Six Sigma Awareness Tool Kit

6 Overview Greenbelt Training 1 Concept of Competitive Quality Historically, Quality has been defined as “ Meeting the Specifications” “ Fitness for Use” Today the most acceptable definition of Quality is “ Meeting the Customer’s Requirements” 2 Concept of Competitive Quality Will just meeting the requirements ensure future Market share ? NO ! 3 Concept of Competitive Quality The definition of Competitive Quality is : “ Product, Process or Service Uniformity around a Target Value” 4 The most noticeable difference in this operational definition of Quality is that it requires CONTINUOUS IMPROVEMENT Uniformity Around A Target Value WHY CHANGE ? The reason is simple, to remain or become competitive and thereby increase Profitability and Market Share 6 The Impact of Added Inspection If the likelihood of detecting the defect is 70% and we have 10 consecutive inspectors with this level of capability, we would expect about 6 escaping defects out of every 1,000,000 defects produced. 1,000,000 ppm 6 ppm 3.4 ppm 7 You can save your self by producing quality not by Inspection Sigma - The Standard Deviation 1 Sigma - 68% 2 Sigma - 95% 3 Sigma - 99.73 % m = S (X – X)2 n 1 Upper Specification Limit (USL) Target Specification (T) Lower Specification Limit (LSL) Mean of the distribution (m) Standard Deviation of the distribution (s) p(d) USL T 3 8 What is Six Sigma Before A 3 process because 3 standard deviations fit between target and spec Target Customer Specification 3 1 2 3 0.27% Defects up-to 6.6 % Target Customer Specification After 1 2 3 4 5 6 6 ! No Defects! 9 Reducing Variability Is The Key To Six Sigma Philosophy of Six Sigma Six Sigma focuses on :  Continuous Improvement of Processes Defect Prevention through the use of Statistical tools as opposed to Defect Detection through inspection. 10 The Many Facets of Six Sigma • Focus and commitment to quality must be driven by top leadership • Leadership must be fully engaged and accountable for success Leadership Process • Focus is on statistical process capability and process variation analysis • Products must be designed to be manufactured within process capability • Process capability must be quantified 6 • Six Sigma provides classical problem solving tools enhanced with a fundamental knowledge of statistics and variation Tools Metrics • Goal is defect free products and processes • Focus is on true capability (rolled throughput yield) rather than end of line yield • All decisions must be data driven 11 How Good is Six Sigma 99% Good (3.8 Sigma) 20,000 lost articles of mail per hour Unsafe drinking water for almost 15 minutes each day 5,000 incorrect surgical operations per week 99.99966% Good (6 Sigma) Seven articles lost per hour One unsafe minute every seven months 1.7 incorrect operations per week Two short or long landings at most major airports each day 200,000 wrong drug prescriptions each year No electricity for almost seven hours each month One short or long landing every five years 68 wrong prescriptions per year One hour without electricity every 34 years 12 What does Six Sigma means to [Company Name]…… Top Line Growth - satisfied customers are repeat customers Bottom Line Growth - it costs less to do it right the first time 13 Industry Overview The Six Sigma Define-Measure-Analyze-Improve-Control (DMAIC) model consistently yields rapid, measurable, benefits Avery Dennison Results: • Implemented 6s in March, 1998. • 52 completed projects Forecast „99 savings = $18,500,000 Actual savings (through 2Q)=$5,800,000 • 130 projects planned or in progress Annualized savings = $35M „99 impact = $12.3M AlliedSignal Results: • Implemented 6 1994 for Operations. in • Initial 4 months, 600 projects reduced defects by 68%. • Saved $175M at bottom line in „95; $500M in „98 ! (not including overhead, inventory, indirect charges, or avoidance. • Over $2B in savings realized since 1992. • Fastest rate for implementing 6 yet! GE Results: • Implemented 6 efforts in late 1995. • Targets savings over $10 B during next few years by reducing its current COPQ ($7B /yr.) to less than $1B annually by: - Reducing scrap parts. - Reducing reworked parts. - Rectifying transaction mistakes. Motorola Results: • Implemented 6 program in 1987 when it was performing at a 4 level. • By 1992 it averaged a 5.21 level. • Sales productivity increased from $68.9K to $110.1K per employee and savings due to US operations improvements were over $2.2 Billion. 