Presentation on Six Sigma Green Belt IMC

INTERNATIONAL MANAGEMENT CONSULTANCY Al Ferdous Tower # 21, 4th Floor, Office #401 Salam Street P.O. Box: 43191, Abu Dhabi, UAE. www.imcinstitute.ae | [email protected] | +971 2627 2111| +971 56994 6111 Lean Six Sigma Green Belt + Black Belt Master Class Training v 3.0 1

Lean Six Sigma Green Belt + Black Belt Training

Cultivating Statistical knowledge All participants have to download a Global Statistical Tool named Minitab Tool . This tool comes in 2 ways : Licensed Version Free Trial Version Free Trial Version could be installed in one of the two ways : https://www.minitab.com/en-us/downloads/ https://www.minitab.com/en-us/products/minitab/free-trial/ Minitab software will automatically give one month free trial if you have not used it before and will be used for all statistical analysis and calculations during our sessions.

4 Lean Six Sigma Green Belt Coverage ….

5 Module 1 Concepts

Pillars of Six Sigma 6 Who Is the Customer Quality as per Customer Defect

There are 2 categories of customers in any organization – Internal Customers – Top Management, Colleagues, Teams, Support Teams in the Organization. External Customers – Ideally Clients 7 Customers

Quality - What are our Client’s Priorities? Different Clients have different priorities and perceptions about ‘Quality’ Zero Defect Within Budget Conform to Specs On Schedule

Product & Process Quality 9 Quality of end-products (Deliverables) depends on - Quality of the Inputs/Outputs Quality of the Processes AND

What is a Process? 10 Process: A sequence of steps performed for a given purpose People Tools PROCESS Process: A set of activities, methods, practices and transformations that people use to develop and maintain software and the associated products Procedures Procedure

Deming’s Cycle: PDCA 11 PLAN DO CHECK ACT

Eight Quality Management Principles 12 1. Customer focus 2. Involvement of people 3. Win-win relation with suppliers 4. Systems approach to management 5. Process approach 6. Leadership 7. Continual improvement 8. Factual approach to decision making

Why Do Projects Fail? 13 Estimation Requirements management (Understanding + Change control) Project planning and monitoring Resources (Availability / Skills) Managing client expectations Lack of processes/standards Configuration management Reviews / Testing Staff attrition +…..

What is Six Sigma It is Customer focused attitude – View quality externally from the customer’s perspective and Measure the same as the customer does. Customer focused All decisions are data driven, not on gut feel Method to achieve defect level of 3.4 per million opportunity. Method to achieve breakthrough improvements in process A Philosophy which mandates operational excellence A Set of tools designed to reduce variation in processes and operations Discipline to meet Customer expectations every time – Continuous improvement cycle Systematic [Measurable] Fact Based and Process Focus. 14

Key Goals of Six Sigma Understand the mindset of Six Sigma Understand the concepts and philosophy of Six Sigma Understand the basics, methodologies of Six Sigma How to handle a Six Sigma Project Sustain the Improvement for times to come Continual Monitoring Savings to the Organization 15

Improve Customer Satisfaction Reduce Cycle Time Reduce Turnaround Time Reduce cost of Poor Quality Effective Process Management for continuous improvement Reduce Defects or Non-Compliance in the process To do a project, the immediate solution to the problem is not known. 16 SIX SIGMA - BENEFITS

Six Sigma Yellow Belt / Awareness Sessions is targeted for People who are : - New to Six Sigma methodology and terms - Interested in the basics of Six Sigma - Keen to Understand what a Six Sigma Project is all about - Try to implement the initiative in day to day process Detailed Six Sigma Green Belt Trainings includes : - Predefined Problem Opportunity as a GB Project. - Detailed Training on all Phases of a Six Sigma Project - Usage and application of different tools - Breakout Exercises with sufficient hands on - Written Test - Gets the Green Belt Certification after completion of GB project - Average completion period 3-4 months. 17 TYPES OF TRAININGS

Detailed advanced Six Sigma Black Belt Trainings includes : - Predefined Criteria set for Black Belt Opportunity - Problem Opportunity as a BB Project. - Detailed Training on all Phases of a Six Sigma Project - Understanding the Roles & Responsibilities of a Black Belt - As an Executioner - As a Mentor - Usage and application of different tools, techniques and instruments - Detailed and Extensive Breakout Exercises with sufficient hands on - Written Test - Gets the Belt Belt Certification after completion of BB project - Average completion period 4-6 months. 18 TYPES OF TRAININGS

The Six Sigma Evolutionary Timeline 19

Identify Root Causes 20

Common tools for gathering VOC data 21 Method Description Advantages Disadvantages Interviews Information obtained from customers either by telephone or in person. Detailed information Follow up Expensive Talent of the interviewer Surveys (Most common tool used) A set of written questions that is sent to selected customers to obtain information that can be formatted into data Objective data Easy to interpret Poor response rate Different answers based on type of questions Focus groups A collection of customers who answer questions from a facilitator Follow-up questions Observing non-verbal behaviors Expensive Skill of the facilitator Observing the Customer Seeing the customer using your product or service Unfiltered information No follow up Complaints Information obtained while someone complains about a Situation Opportunity to make amends Few people complain

