ASQ RRD series webinar
Robustness Thinking in Design for Reliability
A Best Practice in Design for Reliability
Dr. Matthew Hu SVP Engineering and Quality HAYLION Technologies March 11, 2021
https://attendee.gotowebinar.com/register/2625796907172545805
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Popcorn Story
Who Never Tasted Popcorn before?
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Not simply a REHASH of the lessons learned
Microwaves popcorn-making is not simply a REHASH of the lessons learned in conventional way of popcorn-making, but a fundamentally different and proactive methodology.
DO THINGS RIGHT (DMAIC Mind-Set) Conventional Way of Popcorn-making, fixing existing process, following conventional experience and procedures and … However, quality of popcorn is still heavily based on experience and …
DO RIGHT THINGS (Design Thinking Mind-Set) Microwaves Way of Popcorn-making Right and Robust Technology, Good Quality of popcorn is not based (insensitive to) on experience
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Better, Faster and Cheaper
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Begin With the End in Mind (Covey - 7 Habits of Highly Effective People)
• Will your customer always use your product under best conditions? • Will your product always be manufactured under best conditions? No !
Variation Happens!!
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Objectives
• Define robustness • Explain product development using Robust Engineering versus traditional product development • Explain Robust Design for Reliability • Define Objective Function, Basic Function, and Ideal Function • Explain how Ideal Function and Two-step Optimization lead to robust technology development and achieve "Better, Cheaper, Faster" product development • Explain how to conduct a preliminary robustness assessment • Explain the value of robustness assessment • Case study in robust autonomous driving technology development
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Story in 1970 – SONY TV
0.3% Defects Customer satisfaction decreased when colour density deviated from target Made in Japan
Target 100% Compliance X But American families liked Made in USA the TV made in Japan better
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Story in 2018– Self-Driving Vehicle Accident
Arizona News https://www.theguardian.com/us-news/2020/sep/16/uber-self-driving-car-death-safety-driver-charged
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Reliability vs. Robustness Story
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Dr. Matthew Hu Introduction
Dr. Deming 4-Day Seminar
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Un-Reliability
Failure Rate
Infant Mortality Wear-out period (Early failure period)
Manufacturing Usage variation Inner variation & variation deterioration
Best period Constant failure rate Time
Production Processes Usage Environment Under Statistical Control? Under Statistical Control? Not Usually!!
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Control Methods and Tools In Bathtub Curve
Infant Mortality Useful Life Wear Out Life Controls Design Controls Controls • Selection of Low Defect rate Parts, Joints, • Selecting high reliability Parts, Joints, • Selecting appropriate internal stress Interconnects, Fasteners, etc. Fasteners, Interconnects, etc. levels and materials to meet • Selection of Low Defect Rate Processes • Designing with the minimum number of expected time to failure or • Designing Within Process Capability Processes, Parts, Joints, Fasteners, replacement • Robust Design Interconnects, etc. • Designing with low stress levels on Parts, Tools Conformance Joints, Fasteners, Interconnects, etc. • Robust Design Tools Tools • DOE • DFA – Part count reduction • Reliability Data and Statistical Models • Reliability Data and Statistical Models • Critical Parameter Management • Physics of Failures – Failure Mechanisms • DFMEA/PFMEA/DFA/DFM • Physics of Failures – Failure Mechanisms • DOE • DOE • Statistical Tolerancing • Deterministic design • Prediction • Design and Process Capability (Cp, Cpk) • Prediction • Test to Failure/ALT • Mistake-Proofing, SPC, Control Plans • Test to Failure • HALT and HASS • HALT/ALT
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 The Challenge of Reliability Theory Assumptions
Probability models under the assumption:
• Processes under statistical control? – Probably not!!!
• Lagging indictors of reliability performance – The design is created before testing – Usage feedback is even much later
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Reliability and Robustness (An Engineering Measure of Reliability) Target Target
Requirement
(.)
(.)
f f
s s
Quality Loss Quality Prob. Density, Prob. Density, Prob. Density, Prob. Density, m m Product Performance Product Performance Reliability: probability of a product performing Robustness: ability of a product to perform its its intended function for a specified life under intended function consistently in the presence of the operating conditions encountered. uncontrollable user environment (noise) during its intended life. In other words, the product is Q: How do you know the f(.) when a design is new? insensitive to noise.