14 Cost Opportunity Cost of Failure (% Revenue) 40% 35% 30% 25% 20% 15% 10% 5% If [Company Name] is a 3 Company, Cost of Failure is Estimated to be at Least 15% of Revenue Defects per Million 3.4 Sigma 233 5 6210 66,807 3 2 308,537 1 691,462 6 4 15 A $4.5 Million Cost Reduction Opportunity! Six Sigma Saving in GE Six Sigma Cost and Benefits 2500 2000 1500 2000 Cost Benefits Net 1500 1050 1450 $ Million 1000 500 0 1996 -500 200 380 170 -30 700 320 450 550 1997 1998 1999 Mostly variable cost productivity and asset utilization 16 Up front investment and staying power Significant impact on the bottom line How much Cost Reduction is possible Traditional Quality Costs (Easily Identified) Inspection Warranty Rejects (tangible) Scrap Rework 1.5 % COQ Hidden Quality Costs (Difficult to measure) Lost sales Customer Sat Long cycle times Field Modifications (intangible) Overtime Late delivery T&L 15 % COQ More Setups Expediting costs Customer Productivity Loss Engineering change orders Lost Opportunity Excess inventory Lengthy Installs Lost Customer Loyalty Employee Morale, Productivity, Turnover 17 . . . Six Sigma Reveals hidden facts and capabilities Six Sigma as a Goal Distribution Shifted ± 1.5s  2 3 4 5 PPM 308,537 66,807 6,210 233 3.4 Defects Per Million Opportunities 6 18 Process Capability Harvesting the Fruit of Six Sigma Sweet Fruit Design for Manufacturability 5 s Wall - Must Address Designs Bulk of Fruit Process Characterization and Optimization ---------------------------------- 4 s Wall - Must Improve Internally Low Hanging Fruit Seven Basic Tools ---------------------------------- 3 s Wall - Demand improvement Ground Fruit Logic and Intuition The walls crumble faster when addressing process issues 19 Attacking the Problem Practical Problem Rejects for bad estimation of cost =20% average Statistical Problem • Process characterization data set is non-normal. • After normalization: ST = 5.50 LT = 2.36 • DOE Results: - Technology - Labour rate - Interaction Statistical Solution - 52% - 24% - 19% Practical Solution 20 • Install standard measurement system for each technology • Reward & recognition policy to retain experienced labours in order to increase productivity The Focus of Six Sigma Y       Y Dependent Output Effect Symptom Monitor       f(X) X1 . . . Xn Independent Input-Process Cause Problem Control 21 Would you control shooter or target to get the Gold Medal at Olympics Controlling the Output Y OUTPUT SIGNAL = F (x) IN-PROCESS PARAMETERS RELATIONSHIP or EQUATION THAT EXPLAINS Y IN TERMS OF X Distance traveled Money to Spend Determined by Determined by Car speed, traveling time Income, Commitments, Credit Rating OUTPUT (Y) IS DETERMINED BY THE VALUES OF THE IN-PROCESS PARAMETERS (X‟s) 22 Controlling the Output Y OUTPUT SIGNAL = F (x) IN-PROCESS PARAMETERS RELATIONSHIP or EQUATION THAT EXPLAINS Y IN TERMS OF X Distance traveled Determined by Understanding the 23 F (x)S gives insight into the right CarCar speed, traveling time Speed Amount of wear on brakes Selection of CDs available Amount of gas in the tank Time since last service Traveling time Number of passengers Weather Car inside temperature Controlling the Output Y OUTPUT SIGNAL, = Y F UNDERSTANDING OF (x) SELECT IMPORTANT F (x)S VERIFY UNDERSTANDING OF F QUANTIFY WITH CORRELATION How we’ve been taught to search for F 24 Controlling the Output Y OUTPUT SIGNAL, = POOR OR NO F UNDERSTANDING OF (x) BRAINSTORM F (x)S Y NO CORRELATION Thousands of (x)s to choose from....... Without an understanding of F - it‟s your opinion vs mine ! 