VOC to CTQ to Specification Limits

ISO 9001 : Process Orientation – Laying down a process. CMMI : Further focus on Key areas for further improvements. Six Sigma : Tools and Techniques of Improvement 23 QUALITY INITIATIVES - RELATION

Sponsor : Sponsors Projects, review and monitor progress, ensure implementation of improvement suggestions. Black Belt : Preaching Six Sigma culture across Organization, monitors and facilitates Six Sigma Projects across organization along with GBs. Helps and monitors GBs as & when reqd. A person need to complete his project as per criteria set to become a Black Belt. Green Belt : Monitor and facilitates individual Green Belt projects till implementation. Work closely with Black Belt. A person need to complete his project as per criteria set to become a Green Belt. Yellow Belt : Awareness of Six Sigma culture and methodologies with basic application knowledge. A person would have a Yellow Belt Certificate based on criteria set. Team Members : As associates in a Six Sigma Project -Learners 24 PARTIES TO A SIX SIGMA INITIATIVE

Any problem or pain area might not lead to a Six Sigma Project. Any problem or pain area which DOES NOT have an immediate solution would only apply for an Ideal Six Sigma Project. 25 WHAT COULD BE A SIX SIGMA OPPORTUNITY

A Six Sigma Project is a methodical approach which follows statistically and qualitatively driven way to achieve definite improvement in a process or even introduce a new process. It leads to definite savings to the Organization, may it be in P/Days or in financial values. It has to be constantly reviewed and re-tuned to sustain the improvement. 26 SIX SIGMA PROJECT

DMAIC Vs DMADV Process Management 27

Types of Six Sigma Project - Phases DMAIC : Improvement of an Existing Process Define - What is the Problem Measure - How much is the Problem Analyze - Why is the Problem Improve - How to improve upon the Problem Control - How to control the improvement DMADV : Introduction of a Totally New Process . Define - What is the Problem Measure - No Data, since process not in place Analyze - How to introduce a new process Design - What would be the new process-Design Verify - Check completed Design and transition to Customer 28

When to Implement DFSS DFSS stands for Design for Six Sigma. DMADV are the steps to achieve DFSS: Define, Measure, Analyze, Design and Verify. DFSS is used when you are making new products or services. When the current product or process is not required then it is time for a new product and you need DFSS or design for six sigma. DFSS is used when you want the high-quality product without defects, faster time to market, optimize the design, meet customer expectations, and be successful for the first time. 29

Design for six sigma types DFSS stands for Design for Six Sigma. DMADV are the steps to achieve DFSS: Define, Measure, Analyze, Design and Verify. There are 3 types of DFSS six sigma, DMADV – Define, Measure, Analyze, Design and Verify IDOV - Identify, Design, Optimize and Verify. DCCDI - Define, Customer, Concept, Design, Implementation All these approaches to DFSS are more or less achieves the same results that are to design the product to the customer expectations minimizing the defects. 30

Key aspect involved in implementing DFSS DFSS stands for Design for Six Sigma. DMADV are the steps to achieve DFSS: Define, Measure, Analyze, Design and Verify. Ask the following questions to implement the DFSS and these brainstorming questions will allow the entire development team in the organization to design a flawless product. What are the resources, required to make a successful project? What are the design tools required for success? What is the time frame required for the project? What is the budget required for successful project completion? What is the optimization required for the projects and the processes? What are the design requirements needed for products or services to meet customer expectations? What are the components required for making the products or services and what is the integration between the components required? 31

Deming’s Approach Eighty five percent of the reasons for failure to meet Customer Expectations are related to deficiencies in systems and processes rather than the employee. The role of management is to change the process rather than pushing individuals to do better. 32

Kano Model 33 Three Stages of Customer Satisfaction : Must Be Satisfier/Comforter Delighters

Cost of Good Quality 34 Prevention costs : The costs of activities specifically designed to prevent poor quality in products or services. Some of the examples are : Establishing Product Specifications Quality Planning Capability studies New Product Development and Testing Development of a QMS Proper Employee Training Reviews : Design / Quality/ Procedures Prototype testing Quality design Equipment maintenance Field testing Fixture design and fabrication Servicing Appraisal costs : The costs associated with measuring, evaluating, or auditing products or services to assure conformance to quality standards and performance requirements. Some of the examples are : Audits Document / Drawing / Proposal checking Equipment calibration Inspection : Receiving Inspection, In- process, Outgoing Inspection , PDI , Prototype Inspection Laboratory testing Procedure checking Prototype inspection Shipping inspection