Computing probability of success requires Assessing robustness requires knowledge of m, s knowledge of m, s, f (.)
ASQ Reliability and Risk Division series webinar 3/11/202114 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Back to Basic
◼ Work with the failure mechanisms
◼ And their relations to Variation!
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Reliability in a World Full of Variation
Without Variation Variation Creates Problems: No World! - Deviations Life is Variation! - Disturbances - Noise
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 What is Robustness?
Webster’s dictionary defines robustness as: • being powerfully built, sturdy • boisterous, rough • marked by richness and fullness
Dr. Taguchi defines robustness as: • the state where the technology, product, or process performance is minimally sensitive to factors causing variability (either in the manufacturing or user’s environment) at the lowest cost.
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 A Simple Connector Reliability Design
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Typical DfR Process Used by Companies DFR Stages & Activities CONCEPT DESIGN DEVELOPMENT MANUFACTURING SUPPORT PHASE PHASE PHASE PHASE PHASE DEFINE IDENTIFY Reliability Key Reliability Risks Objectives ASSESS Proposed Design Reliability QUANTIFY Analyze & Improve Reliability ASSURE Reliability
SUSTAIN
Monitor & Control Reliability DFR DFR STAGE
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Typical DfR Process Used by Companies DFR Stages & Activities CONCEPT DESIGN DEVELOPMENT MANUFACTURING SUPPORT PHASE PHASE PHASE PHASE PHASE
DEFINE IDENTIFY Reliability Objectives Key Reliability Risks ASSESS Proposed Design Reliability QUANTIFY Analyze & Improve Reliability ASSURE Reliability
SUSTAIN
Monitor & Control Reliability DFR DFR STAGE Life Data Demonstration Requirements & Change Point DOE FRACAS Goals Analysis Analysis Testing Accelerated Knowledge Environment & Robust Design Risk Assessment FMEA Testing Management Usage System Degradation Reliability Supplier Control Critical-to-Reliability Reliability Analysis Model Post Production (CTR) Simulation Failure Analysis PFMEA Warranty Data Requirements Baseline Reliability Critical Design Analysis Cascade Reliability Parameters Burn-in
Allocation Management Manufacturing Prognostic Health ACTIVITIES Physics of Prognostic Capability Management Failure Health Control Plan &
DFR DFR Management Process Control
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Charts Prepared by: Matthew Hu The Challenge of Reliability Design at The Lowest Cost
Reliability Efforts included Results
− d/pFMEA − Connectors were still − HALT disconnected as unexpected − Design of Experiments − One of the main reasons for − 8D an automotive safety recall − Weibull analysis conducted − SOP developed − SPC implemented − Operator trained
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Conventional Approach
• Design-Test-Fix • No cost integration • Test parts under severe conditions • Lack of leading indictor • Ignore deterioration, wear & degradation • Identify causes • Low knowledge gain • Minimize failures • Reliability checking by life test (reactive) • May lead to controlling some factors • Determine failure modes • Unstructured • Predict operational lives • Warranty claim and fire fighting • Eliminate the symptoms downstream • 1 symptom--1 problem--1 solution • High uncertainty • Trade off unknown • Lack of engineering confidence • Does not encourage technologies development
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Design Domains
Performance Target Performance CTQ/CTR Variation Reduction!
Critical Characteristics
y = f (x1, x2 , x3 )
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 The Challenge of Reliability Design at The Lowest Cost
Voids & Bubbles
Cross Section for Energy Variation
ASQASQ Reliability Reliability and and Risk Risk Division Division series series webinar webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Advantage of Robust Product Development Conventional Approach New Approach: Predictive Build-Test-Fix Design- in Quality Symptoms Focused Concept, Functionality Fire Fighting Effort Optimization Too many Product, process Effort Concept, Functionality iterativeOptimization optimization cost reduction loop New Paradigm
Design-Optimization-Verify-Launch
Reactive Reactive DesignQuality
Conception Design Freeze Mass Production Conception Design Freeze Mass Production Matthew Hu Fire From To • Evolving design requirements • Disciplined CTQ flowdown • Extensive design rework • Controlled design parameters • Product performance assessed • Product performance modeled and by “build and test” simulated • Performance and producibility • Designed for robust performance problems fixed after product in and producibility use • Functionally integrated product • Functionally serial product development development ASQ Reliability and Risk Division• Qualityseries webinar “designed 3/11/2021 in” • Quality “tested in” Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Control Factors & Noise Factors
Availability of Control Existence of Noise Factors Factors Typical company spends 70% of Engineers’ Fire Prevention by Time to fire- Robust Optimization fight.
