book

World Class Quality: Using Design of Experiments to Make It Happen

Name: World Class Quality: Using Design of Experiments to Make It Happen
ISBN: 0814404278

by Keki R. Bhote, Adi K. Bhote

November 1999

Intermediate to advanced

488 pages

10h 37m

English

AMACOM

Read now

Unlock full access

Copyright
List of Figures
List of Tables
Foreword
Preface to the Second Edition
Acknowledgments
I. Introduction
1. The Need for and Objectives and Benefits of Design of Experiments
A. The NeedThe Killing of Locusts Chinese StyleHenley's LawB. The ObjectivesC. The Benefits
2. Fads, Potions, Nostrums of the Quality Movement in the Last 50 Years
Lurching From Fad to FadEffective Quality Systems
3. A Powerful Tool Kit for the 21st Century
Elementary ToolsThe Seven Tools of QC—The Kindergarten ToolsThe Seven Quality Management Tools—Another Needless ComplicationThe 8-D Methodology—A BoondoggleThe Engineering Approach (Observe, Think, Try, Explain)Worker InvolvementThe Ten Powerful Tools for the 21st Century7Tool 3: Quality Function Deployment (QFD)—Capturing the Voice of the Customer8Applicable AreaThe NeedObjectivesBenefitsMethodologyNotes of CautionTool 4: Total Productive Maintenance (TPM)—Improving Equipment Productivity9Applicable AreaThe NeedObjectivesBenefitsMethodologyNotes of CautionTool 5: Benchmarking—Learning and Adapting From the Best10Applicable AreasThe NeedObjectivesBenefitsNotes of CautionTool 6: Poka-Yoke—Eliminating Operator Controllable Errors11Applicable AreasThe NeedObjectivesBenefitsMethodologyNotes of CautionBeforeAfter (Poka-Yoke Solution)Best SolutionTool 7: Next Operation as Customer (NOAC)—White Collar Quality, Cost, and Cycle-Time Improvement12,13Applicable AreasThe NeedObjectivesBenefitsMethodology: The 10 Step Roadmap for NOACNotes of CautionTool 8: Supply Management—Breakthrough for Supplier Quality, Cost, and Cycle-Time Improvement,14Applicable AreasThe NeedObjectiveBenefitsMethodologyNotes of CautionTool 9: Total Value Engineering—Going Beyond Traditional Value Engineering15Applicable AreasThe NeedObjectivesBenefitsMethodologyNotes of CautionTool 10: Cycle-Time Reduction—The Integrator of Quality, Cost, Delivery and Effectiveness14Applicable AreasThe NeedObjectivesBenefitsMethodologyNotes of CautionAwareness and Implementation of These 21st Century Tools

