Introduction to Six Sigma (Basic)
Download as PPTX, PDF5 likes453 views
This document provides a comprehensive overview of Six Sigma and its foundational principles, including statistics, data types, sampling methods, and quality control tools. It explains concepts such as population and sampling, confidence intervals, and standard deviation, alongside Six Sigma methodologies like DMAIC and DMADV. Through detailed examples, it illustrates the importance of reducing process variability and achieving high-quality outputs in organizational practices.
Introduction to Six Sigma (Basic)
- 2. Let’s learn something about… Six Sigma
- 3. This team is Awesome! Do you agree??? GREAT!!!
- 4. Now before we jump to Six Sigma…
- 5. How many of us still remember the basics of STATSTICS???
- 6. Do you think it’s IMPORTANT?
- 7. AWESOME!!!
- 8. Let’s discuss - • Basics of Statistics • Data & It’s Types • Population & Sampling • Confidence Level & Interval • Average Mean, Median & Mode • Data Presentation • Variance & Standard Deviation • Six Sigma – DMAIC & DMADV • DMAIC – an introduction • Common Quality Tools
- 9. Statistics - The science of collecting, organizing, presenting, analyzing & interpreting Numeric data to assist in making more effective decisions.
- 10. Statistics - DESCRIPTIVE – To summarize & describe ‘data’ or ‘information’ collected through an experiment, a survey or historical record. INFERENTIAL – Techniques by which decision are made only on a sample observed. Probability, hypothesis testing, correlation & regression analysis are being used.
- 11. Data - Collection of RAW figures like numbers, symbols, characters, images, audio, video etc. representing an information. It gives the status of the past activities and enables us to make decisions. Data must be interpreted by a human or machine to derive its meaning.
- 12. Data - QUALITATIVE DATA – NOMINAL (with no inherent order) ORDINAL (with an ordered series) BINARY (with two options) QUANTITATIVE DATA – DISCRETE (Counted) CONTINUOUS (Measured)
- 13. • Population is the collection of all individuals or items under consideration in a statistical study • Sample is that part of the population from which information is collected. • Sample is a subset of Population. • Sample should be a true representative of population. • Sample can inform us about the population Population & Sample
- 14. 2 Types of Sampling frame – • Probability Sample • Simple Random Sampling • Systematic Sampling • Stratified Sampling etc. • Non Probability Sample • Convenience Sampling • Quota Sampling • Purposive Sampling etc. Sampling
- 15. Probability Sample Systematic Sampling Simple Random Sampling Stratified Sampling
- 16. Non Probability Sample Quota Sampling Convenience Sampling Purposive Sampling
- 17. Confidence Level– Probability that the value of statistical parameter (eg arithmetic mean) for any sample is also TRUE for the Population A survey asked 2,000 Indians over 18 years, years, whether they were in favor of the smoking the smoking ban in restaurants. Overall, 75% of Overall, 75% of the respondents answered 'yes'. answered 'yes'. The confidence level for the for the survey had been set at 95%, the margin of the margin of error was set to 2%. It means that if the survey is conducted again, there is a 95% probability to get the same result. Confidence Level & Confidence Interval Confidence Interval– Also called as Margin Of Error (+/-) Probability that the value of statistical parameter (eg arithmetic mean) of a sample is between the margin Interval for the entire Population A survey asked 2,000 Indians over 18 years, whether they were in favor of the smoking ban in restaurants. Overall, 75% of the respondents answered 'yes'. The confidence level for the survey had been set at 95%, the margin of error was set to 2%. It means that if the responses for the entire population will be between 73% (75%-2%) and 77% (75%+2%).
- 18. • Equal size of each class into range of a variable is divided. • Used to represent data through Frequency Tables, Graphs, Histograms, Bell Curves etc. Data Presentation – CLASS INTERVAL Class Interval (Scores) Frequency (occurrence) Cumulative Frequency 0-20 21-40 41-60 61-80 81-100
- 19. • Number of occurrence • To summarize large volume of data • Used to represent data through Graphs, Histograms, Bell Curves etc. Data Presentation - FREQUENCY Class Interval (Scores) Frequency (occurrence) Cumulative Frequency 0-20 0 0 21-40 0 0 41-60 4 4 61-80 10 14 81-100 13 27
- 20. • Number expressing the central or typical value in a set of data. • MEAN = Sum of values / Count of values (n) • Also called Arithmetic Mean _ • Denoted by X Average Mean, Median & Mode • Represents the middle number in a given sequence of numbers when its orders by rank. • MEDIAN = n/2 OR (n+1)/2 • Most frequent value in a set of data.