25 Model Y Capturing the measurement on a Customer unit basis = Understanding our output as the Customer sees it F (x) F Verify Correlation to find Learning from variance in performance on Customer Y Unlocking the process keys that control Customer impact 26 What are these opportunities….. Cost of Quality SPAN Customer Satisfaction 27 Cost of Quality Sigma Level 2 3 4 5 Defects/Million Opportunities Cost of Quality Not Competitive 25 - 40% of Sales 15 - 25% of Sales 5 - 15% of Sales 308,000 66,800 ($12b) 6210 ($7b) Industry Average 233 ($4.5b) 3.4($0.3b) World Class 6 28 <1% of Sales SPAN Order-to-Delivery Time vs Customer Want Date 40 Day Span 5% of orders are >15 days early to request 5% of orders are >25 days late to request -15 days Early 0 Customer Want Date On Time +25 days Late 29 Difference between Mean & variance Average River Depth - 4ft Focus on Average can turn your business red 30 Outside In Thinking Delivery cycle time (days) Insight Through Variance What WE see Baseline 12 24 13 7 16 8 20 25 14 10 11 30 16 Mean 15.8 Std Dev 7.0 Improved? 27 7 15 4 18 6 23 6 2 24 2 6 5 11.2 9.0 11.2 15.8 What customers feel • Using mean-based thinking, we improve average performance by 29%, and we break out the champagne ... But our customer only feels the variance and cancels the next order! • 31 Customers feel variance, not the mean The Eye of the Beholder Customer’s View How did [Company Name] influence my AC Performance? Missing data during download Customer Process A B C [Company Name] Process How did I do against my AB Obligations? Missing data during logical execution [Company Name]’s View 32 Where in [Company Name]… ... Can Be Applied To Every Business Function Business Development Operations Training HR CRT 6 Sigma Methods IT Finance Admin / Transport Projects Quality 33 Why Now? Customers & Competitors are adopting elements of this business improvement process: – Customers: • HP • Intuit – Competitors: • Wipro Spectramind • EXL • Hughes 34 Driven by Customer Excellence at Lowest Cost Table of Contents Six Sigma Overview What is Six Sigma Overview of Scope Linkage to Vision Pre-Tea Roles and Responsibilities as Leaders/ Sponsor Pre-Tea Criteria for Project Selection Criteria for BB/GB Selection Introduction to Process Pre-Tea Post-Tea Post- Tea 35 Implementation Strategy Train…. Apply….. Review….. Every Participant arrives to training with a well defined project with measurable savings opportunities! Integrate training with metrics performance to maximize the bottom line impact. 36 Six Sigma Program Structure Define Measure Analyze Improve Program Direction, Support, and Marketing Management Leadership Black Belt Process Owner Control Change Agents and Process Leaders Green Belt Organizational “Buy-in” Program is structured to build a self-sustaining critical mass of process improvement competencies. 37 Champion / Functional Leader Role 1. Lead the Six Sigma efforts overall in their BU 2. 3. 4. 5. 6. 7. 8. 9. Provide Strategic Direction for Six Sigma Project teams Track the Project’s Progress, Offer rewards as appropriate Help the Black Belt / Green Belt overcome roadblocks, including seeking collaboration Help find resources for the team as Needed, Allocate resources when authorized Keep Black Belt / Green Belt focused on desired results When immovable objects block the road, Redirect Project / Team activities Serve as the Team’s Champion from Top-To-Bottom of Entire Business Ensure that Project Solutions are well implemented, Gains are sustained and on-going responsibility transfers to Process Owner 38 Six Sigma Roles • Champion/Sponsor/Functional Leader – The Champion Or Sponsor Is The Person(s) Who Is Accountable For And Sanctions A Six Sigma Project. The Champion Or Sponsor Is Involved In Project Team Chartering, Reviews Progress, Helps Remove Organizational Barriers To Project Success, And Is Often The Decision-Maker For Approval Of Final Recommendations • Master Black Belt – Full-Time Positions Dedicated To Supporting Six Sigma Efforts. “Expert” Resources To Black Belts And Teams On The Six Sigma Tools And Techniques Coach And Assist Black Belts And Team Members. Train The Black Belts, Champions, And Employees As Needed. • Black Belt (Team Leader) – Full-time