R&D Advanced Product Product Design Mfg. Customer Engineering Planning Process Design Recycle
Upstream Downstream A slide from ASI
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Robustness Solves the Problem Target
Robustness low variation of ideal performance around the target value IN SPITE OF the effects of Noise Factors (uncontrollable user environment) Defects
Lower tolerance Upper tolerance limit limit
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Deterministic Design
Composite effect of System or Under a simple stress-strength framework, design potential electrical, component nominal are chosen to provide “strength” that exceeds thermal, capacity to the “stress” the product will experience mechanical or Safety Margin endure stress chemical loads It is known that stresses and strengths may vary, so safety margins are selected to minimize risk of failure, based on DESIGN 1 – Rules of Thumb Smaller safety – Past Experience margin, Mean Mean Stress Strength Without some probabilistic analysis, understanding of the nature of the variability and how it combines to Safety Margin affect performance is limited DESIGN 2 Larger safety Consider two designs from a solely deterministic margin perspective. Which is less likely to fail? Mean Mean Stress Strength
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Probabilistic Design
External How can design be insensitive to noise…? Environment Piece-to-piece Customer Variation Usage Safety Margin • Probabilistic analysis helps one understand the shape and dispersion Aging, Systems Degradation of variability caused by noise Interaction
• The interference region between DESIGN 1 stress and strength defines the Smaller safety Mean Mean probability of failure--this determines margin, higher Stress Strength reliability reliability Interference • A design with a larger safety factor DESIGN 2 Safety Margin region may have lower reliability depending Larger safety margin, lower upon stress and strength variability reliability Mean Mean Stress Strength
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 The 5 types of Noise factors The 5 types of Noise factors that disturb ideal function
Noise Factor Caused by Inner Noise 1 Piece-to-piece variation of part properties (such as Production rate (Capacity) dimensions) 2 Changes over time in dimensions or strength (such as Exposure to repetitive wear out, fatigue, deterioration, chemical, degradation) demand Outer Noise 3 Customer usage and duty cycle Conditions of use (Demand) 4 External operating environment Climatic and application conditions 5 Internal operating environment (error states from on Component & system component being received as a noise factor by interactions and another) interfaces
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Noise Impact in Bathtub-Curve
Strength Noises Failures 1. Piece to piece/comp.-comp. variation Change Over Time 2. Aging/Wear out/ strength over time / cycles Stress Strength Affected by Stress Noise Affected by Failures Occur Outer Noises Inner Noises Conditions of Use 3. Customers usage and duty cycle or DFR CFR IFR usage profile Operating environment a. infant b. useful life c. wear time 4. External (climate/location) mortality out 5. Internal subsystems /components noise #1 noises #3/4/5 noise #2 interactions / interfaces Affected by Customer Usage Variation Affected by Mfg. Variation
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Robust Design
Make design insensitive to the uncontrollable user environment (noise).
• Concentrates on:
− identifying the “ideal function(s)” for a specific technology or product/process based on its energy transformation, then selectively choosing the best levels of design parameters that optimize performance reliably (even in the presence of factors causing variability) at lowest cost.
− application of two-step optimization.