- 21. • Presents grouped data with rectangular bars • Lengths proportional to the values that they represent Data Presentation – Bar Charts 0 10 20 30 40 50 60 70 80 90 100 A B C D E F G H I J K L M N O P Q R S T TEST SCORES Scores %
- 22. • Diagram consisting of rectangles whose area is proportional to the frequency of a variable. • Width is equal to the class interval Data Presentation – Histogram
- 23. • Shows distribution for variable from the mean value • The most common is also called Normal Distribution Data Presentation – BellCurve
- 24. • Difference between expected and actual output. • Variance is everywhere….every process has variance. • The least the Variance, better the process efficiency. • Always aim to control the variance. • Sum of squared differences from the mean divided by the total sample count. • Used to calculate Standard Deviation Variance
- 25. • Shows the variation in data • If the data is close together, standard deviation will be small • Denoted by a greek letter SIGMA Standard Deviation
- 26. Normal Distribution & Standard Deviation 1 Sigma - 68% 2 Sigma - 95% 3 Sigma - 99.73 %
- 27. Class Interval Mid Number (M) Frequency (F) (FM) MEAN _ X Deviation from Mean _ ( M – x ) Squared Difference from Mean X2 ( FX2 ) 0-10 5 2 10 49.33 -44.3 1962.49 3924.98 11-20 15 6 90 49.33 -34.3 1176.49 7058.94 21-30 25 12 300 49.33 -24.3 590.49 7085.88 31-40 35 25 875 49.33 -14.3 204.49 5112.25 41-50 45 34 1530 49.33 -4.3 18.49 628.66 51-60 55 30 1650 49.33 5.7 32.49 974.7 61-70 65 20 1300 49.33 15.7 246.49 4929.8 71-80 75 15 1125 49.33 25.7 660.49 9907.35 81-90 85 5 425 49.33 35.7 1274.49 6372.45 91-100 95 1 95 49.33 45.7 2088.49 2088.49 n=150 7400 48083.5 Calculating Standard Deviation - Average Mean = FM/Sum of Frequency (F) = 7400/150 = 49.33 Standard Deviation (Sigma) = = 48083.5/150 = 17.9
- 28. Mean49.3 1123 2 3 (49.3+17.9) 67.2 (49.3-17.9) 31.4 (31.4-17.9) 13.5 (67.2+17.9) 85.1 =17.9 Understanding Standard Deviation - 1 Sigma - 68% 2 Sigma - 95% 3 Sigma - 99.73 %
- 29. Target Customer Specification Target Customer Specification 1 2 3 0.27% Defects 3 After 6 No Defects! Reducing Variability Is The Key To Six Sigma How Six Sigma works?