II. Preamble to the Design of Experiments
4. The Measurement of Process Capability
Why Variation Is EvilA Football AnalogyCase StudiesThe Ford versus Mazda Case StudyThe Nippon Telegraph and Telephone Case StudyThe F-16 Fighter Jet Case StudyThe Build-Up of TolerancesThe Enormous Cost of Inspection, Test, Rework, and ScrapIntroduction to Process Capability: CpCpk, A Better Measure of Variation and Process CapabilityPitfalls to Avoid in Cp,Cpk MeasurementsPitfall 1. Measuring All ParametersPitfall 2. Measuring Cp,Cpk ConstantlyPitfall 3. If Specification Limits Are Not Correct, Cp and Cpk Are MeaninglessPitfall 4. Extending Cp Measurements to Suppliers Without Doing So in Your Own CompanyWorkshop Exercise 1: The Press BrakeQuestionsAnswersWorkshop Exercise 2: Supplier Tolerances versus CostAnswerWorkshop Exercise 3: Customized Matching of Light Emitting Diodes (LEDs)AnswerQuestions for Top Management
5. Variation: An Industrial Epidemic
The Many Sources of VariationPoor ManagementPoor Product/Process SpecificationsPoor Component SpecificationsInadequate Quality Tools and SystemsPoor Manufacturing PracticesPoor Supplier MaterialsOperator Errors
6. The Three Approaches to DOE: Classical, Taguchi, and Shainin
Ineffective Approaches to Problem Solving and Variation ReductionBackground on the Three Approaches to DOEClassical DOETaguchi DOEShainin DOEThe Motorola Case Study: 10:1, 100:1, 1,000:1 Quality ImprovementThe Pot of GoldBob Galvin's Inspiring LeadershipToolsThe Arcade Pilot DOEThe Five-Plant Pilot Run in the U.S.The Worldwide Spread of DOEFundamental Weaknesses in Classical and Taguchi Approaches1. Techniques: Limited and Ineffectual for Classical/Taguchi; Versatile and Powerful for Shainin2. Clue Generation: ''The Parts Are Smarter Than the Engineers!''3. Effectiveness: Low Success Rates for Classical/Taguchi; High for Shainin4. Cost: High Cost of Failure in Classical/Taguchi versus Lower Costs of Success in Shainin5. Complexity: Incomprehensible Statistical Jargon in Classical/Taguchi versus Simple, Logical Concepts in Shainin6. Time in Instruction: Three Days to Two Weeks of Seminars for Classical/Taguchi versus One Day for Shainin; Time in Execution: False Starts in Classical/Taguchi versus Sure-Fire Clues in Shainin7. Statistical Validity: Confounding and Saturation in Classical/Taguchi versus Clear Separation of Main and Interaction Effects in Shainin8. Applicability: Shainin Best at the Design Stage of a Product or Process, in Production, at the Supplier, in the Field, Even in Administrative Applications9. Ease of Implementation: Classical/Taguchi—Difficult, Slow, Tedious; Shainin—Easy, Fast, Exciting10. Disruption of Production: Classical/Taguchi—Disruptive; Shainin—NondisruptiveAn Open Challenge
7. A Synopsis of the 10 Powerful DOE Tools
Linking the 10 DOE ToolsClue-Generation Tools—Filtering Out Unimportant VariablesFormal DOE Tools—Homing In on the Red XSafeguards Against RetrogressionA Generic Problem-Solving FrameworkA Capsule Summary of the Ten DOE ToolsThe Green Y, Red X, Pink X, and Pale Pink XDefining and Quantifying the Problem: The Green Y—A ChecklistMeasurement AccuracyPrecision, Bias, Accuracy, Discrimination in MetrologyReducing Instrument VariationsReducing Within-Instrument VariationReducing Instrument-to-Instrument VariationReducing Operator-to-Operator VariationMeasurement Accuracy ChecklistConverting an Attribute into a VariableLikert ScaleLikert Scale Achieves Smaller Sample SizesMeasurement Case Study: Electric Razor
III. ''Talking to the Parts''—A Set of Clue-Generating Tools
8. The Multi-Vari Analysis: Homing In on the Red X