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Robustness – An Approach to Make Money
• Robustness reduces performance variations and achieves Six Sigma quality • Avoids failure modes • Achieves customer satisfaction • Also shortens development time to market – reduces build/test/fix cycles
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Robust Design for Reliability
Ideal Function Minimize Sensitivity to
Reduce Rate of change of product Output Output
Response Noise parameters Input Energy - Concept / System Design -Tolerance Design - Robust Parameter Design ROBUST DESIGN O U Less Sensitive T P U T Sensitive
Input/Signal Robust Design for Reliability
Redundancy Capable Manufacturing Process (Cost of failures vs. cost of providing redundant components
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Sensitivity Rate of change System Thinking & Robustness - Concept / System Design -Tolerance Design - Robust Parameter Design - Tolerancing
Reliability Improvement Redundancy The achievement of higher reliability can also be viewed as an (Cost of failures vs. cost of CapableManufacturing providing redundant Process improvement to Robustness. components
Design Space Optimal Points Searching
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Sensitivity Rate of change Create Robust Design - Concept / System Design -Tolerance Design - Robust Parameter Design - Tolerancing
Reliability The principle of parameter design is a powerful methodology to increase Improvement Redundancy (Cost of failures vs. cost of CapableManufacturing the distance from the failure mode. providing redundant Process components
Exploiting Non-Linearity
ROBUST DESIGN
O Y2 Less Sensitive s U Y T P U Y1 T Sensitive
X1 X2
Input/Signal sX X2 results in less variation in Y
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Robust Engineering
Robust Engineering Emphasizes CI* Robust Design Principles on Three Main Design Stages
Customer Design Space Space (STRESS) Input Signal (STRENGTH) (M)
Voice of System Satisfied Customer Uncontrollable User Output Response • System Design Conditions (y) Control Factors Customer 1. Identify and Select Proper System Output Response (s) Uncontrollable (C) Factors Product Reliability • Parameter Design Requirements 2. Measure Functions using S/N** Ratio or Equivalent
System Design • Tolerance Design Specifications 3. Take Advantage of Interactions between Control &
Subsystem Design Noise Factors Specifications
Component 4. Use Orthogonal Arrays Design Specifications 5. Apply Two-step Optimization
Upfront Critical Thinking and Discipline Note: *CI: Continuous Improvement, S/N=Signal-to-Ratio
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Model for Robustness Thinking
Noise Factors
J Input Ideal Function Signal System E.g., Energy Out E.g., Energy in L Error States Control Factors
Ideally 100% of input energy (the signal) should convert into 100% ideal function.
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 (Shrink and Shift) Sensitivity Rate of change Results of Robust Design Effort - Concept / System Design -Tolerance Design - Robust Parameter Design - Tolerancing
Reliability Improvement Results of Robust Design Efforts Redundancy (Cost of failures vs. cost of CapableManufacturing providing redundant Process • Reduced variation components • ImprovedReduced Variation targeted performance • ImprovedImproved Targeted reliability Performance Step 2 • ImprovedImproved Reliability customer satisfaction Adjust Mean Select factors to ship mean with minimum impact veracity
Target Step 1 AssessAssess the the Reduce Variability Take advantage of control factors limitationlimitation of of a a affecting variability Target Mean givengiven design design
Target Mean
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Example of Mechanical Crimped Connector
New Approach - Intent: To transfer energy from input energy to form a proper shape. What to measure to understand performance? Quality Problems: • Poor electrical conduction • Poor tensile strength • Poor vibration resistance • High voltage drop • Degraded electrical & mechanical integrity • … etc Measure of Robustness Desired Output S / N = HarmfulOutput
Results: -Pull strength increased -Voltage drop reduced Focusing on basic function, minimizes the difficulty in improving this problem - Improved process capability
Focus what you want. Don’t focus what you don’t want!
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 The Challenge of Reliability Design at The Lowest Cost
Voids & Bubbles
ASQASQ Reliability Reliability and and Risk Risk Division Division series series webinar webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Example of Mechanical Crimped Connector
New Approach - Intent: To transfer energy from input energy to form a proper shape. Quality Problems: Compactness • Poor electrical conduction • Poor tensile strength • Poor vibration resistance • High voltage drop • Degraded electrical & mechanical integrity • … etc Measure of Robustness Desired Output S / N = HarmfulOutput
Results: -Pull strength increased -Voltage drop reduced Focusing on basic function, minimizes the difficulty in improving this problem - Improved process capability
Focus what you want. Don’t focus what you don’t want!