position where the Black Belt Is Accountable, Usually To The Champion, For The Project / Team Results. The Black Belt Is Responsible For The Project / Team’s Progress, Provides Leadership In Planning The Project / Team’s Work, Applies Six Sigma Tools And Teaches Team Members How To Apply Them, Often Leads Team Meetings, And Ensures That Decisions Are Made By The Team In A Timely Manner To Meet Its’ Goals • Green Belt (Team Leader) – Part-time position where the Green Belt Is Accountable, Usually To The Champion, For The Project / Team Results. The Green Belt Is Responsible For The Project / Team’s Progress, Provides Leadership In Planning The Project / Team’s Work, Applies Six Sigma Tools And Teaches Team Members How To Apply Them, Often Leads Team Meetings, And Ensures That Decisions Are Made By The Team In A Timely Manner To Meet Its’ Goals • Team Members – Team Members Are The Individuals Who Comprise The Six Sigma Project Team. Team Members Are Individually And Collectively Accountable For Specific Tasks That Will Result In The Team’s Final Recommendation. When Team Members Are Responsible For A Particular Aspect Of A Project, They Often Will Make Their Own Decisions 39 BQC – Roles & Responsibility Six Sigma Steering Committee - BQC – Assures [Company Name]’s Six Sigma implementation plan: • Has appropriate resources allocated • Has appropriate scope and is involving all required elements of the organization • Is consistent with [Company Name]’s culture of Exceeding customer expectation – Develop a communication plan to energize the organization around the Six Sigma implementation. – Remove roadblocks to facilitate implementation. – Review the status of training / project implementation on a regular basis. – Establish [Company Name]’s business metrics and goals, assess progress towards goals, analyze strengths and weaknesses of implementation, and provide strategic direction as necessary. – Verify the financial (as calculated by Finance) impact of projects implemented. – Provide a forum to share best practices within the organization. 40 Table of Contents Six Sigma Overview What is Six Sigma Overview of Scope Linkage to Vision Pre-Tea Roles and Responsibilities as Leaders/ Sponsor Pre-Tea Criteria for Project Selection Criteria for BB/GB Selection Introduction to Process Pre-Tea Post-Tea Post- Tea 41 Customer Focus Start With The Customer 1. 2. Measure the same as the customer does Determine your capability as the customer sees it 3. 4. Understand the variance in the output signal Find the in-process keys to impact the customer 42 Customer Focus What I Want to be What I Am (?) What I Am(?) What I Am(?) What the Customer Wants Exceeding Customer Expectations Competition ? Unhappy Customers Status Quo Delighted Customers Performance Continuum 43 How Important Is This GAP? Project Selection Objective: 1. What exactly is the problem being addressed in measurable terms? Need: 2. Why is the project worth doing? 3. Is the project tied to a high importance customer CTQ? 4. What are the consequences of not doing this project? 5. How would it fit with business initiatives and targets? Scope: 6. Is the scope reasonable? Can the problem be effectively broken apart into projects with reasonable scope? Expectations: 7. Is there a clear owner in the organization for the problem and for the benefits of improving? 8. What specific goals would project be trying to achieve? What would constitute stretch results? 44 Project Objectives What exactly is the problem being addressed in measurable terms? PROBLEM The problem this project is going to solve is: - Takes too long to submit the quotation to customer - The productivity is very poor in our company - There is too much documentation in our work MEASURABLE In measurable terms that means: - Improve time of submission of quotation from the time enquiry comes from customer - Improve productivity by 15 % - Reduce the paperwork by 20% 45 Project CTQs (Critical to Quality) Who are your customers? What do you provide your customers? What is critical to quality for your customers? What are your internal processes for providing your product or service to customers? What CTQ is this project addressing? 