- 30. Do we know? Defect & Defective Cost Of Poor Quality Process Capability
- 31. The Process Capability is a measurable property of a process to the specification. Expressed as a process capability index (e.g., Cpk or Cpm) Two parts of process capability are: 1) Measure the variability of the output of a process 2) Compare that variability with a specification or product tolerance. Process Capability Cpk = (USL – LSL)/ 6 σ
- 32. FOCUS Of Six Sigma Which one should be controlled??? Input OR Output ‘Y’ OUTPUT EFFECT RESULT Dependent Symptom Monitor ‘X’ INPUT CAUSE FACTOR Independent Problem Infuse
- 33. Six Sigma Increase Profits 3.4 Defects per Million Opportunities
- 35. Overview of Six Sigma
- 36. Overview of Six Sigma
- 37. DMAIC vs DMADV
- 38. DMAIC
- 39. DMAIC
- 40. DEFINE Identify Customers & Project CTQ Develop Project Charter Project Scope
- 41. • VOC to be further transformed into CTQs. • EG. A requirement of the customer is that the tiffins should be delivered on time. Thus for the customer, Delivery is Critical to the Quality (CTQ). • TOOLS USED – • VOICE OF CUSTOMER/CTQ TREE • AFFINITY DIAGRAM DEFINE – IDENTIFY PROJECT CTQ
- 42. DEFINE – #CTQ TREE Delivery On Time – Within +/- 30 min of Lunch Time
- 43. • BUSINESS CASE – WHY? • PROBLEM STATEMENT – WHAT? • GOAL STATEMENT – WHEN? • ROLES – WHO? • TOOLS USED – • STAKEHOLDER ANALYSIS • GNATT CHARTS DEFINE – DEVELOP PROJECT CHARTER
- 44. DEFINE – #PROJECT CHARTER
- 45. • Clearly define In & Out of Scope Items • TOOLS USED – • SIPOC • STRATIFICATION ANALYSIS DEFINE – IDENTIFY PROJECT SCOPE
- 46. DEFINE – #STRATIFICATION ANALYSIS Is Is Not Distinctions Geography INDIRAPURAM VAISHALI & VASUNDHARA VAISHALI & VASUNDHARA are sub contracted Output Delivery time Mixups, Hygiene, Temperature Customer Lower and Middle Income Higher Income Premium service for higher income group Time After Jul’18 Before Jul’18 Increased employees in Jul’18
- 47. DEFINE – SIPOC
- 48. MEASURE Current Process Flow Data Collection, Project Y & MSA Define Performance Standards
- 49. • Determine how the current process performs • TOOLS USED – • Value stream mapping • Process flow diagrams MEASURE– DETERMINE CURRENT PROCESS FLOW
- 51. • Collect Data for the identified CTQ • Translate CTQ to the measurable output Y • Eg for delivery – • No. of deliveries delivered during shipping window • Time taken to travel from DABBAWALA to customer’s place • Actual delivery time perceived by the customer • MEASUREMENT SYSTEM ASSESSMENT • TOOLS USED – • BRAINSTOMING • CHECK SHEETS MEASURE– DATA COLLECTION, PROJECT Y & MSA
- 52. MEASURE– #Gauge R & R
- 53. • What is the definition of a defect? • What is customer’s requirement on delivery time? EG – For Delivery -Capture Target Delivery Time -Get the allowed specification limits on Y MEASURE– DEFINE PERFOMANCE STANDARDS
- 54. 15 15 30 45 LATE DELIVERY LSL USL Visualize customer requirements Customer does not want earlier than this The customer tolerance window is 30 minutes on either side This is the target delivery time 45 30 EARLY DELIVERY 0 (MINUTES) Customer does not want later than this MEASURE– DEFINE PERFOMANCE STANDARDS
- 55. ANALYZE ESTABLISH PROCESS CAPABILITY DEFINE PERFORMANCE OBJECTIVES IDENTIFY SOURCES OF VARIATION
- 56. ANALYZE – ESTABLISH PROCESS CAPABILITY • What are the chances of your process creating defects? • Measure Variation in the current process • TOOLS USED – • HISTOGRAM • BOX PLOT • STANDARD DEVIATION
- 57. ANALYZE – #HISTOGRAM Target time hrs Delivery time hrs 12:30 12:18 12:30 12:26 13:30 13:34 13:30 13:42 13:00 13:07 13:00 13:06 13:30 13:23 13:30 13:41 12:30 12:26 12:30 12:37 Target time hrs Delivery time hrs 13:00 12:38 13:00 13:09 13:00 12:49 13:00 13:05 13:30 13:31 12:00 12:04 12:00 12:08 12:00 12:10 13:00 13:18 13:00 13:01 Target time hrs Delivery time hrs 13:30 13:45 13:30 13:21 12:00 12:17 12:00 11:58 12:00 12:17 12:00 11:46 12:00 11:53 12:30 12:33 12:30 12:30 13:30 13:28 25-25 -15 -5 5 15