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Example of Reliability Improvement in Robust Design (cont’d) Optimized design in presence of uncontrollable usage environment. Noise actors Outer Array Noise Factors M, Y, Compact M1=-0.2 M1=-0.2 M1=-0.2 Crimped -ness CSA Crimped System geometry Wire K1=-5% K2=-7.5%StressK1=-5% K2=-7.5% K1=-5% K2=-7.5% Aging Before After Before After Before After Before After Before After Before After Control Factors Control Factors Sensitivity Analysis Crimp Res. Angle # Material Material CH/CW Strok P.W. leg Of of Type Thinkness ratio length terminal length Press punch 1 2 The Designs 3 SN Ratio Analysis …. Strength … Inner … Array 16 17 L18 18
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Example of DfR Basic Tools Application (cont’d) Crimped Connector Reliability Demonstration Plan & Report
RDM Candidates RELIABILITY/ROBUSTNESS IMPLEMENTATION VALIDATION EVIDENCE Noise Factors Present in the Test
High Impact Noise Factors
High N1. P-to-P variation Impact Related Validation N2. Wear-out / Aging Metric Range Risk Key Function Failure Component Critical Test Demonstrated Validation Evidence & Completion Test from N3. Customer Duty Cycle Assess- /Technology Modes Subsystem Metric Target Comparison to Prior Design Date RCL N4. External Environment ment (Soft / / System N5. System Interactions Hard) Example: Withstand Wires Terminal Pull N 150 N1. Different Wire Size mm CSA% Force Disconne and Wire strength N2. Aging Y/N Y/N
(Pull cted strands Life Test Probability Plot for Optimal_AA, B2_AA Weibull - 95% CI Strength) Complete Data - LSXY Estimates
99 Variable 90 Optimal_AA 80 B2_AA 70 Table of Statistics 60 Shape Scale C orr F C 50 21.1703 222.041 0.978 14 0 40 18.4557 148.370 0.976 12 0
t 30 n
e 4-Jul-99
c 20
r e P 10
5
3 2
1 100 120 140 160 180 200 220 240 260 Pull Force (N)
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 The Essentials of Robust Design
Why Robust Design?
Efficiency And Effectiveness
ASQASQ Reliability Reliability and and Risk Risk Division Division series series webinar webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Sensitivity Rate of change - Concept / System Design -Tolerance Design Robustness "Rules of Engagement" - Robust Parameter Design - Tolerancing Reliability Improvement Redundancy (Cost of failures vs. cost of CapableManufacturing providing redundant Process components
1. Concentrate on Ideal Function, and establish a 7. Work out how to include remaining Noise way to measure it; do not use symptoms of poor quality Factors in tests 2. Identify sources of the five types of noise and 8. Plan a robustness assessment of current design to expected magnitudes compare against ideal performance. 3. Introduce product noise early. Drive the performance 9. Where robustness improvement strategy is obvious away from ideal situation from knowledge of physics, DO IT! 4. Concentrate on the effects of the noise, rather than the 10. Where robustness improvement is not obvious from noise themselves current knowledge of the physics, plan parameter 5. Understand how error states and noise factors cross design studies (using DoE if necessary) to discover system interfaces and boundaries the improvement 6. Develop a noise factor management strategy 11. Management needs to design this into the Product Design Process and check that it is done to an appropriate degree
ASQASQ Reliability Reliability and and Risk Risk Division Division series series webinar webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Case Study- Robustness Thinking In Innovative Problem Solving
IDDOV
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 System Robustness and Redundance
An important result of our Comprehensive Risk Management and Deep Integration process is systems diversity, robustness and redundancy, which are key drivers of the safety of the Alphaba Bus.
System Robustness All critical systems have been designed, optimized in Integrated Vehicle Health Monitor the presence of user conditions, tested and validated through intrusive testing, test track durability testing and extensive on-road mileage accumulation.
System Robustness
Steering and Braking Signal Communications between computers, sensors and actuators have an alternate path if Electrical Power the primary fails.
Vehicle Localization Redundant Collision Signal Communications
Redundant Collision Detection Redundant Collision Detection our vehicle includes a crash-imminent braking system calibrated to work as a backup to the self-driving system that can apply the brakes to Perception Sensors stop the car if necessary.
ASQ Reliability and Risk Division series webinar 3/11/2021 Matthew Hu, Senior Vice President 13681699148 (China); 281-299-4230 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 (USA) Important Takeaway
• Robust design is essential to achieve competitive advantage • Make design insensitive to uncontrollable user environment (Noise) • Early development of robustness is key to proactive quality and reliability Improvement – Capture, front load noise and manage noise – Gain control of your product performance – Optimize robustness – avoid all failure modes • Apply Robust design principles at early stages of product design to “forecast” problems and take preventive action.
ASQASQ Reliability Reliability and and Risk Risk Division Division series series webinar webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Questions?
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230 Thank You!
Matthew Hu, PhD Haylion Technologies Phone: 281-299-4230 Email: [email protected]
ASQ Reliability and Risk Division series webinar 3/11/2021 Dr. Matthew Hu, [email protected]; Phone: 281-299-4230