Purpose and PrinciplesA Card Trick to Illustrate Homing In on the Red X With a Multi-Vari AnalysisThe Multi-Vari Principle Behind the Card TrickThe Futility of the Cause-and-Effect (Ishikawa) DiagramThree Major Families of Variation: Positional, Cyclical, and TemporalPositional Variation (Within Unit Family)Cyclical Variation (Unit to Unit Family)Temporal Variations (Time to Time Family)Preamble to Any DOE StudyDesigning, Conducting, and Analyzing a Multi-Vari Study: A RoadmapA. Designing the Multi-Vari StudyB. Running the Multi-Vari ExperimentC. Interpreting and Analyzing the Multi-Vari ChartCase Study 1: Designing the Multi-Vari—Printed Circuit Board DrillingDetermining Sampling Frequency and Number of Units RequiredWorkshop Exercise 1: Semiconductor Wafer Multi-Vari PlanAnswersConstructing a Multi-Vari ChartCase Study 2: The Rotor ShaftQuantification of Each Family and Subfamily of VariationTwo Types of Multi-Vari ChartsCase Study 3: Radio Rejections—''No Trouble Found'' CategoryWorkshop Exercise 2: Paint Defects on a Washing Machine TubConclusions and Mistakes Made in the Tub Paint Defects ExerciseWorkshop Exercise 3: Tile AdhesionQuestionsAnswersWorkshop Exercise 4: An Unsuccessful Outcome—Belt Passivator Multi-Vari Case StudyQuestionsAnswersCase Study 4: ''Tombstoning''Multi-Vari Applications in Administrative WorkCase Study 5: Hospital Billing ErrorsCase Study 6: Customer Service at a HotelCase Study 7: High Employee TurnoverQuestions for DOE TeamsQuestions for Top Management During DOE Project Reviews
9. The Concentration Chart: Pinpointing Locations of Repetitive Problems
PurposeConstructing a Concentration ChartCase Study 1: Foam Leaks on Refrigerator DoorCase Study 2: Wave Solder DefectsWorkshop Exercise 1: Paint DefectsAnswersWorkshop Exercise 2: Shaft Distortion After Heat TreatQuestionAnswerApplications in Administrative WorkCase Study 3: Shorted Horizontal Output TransistorCase Study 4: A University Recruitment DriveQuestions for DOE TeamsQuestions for Top Management During DOE Project Reviews
10. Components Search: Simple, Smooth Swapping
IntroductionA Humorous Analogy: Incident on a British TrainBypassing the Multi-Vari With Components SearchDisarming the Critics of Components SearchThe Four Stages of Components SearchComponents Search: PrerequisitesA Procedure in 12 StepsCase Study 1: The HourmeterStage 1: Disassembly and ReassemblyTest for Significant Repeatable DifferencesConclusionThe Graphical PlotDecision LimitsAn Alternative ''Rule of Thumb'' to Assess the Importance of a Component Exchange in Stage 2Main Effects and Interaction EffectsStage 4: Factorial Analysis''Talking to the Parts''The Final Solution to the Hourmeter Case StudyEngineering AnalysisResultsAlternative SolutionsLet the Parts, Not the Engineers, Determine the Correct SpecificationsCase Study 2: The Nonfiring BurnerStage 1Stage 2Workshop Exercise 1: A Car Company's Ice Cream Fiasco!QuestionsAnswersFurther DOE ExperimentsAnalysisConclusionWorkshop Exercise 2: Oscillator Time DelayQuestionsAnswersConclusionComponents Search Practices When Disassembly/Reassembly Is Not RepeatableCase Study 3: Refrigerator Door GapWorkshop Exercise 3: The ''Walking'' MachineQuestionsAnswersComponents Search Applications in Processes/Machines/LinesCase Study 4: Spring Torque Differences on Two LinesCase Study 5: Operator Performance DifferencesComponents Search Applications in Administrative WorkWorkshop Exercise 4 (Game): Find the Red X Ball (1)Workshop Exercise 5 (Game): Find the Red X Ball (2)Answer to Workshop Exercise 4: Find the Red X Ball (1)Answer to Workshop Exercise 5: Find the Red X Ball (2)Components Search Questions for DOE TeamsGeneralStage 1Stage 2Stage 3Stage 4Questions for Top ManagementEndnotes
11. Paired Comparisons: An Elegant, Versatile Tool
IntroductionPrerequisitesMethodology Alternative AMethodology Alternative BThe Tukey TestTukey Test ProcedureTotal End-Count and Confidence LevelsTukey Test—An ExampleStatistical Principles in the Tukey TestCase Study 1: Micro-Motor NoiseAction TakenComments on Micro-Motor Case StudyWorkshop Exercise 1: Contact Lens Prescription DefectsQuestionsAnswersCase Study 2: The 4-Megabit Dynamic RAMWorkshop Exercise 2: Out-of-Square GrillsQuestionAnswerCommentsTukey Test Ranking When the Center Is Good and the Tails Are BadApplications of Paired Comparisons in Administrative WorkCase Study 3: Dairy Farm Bacteria LevelsCase Study 4: High Turnover of Temporary EmployeesQuestions for DOE TeamsQuestions for Top Management ReviewSequel to Paired Comparisons Experiments: A Note of Caution
12. Product/Process Search: Pinpointing Process Variables