- 58. 15 30 LATE DELIVERY 45 Measure the process output 45 30 15 EARLY DELIVERY 0 (MINUTES) NUMBEROFDATAPOINTS Dabbawala has an average delivery time () 2 minutes late and a standard deviation () of 10 minutes ANALYZE – #HISTOGRAM
- 59. 15 30 LATE DELIVERY 45 Measure the Process Output 45 30 15 EARLY DELIVERY 0 (MINUTES) USLLSL 30 minutes 10 minutesZ = Process standard deviation Customer tolerance Z = Z = 3 ANALYZE – #STANDARD DEVIATION
- 60. ANALYZE – #SIGMA LEVEL • Z OR SIGMA LEVEL determines the process capability. • Z value = 3*CPk • Z can be calculated as =(USL – Mean)/Standard Deviation
- 61. Target LSL USL Center Process Reduce Spread Excessive Variation in Process T a rg e t U S LL S L T a rg e t U S LL S L Process Off Target ANALYZE – DEFINE PERFORMANCE OBJECTIVE • Check the actual problem of your process – • Shift the target OR reduce the variance • TOOLS USED – • DATA DISTRIBUTION
- 62. ANALYZE – IDENTIFY SORUCES OF VARIANCE • Identify possible factors leading the problem • TOOLS USED – • FISHBONE OR CAUSE & EFFECT OR ISHIKAWA • FMEA • BRAINSTORMING • PARETO
- 63. ANALYZE – #FISHBONE Delivery Time MACHINE MOTHER NATURE MAN METHODS Poor dispatching Delivery person gets lost Delivery person does not show up Poor handling of largeorders Run out of storage space on vehicles sacks Don’tknow routes High turnover Get wrong Did not understand labels No teamwork No training Unreliable bikes Delivery persons own junk Cant locate employees homes Not on std routes Did not understand labels Too few delivery persons Uneven distribution of delivery loads No money for repairs per person Too few delivery persons Too many orders Large items difficult to carry in bus /bikes Too many MATERIALS Sacks too small Too much traffic Weather Bus service unreliable in peak hours Parkingspace problem
- 64. ANALYZE – #PARETO • 80-20 RULE • Identify VITAL few – CUT OFF level to be decided by the team on the basis of • Process Knowledge • Resource Availibility • TOOLS USED – • FISHBONE OR CAUSE & EFFECT OR ISHIKAWA • FMEA • BRAINSTORMING • PARETO Others Location Order Size Parking Traffic Label
- 65. IMPROVE SCREEN POTENTIAL SOLUTIONS VALIDATE SOLUTIONS IMPLEMENT SOLUTIONS
- 66. IMPROVE – SCREEN & VALIDATE SOLUTIONS • Screen all possible solutions • Discover casual relationship & identify the most effective ones • TOOLS USED – • POKA YOKE • 5S • SCATTER PLOTS • REGRESSION ANALYSIS DELIVERYTIMESPAN
- 67. IMPROVE – #SCATTER PLOT • TWO dimensional data visualization • Dots represents values for two different variables • Also called as Correlation Plots • • Discover casual relationship & identify the most effective ones • TOOLS USED – • POKA YOKE • 5S • SCATTER PLOTS • REGRESSION ANALYSIS
- 68. IMPROVE – #5S
- 69. IMPROVE – #POKA YOKE
- 70. CONTROL ESTABLISH NEW PROCESS CAPABILITY IMPLEMENT PROCESS CONTROL
- 71. CONTROL – ESTABLISH NEW PROCESS CAPABILITY • Gauge process performance and establish new Process Capability 15 30 LATE DELIVERY 4545 30 15 EARLY DELIVERY 0 (MINUTES) USLLSL Z = 4.5
- 72. CONTROL – IMPLEMENT PROCESS CONTROL • Control Plan should be in place to ensure sustained improvement • Use Of Control Charts • Documentation of Control Plan • Standardization • TOOLS USED – • CONTROL CHARTS • RUN CHARTS • PROCESS FLOW • SOP 30 20 10 0 No.lunches/person
- 73. CONTROL – #RUN CHARTS • Line Graph of data plotted over TIME • Shows Trend and Pattern in a process • Shows how the process is performing • Unlike Control Charts, RUN CHARTS cant say if a process is stable
- 74. CONTROL – #CONTROL CHARTS • Line Graph of data plotted over TIME • Shows Trend and Pattern in a process • Central Line for Average with Upper Control Limit & Lower Control Limit, helps to gauge is process is stable • Two Types – XBAR & RCHARTS
- 75. DMAIC
- 76. TO BE CONTINUED… HAPPY LEARNING!