46 Project Need Why is this project worth doing? Customer [Company Name] What activities have higher or equal priority? Why is it important to do now? Customer [Company Name] How does it fit with the business initiatives and targets? What are the consequences of not doing this project? Customer 47 [Company Name] Project Need Write down threat and opportunity for short term and long term for the problem you are addressing in your project Threat Short Term 1 2 Long Term 3 4 Opportunity 48 Expectations What specific goals must be met? When must they be met? For each goal, what milestones are critical and must be met? What would constitute stretch results? 49 Project Scoping 50 Project Scoping Why? High level problem Level A Initial Contributor Level B How far down should I scope my project? Why? Why? Secondary Contributor Level C Why? Project Level Level D Project Level ??? When you can no longer answer the “Why?” with confidence, you have arrived at the project level. 51 Project Scoping Why? AHT too high Why? AHT for Pavilion P.L. is too high Why? AHT for Team 11 is too high AHT for New Agents is too high Level A Level B Why? ??? Project Level Level C Level D Our project then becomes in measurable terms: Improving the AHT for New Batches from 30 min. to 20 min. 52 Project Scoping What must this project accomplish? What (if anything) is out of bounds for the team? What resources are available to the team? What (if any) constraints must the team work under? 53 Criteria for BB / GB Selection  Business Acumen  Project and Process Management  Data Affinity  Result Orientation  Relationship Building and Influence  Coaching and Mentoring  Team Leadership  Change Leadership 54 Curriculum for Green Belt  Receive 5 days training  Understand the statistical tools and practice them  Work on the project  Monthly presentation to the Project Sponsor / MBB  Close the project  Clear Green Belt Certification Test 55 Table of Contents Six Sigma Overview What is Six Sigma Overview of Scope Linkage to Vision Pre-Tea Roles and Responsibilities as Leaders/ Sponsor Pre-Tea Criteria for Project Selection Criteria for BB/GB Selection Introduction to Process Pre-Tea Post-Tea Post- Tea 56 DFSS (DMADOV) vs DMAIC DEFINE IDENTIFY NO PROCESS ? YES MEASURE ANALYSE DESIGN NO CAPABLE ? YES OPTIMISE IMPROVE VALIDATE CONTROL 57 DMAIC Process Define the project (Y) & make the team & plan Measure Y & X’s What’s wrong with X’s How much and what I can improve Control Sustain the improvement Define Measure Analyze Improve PLAN 58 DO CHECK & ACT Six Sigma Tools Used………. • Project Scoping • SIPOC • Thought Process Mapping • Quality Metrices • Process Mapping • C&E matrix • FMEA • MSA • Concepts of DOE • DOE Strategies & Analysis • DOE • Control Strategies 59 Measurement Purpose Document Process Map Begin To Link CTQs to Input Variables Establish Measurement Capabilities Establish Baseline Process Capabilities     60 The Funnel Effect Process Map +30 Inputs 10-15 All X‟s C&E Matrix and FMEA Multi-Vari Studies 1st “Hit List” 8 - 10 Experimentation Control Plans / SPC 4-8 3-6 Screened List Found Critical X‟s Controlling Critical X‟s Optimized Process 61 Input, Output & Process Measures Input Measures Process Measures Measures That Are Internal To Your Process. They Include Quality And Delivery Measures Important To Your Internal Customers As Well As Waste And Cycle Time Measures. They Are Correlated To The Pertinent Output Measures. Output Measures Output Measures Are Measures Used To Determine How Well Customer Needs And Requirements Are Met. The Key Quality And Delivery Requirements Placed On Your Suppliers. 62 Steps to Business Process Mapping Develop A Picture Of The Working Process As A Team Suppliers Inputs Start Process Outputs Customers Requirement 64 Define the Boundaries of Business Process START Boundary Boundary Input What Must My Suppliers Provide My Process To Meet My Needs? Process Output How Can I Assure That My Process Output Meets The Needs Of My Customer? 