Product/Process Search ObjectiveProduct/Process Search PrinciplesProduct/Process Search MethodologyProduct/Process Search: When Process Parameters Cannot Be Measured on Individual UnitsCase Study 1: Stale Lettuce/Bag LeakageThe ProcessAn Earlier Green YResultsFollow-Up Experiments and ConclusionCase Study 2: Bubbles in Glass ScreenFlow ChartResultsWorkshop Exercise 1: Plastic Injection MoldingQuestionsAnswersQuestions for DOE Teams as Guidelines for Product/Process SearchQuestions for Top Management in Following DOE Team Progress
IV. Formal Design of Experiments Techniques to Characterize a Product/Process
13. Variables Search: The Rolls Royce in the Search for the Red X
From Clue Generation to Distilling the Red XFormal DOE TechniquesThe Full FactorialLimitations of All Types of Fractional FactorialsTaguchiClassical Fraction Factorials—Only Slightly Better Than TaguchiConfounding Main Effects With Interaction Effects—An ExampleWeaknesses of a Half–Replicate Fraction FactorialVariables Search: Its Advantages Over All Fraction FactorialsVariables Search ObjectivesThe Binary Search PrincipleVariables Search MethodologyStage 1. Ball ParkStage 2. Separation of Important and Unimportant FactorsStage 3. Capping RunStage 4. Factorial AnalysisAn Alternate ''Rule of Thumb'' to Assess the Importance of a Factor in Stage 2Interaction Effects in Variables SearchProduct/Process CharacterizationCase Studies and Workshop ExercisesCase Study 1: The Press BrakeStage 1. Ball ParkStage 2. EliminationStage 3. Capping RunStage 4. Factorial AnalysisA Graphical Plot of Case Study 1, The Press BrakeCommon Mistakes Made by Variables Search ExperimentersStage 1Stage 2Stage 3Stage 4Post Variables SearchWorkshop Exercise 1: Engine Control ModuleQuestionsAnswersCase Study 2: Cracked EpoxyGreen YConclusionsWorkshop Exercise 2: Screen Printing (Unsuccessful)QuestionsAnswersCase Study 3: Tombstone Defects (A Wrong Technique)Case Study 4: Contact Lens Cylinder Not Meeting SpecificationsConclusions and CommentsQuestionsAnswersVariables Search: Software ApplicationsCase Study 5: Software ApplicationsA Combination of Variables Search and Paired Comparisons as a Derivative DOE TechniqueProduct ApplicationsVariables Search/Paired Comparison Case Study 1: Travel MotorVariables Search/Paired Comparison Case Study 2: Instrument IntermittencyVariables Search/Paired Comparisons in Administrative ApplicationsAnalogy With Medical/Pharmaceutical ResearchAdministrative Case Study: Appliance ServicerQuestions for DOE Teams as GuidelinesQuestions for Top Management to Review DOE Team Projects
14. The Full Factorial: Purest Technique for Separating and Quantifying Each Interaction Effect
LimitationsObjectivesPrinciplesBalanceReplicationRandomizationMethodologyGreen YThe MatrixCase Studies and Workshop ExercisesCase Study 1: Motorola Wave SolderingConfirmation Experiment: Capping RunWorkshop Exercise 1: Paint YieldsQuestionsAnswersPostscriptCase Study 2: Moire Effect MinimizationAnalysisResultsCommentary on Case Study 2: Moire Effect MinimizationWorkshop Exercise 2: Drilling OperationGreen YQuestionsAnswersConclusionQuestions for DOE TeamsQuestions for Management
15. B versus C: An Excellent Validation Technique
The Importance of ValidationDisc Brake Case StudyWhat Is B versus C?ObjectivesPrinciplesMethodologyRisk and Confidence LevelsSample Size for B's and C'sImportance of RandomizationRank OrderDecision Rule&b.beta; Risk: Determining the Magnitude of Improvement of B Over CCase Studies and Workshop ExercisesCase Study 1: Cracked EpoxyWorkshop Exercise 1: Wire Bond StrengthQuestionsAnswersCase Study 2: The Press BrakeIncreasing B and C Sample Sizes to Reduce Uncertainty of Predicting the Magnitude of B Improvement Over CModified Tests to Determine That C Is Worse Than BExpanding the Number of Alternatives Beyond Two (Beyond B and C)Workshop Exercise 2: Choice Between Four SuppliersQuestionsAnswersA Special B versus C Stress Tests to FailureCase Study 3: Gear Train MotorIndustrial Applications for B versus CAdministrative Applications for B versus CCase Study 4: Automobile StylingB versus C in Opinion Surveys: A Poor Man's Gallup PollQuestions for DOE Teams as Guidelines for B versus CQuestions for Top Management in Reviewing DOE Team Projects
V. DOE Optimization
16. Scatter Plots to Achieve Realistic Specifications and Tolerances
