-3, 4
how much time,' or 'how far.' Generally measured with some instrument or device.
The most important step in vendor certification is to familiarize the vendor with quality requirements CQE II-104/105
Diagram consisting of intersecting circles, usually used in studies of probability (union and intersection) CQE IX-38
Confirmation by examination and provision of objective evidence that specified requirements have been met CQE V-3
The vernier scale makes possible readings as fine as one thousands of an inch by visually subdividing the
CQE VI-24
smallest divisions on the scale. CQE VI-12
See Primer. See also hardness testing method summary matrix in Primer CQE VI-59 CQE VI-59, 61
Describes a future state, perhaps 5 to 10 years into the future. Not to be confused with a Mission Statement,
CQE II-24
which generally refers to day-to-day views
Ulitizling visual standards and Samples to describe deftects for inspection purposes is weak, but minimally
CQE IV-5
effective.
Written permission to accept for use a completed, but nonconforming, item either "as is," or upon completion
CQE V-19
of rework for a specified number of units or period of time
Believed bureaucracy to be the ideal organizational structure, since everyone would clearly understand their
responsibilities, and that lines of authority would provide more predictable and efficient output.
See book for formulas, graph, and applications. The scale parameter is always the point where 63.21% of
the population fall below it.
If the shape parameter ß(beta) =1, the Weibull distribution is identical to the exponential distribution, if
CQE IX-27, 53
ß= 2, the Weibull distribution is identical to the Rayleigh distribution; if ß is between 3 and 4 the Weibull
distribution approximates the normal
The layout of machines or business process of different types performing different operations in a tight
CQE VIII-14
sequence. Typically a 'U' or 'L' shape to permit single piece flow and flexible deployment of human effort
Work instructions provide details and a step by step sequence of activities. Flow charts may be used to
show relationships of the process steps. Typically the people that perform the activities should be involved in CQE V-4
creation of the work instructions. See also "Procedures"
Control charts plottirrg individual data points on one graph, and a moving range on another (similar to X bar -
R) are the most common and most applicable charts for calibration and testing.
The X-MR chart (for individuals and moving ranges) is the only control chart which may have
specification limits shown. CQE X-18/21
However, there are some drawbacks in the interpretation and use of X-MR charts:
CQE X-20
• All interpretation is faulty if the data distribution is not normal
• X-MR charts do not separate piece-to-piece repeatability of the process
• Averages and limits can have wide variability until 80-100 readings are taken
• X-MR charts are not as sensitive to process changes as the X- R chart
n = sample size, x = a reading (data), X-bar - average of readings in sample, X double bar = average of all
the X-bar's, R=the range, R-bar = average of all the R's. UCL/LCL Upper and lower control limits. These are
the bounderies for 99.73% of the population. See example of how to create the data on CQE X-14.
It may be economical to allow an X-bar chart to go out of control when 6= is appreciably less than the CQE X-12,13
difference between specifications. CQE X-37/43 and 56/57
Plotting a single point by accident on a chart that is based on a sample size of 4 would put that point at risk
of being outside control limits. CQE X-12/14