65 Process Map [Company Name] What You Think It Is... What It Really Is... What It Should Be... What It Could Be... 66 Industry Overview Traditional View Final Test “The Hidden Factory” RTY is the probability that a product will pass through the entire process without rework and without any defects. It is the true yield for a product at the completion of all the individual processes. Six sigma View Value Stream optimization is enabled by elimination of the hidden factory. 67 Rolled Throughput Yield Develop A Better Understanding of Your Operations To Know Where To Begin If this is your process, where do you put your key resources ? A B 0.90 $10 / Unit C 0.90 $ 5 / Unit 400 un/dy D 0.90 RTY COPQ 0.80 $2 / Unit 0.583 $19 $2 /Unit 200 un/dy Capacity 68 700 un/dy 500 un/dy 200 un –Rolled Throughput Yield (RTY) -- A true estimate of process yield Project Prioritization A 0.80 $2 / Unit 700 un/dy B 0.90 $10 / Unit 500 un/dy C 0.90 $ 5 / Unit 400 un/dy D 0.90 $2 /Unit 200 un/dy 0.583 RTY COPQ $19 200 un Capacity Project # A Project # B Project # C 69 Role of Statistics Can you always measure …100% or less What is Population ……what is sample? Roll of statistics in measurement (descriptive / Inference) 1. We only use experience, not data. 2. We collect data, but just look at the numbers. 3. We group the data so as to form charts and graphs. 4. We use census data with descriptive statistics. 5. We use sample data with descriptive statistics. 6. We use sample data with inferential statistics. 70 Basic Statistics  Types of data  Measures of the Center of the data  Mean  Median  Mode  Measures of the Spread of Data  Range  Variance  Standard Deviation  Normal Distribution and Normal Probabilities 71 Measures of Central Tendency  Mean: Arithmetic average of a set of values  Reflects the influence of all values  Strongly Influenced by extreme values x   xi n i 1 n   Median: Reflects the 50% rank - the center number after a set of numbers has been sorted from low to high.  Does not include all values in calculation  Is “robust” to extreme scores The mean and median will be affected by the nature of the distribution of numbers Mode - Most Common Observation Why would we use the mean instead of the median in process Improvement?   72 Different Distributions Sketch in the Means and Medians on each Distribution. Comparison of Distributions. 300 300 Comparison of Distributions. Frequency Frequency 200 200 100 Tail 0 10 20 30 40 50 60 70 80 100 Tail 60 70 80 90 100 110 120 130 0 0 C2 C3 Negative Skew Positive Skew Comparison of Distributions. 100 Frequency 50 0 20 30 40 50 60 70 80 90 100 110 C1 Symmetric Distribution 73 Population Parameters vs Sample Statistics Examples of POPULATION: Examples of SAMPLE: 500 people randomly selected Entire India Average Literacy rate  = Population Standard Deviation m = Population X Mean = Sample Mean ^  = Sample Standard Deviation 74 Computational Equations Population Mean m = X i 1 N i N Population Standard Deviation  =S= (Xi  m ) 2  i=1 N N Sample Mean ^ m =x= N x i =1 n i n i -- Sample Standard Deviation ^  = s=  (X i =1 X )2 75 n -1 Measures of Variability     Range: the distance between the extreme values of a data set. (Highest - Lowest) Variance (  ): the Average Squared Deviation of each data point from the Mean. Standard Deviation (  ): the Square Root of the Variance. The range is more sensitive to outliers than the variance. 76 Calculating Standard Deviation X 1 2 3 4 5 6 7 8 9 10 S Mean  square  X - X (X - X) 2 Variance 2 1 3 5 4 (X i  X ) 2  i=1 N N -1  (X i=1 N i  X )2 Sum N -1 Standard Deviation 77 1.581139 Types of Data   Attribute / Discrete Data (Qualitative)  Categories  Yes, No  Go, No go  Operator 1, Operator 2, Operator 3  Pass / Fail Variable / Continuous Data (Quantitative)  Decimal subdivisions are meaningful  Time, Pressure, Conveyor Speed 78 Variation    “While every process displays Variation, some processes display controlled variation, while other processes display uncontrolled variation (Walter Shewhart). Controlled Variation is characterized by a stable and consistent pattern of variation over time. Associated with Common Causes. Uncontrolled Variation is characterized by variation that changes over time. Associated with Special Causes.   