ObjectiveThe Importance of Realistic Specifications and TolerancesCase Study: Edge Defects in Contact LensesPrinciplesMethodologyQuiz No. 1Quiz No. 2Quiz AnswersDetermining Realistic Specifications and TolerancesCase Study 1: Ignition AmplifierWorkshop Exercise 1: Infrared SolderingQuestionsAnswersCase Study 2: Cylinder Prescription versus Lens AxisConclusionsWorkshop Exercise 2: Digital CircuitQuestionsAnswersUse of Scatter Plots in Administrative ApplicationsCase Study 3: Customer Satisfaction versus Customer Loyalty versus ProfitMultiple Regression AnalysisExampleQuestions for DOE Teams as Guidelines for Scatter PlotsQuestions for Top Management
17. Response Surface Methodology (RSM): To Optimize Interactions
ObjectivesApproachesThe Concept: A Mountain Climbing AnalogyEvolutionary OperationProcedureCase Study: Maximizing Process Yields—Schottky DiodesStage 1Stage 2Pitfalls to AvoidExpanding EVOP to Three Interacting Input VariablesExpanding EVOP to Four Interacting Input VariablesSimplexCase Study: Paint Process OptimizationStage 1Stage 2Stage 3Random Evolutionary OperationProcedureQuestions for DOE Teams as Guidelines in Response Surface Methodology (RSM) ProjectsQuestions for Top Management During DOE Team Project Reviews
VI. The Transition From DOE to SPC: Vital Disciplines
18. Positrol: Freezing Process Gains
The ConceptThe Positrol PlanThe Positrol ChartReasons for RetrogressionCase Study 1: The Pick-and-Place MachineWorkshop Exercise 1: P.C. Board SolderingQuestionsAnswersCase Study 2: CNC LatheCase Study 3: Plastic Injection Molding MachineCase Study 4: Die CastingQuestions for DOE Teams as Guidelines for Positrol ProjectsQuestions for Top Management During DOE Project Reviews
19. Process Certification: Eliminating Peripheral Causes of Poor Quality
Murphy's LawA Generic List of Peripheral Causes of Poor QualityManagement/Supervision InadequaciesViolation of Good Manufacturing Practices (GMP)Plant/Equipment InattentionEnvironmental NeglectHuman ShortcomingsProcess Certification MethodologyCase Study 1: SCAP MetallizationCase Study 2: Punch PressCommentWorkshop Exercise 1: Porcelain Paint ProcessQuestionsAnswers (1 and 2)Workshop Exercise 2: Poor Yields on Memory ChipsQuestionAnswerQuestions for DOE Teams as Guidelines for Process Certification As Well As for Management During DOE Project ReviewsProcess Certification Audits and ScrubsManagement/SupervisionGood Manufacturing PracticeMetrologyTotal Productive MaintenanceEnvironmentPeople
VII. Statistical Process Control: For Monitoring Quality
20. Control Charts: A Technique Whose Time Has Gone
The Roller-Coaster History of Control ChartsThe Weaknesses of Control ChartsCase Study 1: A BushingCase Study 2: Sensor CapacitanceSlipshod Control Chart PracticesQuestions for DOE Teams and Top Management
21. Pre-Control: A Technique Whose Time Has Come
The Discovery of Pre-ControlThe Mechanics of Pre-Control in Four Easy StepsRule 1. Establishing Pre-Control Lines (Limits)Rule 2. A Shortcut to Determine Process CapabilityRule 3. Pre-Control in Production: Sample Size; Continuing/StoppingRule 4. Frequency of SamplingSpecial Circumstances Requiring Modifications of Pre-Control StepsThe Theory of Pre-Controlα Riskβ RiskThe Enormous Statistical Power of Pre-ControlCharting Pre-Control: Easing the Operator's BurdenTackling Multiple Quality CharacteristicsReasons Why Some Companies Are Opposed to Pre-ControlA Capsule Comparison of Pre-Control and Control ChartsThe Use of AveragesFeedback of ResultsCase Study 1: Wave SolderingWorkshop Exercise 1: BushingQuestionsAnswersWorkshop Exercise 2: Sensor CapacitanceQuestionAnswerWorkshop Exercise 3: The Wire BonderQuestionsAnswersQuestions for DOE Teams on Pre-ControlQuestions for Top Management During DOE Project Reviews
VIII. Linkage of DOE to Reliability
22. Multiple Environment Over Stress Tests: As Effective for Reliability as DOE Is for Quality
Quality versus ReliabilityFundamental Shortcomings in Traditional Reliability PracticesModest Reliability AdvancesRevolutionary Reliability Test PrinciplesRelationship of Failure and StressStress Acceleration in Thermal CyclingStress Acceleration in VibrationHighly Accelerated Life Tests (HALT)PrinciplesMethodologyHALT: Pros and ConsDefinitions of LimitsHighly Accelerated Stress Screening (HASS)Multiple Environment Over Stress Tests (MEOST)HistoryThe Objectives of MEOSTThe Benefits of MEOSTMEOST Methodology: A RoadmapStresses and LevelsSample SizesThe Seven Stages of MEOSTStage 1: Single Stress to Design Stress LimitStage 2: Single Stress up to Maximum Practical Overstress Limit (MPOSL)Stage 3: MEOST to Maximum Practical Overstress Limit (MPOSL)Stage 4: MEOST at Pilot RunStage 5: Truncated MEOST in Ongoing ProductionStages 6 and 7: MEOST on Good Field ReturnsA Comparison of Overall Effectiveness: HALT/HASS versus MEOSTThe Close Linkage Between MEOST and Design of Experiments (DOE)1. Solving Elusive Problems2. Tackling ''No Trouble Found'' (NTFs) or ''No Apparent Defects'' (NADs)3. Evaluating the Effectiveness of Engineering Changes4. Cost ReductionConclusion
IX. A Logical Sequence in the Use of DOE Tools
23. Case Studies in Sequential DOE Tools to Solve Chronic Problems
Introduction: Difficulties Experienced by Beginning Practitioners in Implementing DOE1. The Root Cause Red X May L-ie Five or Six Layers Deep2. Not Turning the Experiment ''On'' and ''Off''3. Not Establishing Realistic Tolerances4. No Positrol5. No Process Certification (too much ''noise'' at the start of an experiment)A Continuum of Linked DOE ToolsCase Study 1: Wave SolderingDOE Experiment 1: Multi-VariDOE Experiment 2: The Concentration ChartDOE Experiment 3: Paired Comparisons of BoardsDOE Experiment 4: Paired Comparison of Unsoldered ConnectionsDOE Experiment 5: B versus C ValidationDOE Experiment 6: Variables SearchConclusionsDOE Experiment 7: B versus C ValidationDOE Experiment 8: Optimization Through a Full FactorialConclusionsDOE Experiment 9: Final Optimization With Scatter PlotsPositrolProcess CertificationCase Study 2: Dome Tweeter SensitivityDOE Experiment 1: Components Search on Entire AssemblyDOE Experiment 2: Paired Comparisons on DomeDOE Experiment 3: Paired Comparison on MagnetDOE Experiment 4: The Full FactorialConclusionDOE Experiment 5: Multi-Vari Study of Glue Weight on DomeDOE Experiment 6: Multi-Vari Study of Glue Weight on Coil and Dome (not shown)DOE Experiment 7: B versus C—Automatic versus Manual Applications of GlueDOE Experiment 8: Position of the Voice Coil in the Air GapDOE Experiment 9: Change in the Magnetizing Unit (B versus C test)DOE Experiment 10: Final ResultsConclusion
X. From Classroom Instruction to True Learning on the Job
24. Learning by Doing
Practice, Practice, PracticeEnergizing a Problem-Solving CultureStep 1: Top Management Commitment and InvolvementStep 2: Conversion of the Design EngineerStep 3: Quality Management—From Policeman to CoachStep 4: Production ManagementStep 5: SuppliersStep 6: The Use of Skunk Works—An Unobtrusive Pilot EffortStep 7: The Untapped Gold Mine of Line Workers' ContributionCase Study: Dendermonde, BelgiumJoy in the Workplace
References

Content preview from World Class Quality: Using Design of Experiments to Make It Happen

Chapter 17. Response Surface Methodology (RSM): To Optimize Interactions

Objectives

The purpose of scatter plots, detailed in the last chapter, is to determine the best levels of noninteracting input variables to optimize a Green Y.
The purpose of Response Surface Methodology (RSM) is to determine the best combination of levels of two or more interacting input variables (identified in previous DOE experiments) to achieve a maximum, minimum, or optimum Green Y. (Response, output, and Green Y are synonymous terms.) For the place of RSM in the problem-solving roadmap, see Figure 7-1.

Approaches

There are a number of empirical, statistical, and mathematical approaches to optimization and RSM—

Evolutionary Operation (EVOP),
Simplex;
Random Evolutionary Operation ...

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Read now

Unlock full access

More than 5,000 organizations count on O’Reilly

O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.

Julian F.

Head of Cybersecurity

I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.

Addison B.

Field Engineer

I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.

Amir M.

Data Platform Tech Lead

I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.

Mark W.

Embedded Software Engineer

Publisher Resources

ISBN: 0814404278

Cloud Computing

Data Engineering

Data Science

AI & ML

Programming Languages

Software Architecture

IT/Ops

Security

Design