Process A shows controlled variation. Process B shows uncontrolled variation X-Bar C hart for Proc es s A UCL=77.20 80 X-B ar C hart for P roc es s B UCL=77.27 X =70.98 LCL=64.70 60 Sample Mean 75 Sample Mean 70 X =70.91 70 65 0 5 10 15 20 25 LCL=64.62 50 0 5 10 15 20 25 Sample Number Sample Number 79 Special Causes Stratification Customer Type Geography Etc. Company Process ALL DATA n = 2000 0 (-11, 38) 49 Sample Size , Median (Min, Max) SPAN The most powerful potential process labels are those that are known at the beginning of the process. 80 Stratification ALL DATA n = 2000 0 (-11, 38) 49 Dashboard n = 899 -2 (-9, 21) 30 Non-Dashboard n = 1101 1 (-12, 70) 82 North n = 261 -2 (-8, 8) 16 South n = 297 -2 (-10, 24) 34 East n = 103 -1 (-8, 15) 23 West n = 238 -1 (-8, 23) 32 Commercial n = 119 -2 (-10, 5) 15 Government n = 74 -2 (-8, 9) 17 Industrial n = 68 -2 (-7, 40) 47 Even with small sets of Data, the median difference appears. Credit A n = 71 -3 (-10, 0) 10 Credit B n = 41 0 (-7, 6) 13 Credit C n=7 5 (-8, 31) 39 81 Stratification Key Learning Points: • The first thing you must do is Separate the Processes. We call this Stratification. If you don‟t Stratify (isolate) the processes, you will have more than one central tendency in the data set and you will never figure out what drives variance. • If you think you have found the right label to stratify the processes, make sure you double check it to see if there is another label that is influencing the way the data appears. In this case, the real process label was Credit Rating, but it appeared in the Dashboard/Non-Dashboard data. You can double check by cross-cutting the data (look at Credit Rating and Dashboard at the same time in a tabular format), or by continuing to segment to see if the central tendency indicator still moves even though we thought it was an isolated process. In this case, if you continue with Dashboard as an isolated process, you will see the median move for various segmentations (especially Credit Rating). • Once you have Stratified (isolated the processes) and you have a segment that reflects several different levels of Variance, you have the first clues to find the critical x‟s that drive variance. • When you find a critical x for one of the processes, check to see if it is also the critical x for the other processes. Often the factors that drive variance in one of the processes, also drive it in another. 82 FMEA Model Prevention What made failure mode to take place. Ask 5 Why’s…. Detection Detection What manner my process was not able to obey me Cause Material or process input Failure Mode (Defect) Process Step Effect External customer or downstream process step. Controls Because of your process what all I will not be able to do 83 Measurement System Analysis A measurement system will not willingly disclose the type of distortion, inaccuracy or imprecision it is transmitting to our data. We must actively force it to reveal its hidden effects. CAUTION: Objects in mirror are closer than they appear 84 Measurement System Analysis Parts (Example) Inputs Process Outputs Inputs Measurement Process Outputs • Observations • Measurements • Data Product Variability (Actual variability) Measurement Variability Total Variability (Observed variability) Measurement System Variability Investigated through “R&R Study” 2 Actual(Part) + 2  Meas.System = 2  Observed (Total) 85 The Measurement System will transmit variation to our data. Establishing the Process Capability LSL USL Short-Term Capability Long-Term Capability 86 Over time, a process tends to shift by approximately 1.5 . Visualizing the Causes Within Group • Called  short term (st) • Our potential - the best we can be Time 1 Time 2 Time 3 Time 4 • The  reported by all 6 sigma companies • The trivial many st + shift = total Between Groups • Called  shift (truly a measurement in sigma's of how far the mean has shifted) • Indicates our process control • The vital few 87 Analyse Purpose  To reduce the number of Process Input Variables to a manageable number To determine high risk inputs from Failure Modes and Effects analysis To determine the presence of and eliminate Noise Variables through Multi-vari Studies    To plan the first improvement activities 88 Sources of Variation A common method of analysis at this stage is the variables tree. Try thinking about your process in this manner........ Customer Service Example Not resolved the call Skill to Skill Agent to Agent Call to Call Customer to Customer Type of call 89 Tools Used…….. Time Series Plot 15 Scatter Plot 15 HrsVar 10 HrsVar 5 10 10 5 5 0 Index 0 May Jun Jul Date ANOVA (Analysis of Variance) Main Effects Plot - Means for HrsVar 13. 0 10. 5 HrsVar 8. 0 5. 5 3. 0 D at e Cu s t o m er Sal es m an Box Plot 15 1000 Pareto Chart Pareto Chart for : Defects 100 80 900 800 700 600 500 400 300 200 100 20 40 60 rsVar H 10 5 0 Water Util Mining Paper Count 0 0 . ev tD i gh We rB Ai bl e ub l or Co f or De n t io ma Defect Count Percent Cum % C us tom er 431 44.2 44.2 293 30.0 74.2 1 32 1 3.5 87.7 1 20 1 2.3 1 00.0 90 Percent Regression….. Some examples: Y=Gas Mileage 30 (mpg) 20 10 Y=Son’s Height 80 60 40 0 .5 1 1.5 2 X=Car Weight (tons) Y=Grades (of 100%) Y=Selling Price 80 60 40 60 70 80 X=Dad’s Height (inches) (Thousands) 35 25 5 0 .5 1 1.5 2 X=Study Time(Hours/Night) 1 6 14 22 30 X=Age of Car 91 Regression….. How do you find a line that “fits” the data? What we are looking for is a line which will minimize the distances from the plotted points to the line.... Deviations (distances) “How much the line missed by” * * * Regression Line Response Variable (Y) * * * * * * Scatter Plot Points (actual data values) * 92 Input Variable (X) Regression….. The R2 Statistic is defined as the sum of squares of errors divided by sum of the square of difference from average: Y Measured Error Measured Predicted Y=a+bX r  1 2 ^  (yi  yi ) n 2  (y  y ) i 1 i i 1 n 2 X 93 Improve What will you do for Improve – – Identify solutions. Develop change management plan. – – Conduct cost / benefit analysis. Create implementation plan. 94 Improve What will you do for Control – – Define and implement ongoing measurement / monitoring plan. Document procedures. 95 Control In the physical world, the law of entropy explains the gradual loss of order in a system. The same law applies to business processes. Unless we add “energy” (in the form of documentation and ongoing process controls); processes will tend to degrade overtime, losing the gains achieved by design and improvement activities. Target The quality plan is the structure through which we add this “energy” to business processes. g 96 Control Three Main Control Mechanism…….. Avoid Potential Problems Control Potential Problems Risk Management SPC Mistake Proofing 97 Project Sign-off Answer the following questions before the project is signed off: • • • • What can go wrong and derail improvements ? What controls are in place ? Can you show me your closure plan ? What happen when the people change ? • • • • Are there any follow up on projects ? Is all documentation completed ? Is the savings verified by finance ? Is the audit plan in place ? 98 Project Sign-off Finalize Financial Results •Calculate tangible benefits •Determine implementation costs •Calculate net financial gain •Calculate the intangible benefits e.g. cost avoidance, customer retention Tangible Benefits - Implementation Costs = Net Financial gains (Over one financial year) Bank 99 Documentation Complete Documentation Package ……..Compile and organize a record of the key aspects of your six sigma project Typical Elements of the Documentation Package •A description of the project •Problem statement & business case •A list of CTQs + Xs •Hypothesis tests •Process capability analysis •Control parameters •Audit Plan/ owner •Financial results •Operational metrics •Lessons learned and best practices Project to be signed off by GB/BB, MBB, Financial controller, Process Owner, Champion. 100 Reward & Recognition ?... 101