Internal Failures – Costs associated with defects found before the product or service reaches the customer. Internal Failures may include, but are not limited to, the following examples: Excessive Scrap Product Re-work Waste due to poorly designed processes Machine breakdown due to improper maintenance Costs associated with failure analysis . COPQ (Cost of Poor Quality) 35 External Failures – Costs associated with defects found after the customer receives the product or service. External Failures may include, but are not limited to, the following examples: Service and Repair Costs Warranty Claims Customer Complaints Product or Material Returns Incorrect Sales Orders Shipping Damage due to Inadequate Packaging

Tollgate Reviews After the completion of each phase, projects will be presented to the leadership team called gate reviews or tollgate reviews. The purpose of the reviews is to monitor team progress, reinforce priorities and align project to business strategies. This gives an opportunity for the leadership team to provide ongoing coaching to the project team. Typical participants at the tollgate reviews are Green Belt, Black Belt, Project Sponsor, Master Black Belt and other Stakeholders The tollgate review has to be formal and should be setup by the project leader. Team sends the PowerPoint file 2-3 days in advance. The possible outcomes from the tollgate review are: Everything looks great, you can move on to the next phase of the project Some minor changes are required, make these changes and move ahead Major changes are required, make the changes and setup another tollgate review for approval Project is no longer aligned with the business and should be stopped Tollgate reviews help reduce project risk as you are not waiting until the end of the project to share your findings with the leadership team. Need to have tollgate review after each phase

Project Benefits 37 Level 1: HARD – Impact Profitability Increase market share Improve capacity of production and hence sell more units Increase product pricing Reduce cost of operations Reduce transaction cost by making process more efficient Reduce selling expenses by removing non-value added activities for sales Reduce distribution costs by having an e fficient route for product delivery Improve employee efficiency Reduce wastage in processes Reduce energy consumption in buildings Level 2: HARD – Impact Cash Flow Reduce the inventory that a company carries without impacting revenue or costs Get payment from customers on-time or earlier Delay payments to suppliers Level 3: SOFT – Future Opportunities Improve efficiency of people but not impact headcount Eliminate the chance for getting penalty Reduce usage of office space but not put out for rent Improve production capacity but not yet sell the additional items that can be produced Level 4: SOFT – Non Financial Benefits. Customer Satisfaction Employee Satisfaction Safety, Risk Reduction Other Non Financial Benefits

Benchmarking In business, benchmarking is a process used to measure the quality and performance of your company’s products, services, and processes. These measurements don’t have much value on their own—that data needs to be compared against some sort of standard. Types of benchmarking : Internal Competitive Strategic 38

39 Module 2 Define Phase

Why Define? 40 Success of a Six Sigma Project depends on proper definition of its scope Over 50% of projects fail or gets delayed due to inadequate scope definition Proper definition of Scope facilitates project execution and result in a timely manner

Project Charter 41 A team charter is a written document and works as an agreement between Sponsor and the team about what is expected from the project. WHY? Clarifies what is expected of the team. Keeps the team focused. Keeps the team aligned with organizational priorities. Transfers the project from the Sponsor(s) to the project team Please note that a Project charter is a live document and it can be modified at any point of time based on the findings of the Project team.

Elements of a Charter 42

Business Case - Example Our Customers of ABC Project use online customer service and product registration website extensively to get the problems solved. This improves the customer satisfaction and also increases the revenue. It also reduces inbound calls to our call center. Recently it has been observed that the site availability and response time has deteriorated considerably, leading to customer dissatisfaction and increased volume of inbound calls. A 5% decrease in availability could result in 200 more inbound calls per day resulting in additional cost of $ 2,000,000 per annum. In addition there is a potential revenue loss of approx amount of $ 5,000,000 per annum. 43

Problem Statement - Example Since Apr’05, the Customer Service website of AMEX Project is down for average 2 hours a week during working hours. This is significantly higher compared to 2003 average downtime of 30 mins per week. This negatively affects the customer satisfaction and cost on call center. Approach - Be Specific and Concise - Include Measurement - Should not include CAUSE - Should not state SOLUTIONS 44

Goal Statement - Example Improve Website Availability above 99% each day by January 2006. Improve Quality of Reconciliation Output from 80% as of current to 98% by December 2005. 45

SMART Problem & Goal Statements 46

Sample Project Charter 47

SIPOC Analysis

Benefits of SIPOC tool 49 SIPOC tool help: To know who supplies input to the process To know what are the inputs to the process To Know what are specifications of Inputs To know step by step flow of process To know the outputs of process To know the customer requirements / Specifications from the process To know the customer of a process

Sample SIPOC 50

51 Module 3 Measure Phase

Operation Transportation St o r a g e D e l a y D e c i s i o n Document Input to a process A2 Process flow c onne ct o r Start, stop of Process Off page c onne ct o r 2 - P a r t M e t ri c Inspection / M e a s u r e m e n t / Process Step Operation & Inspection Procedure Detailed Process Mapping 52 A On page c onne c tor Three Components : Value Added Activities Non Value Added Activities Value Enabling Activities

Data Collection Plan A data collection plan lists the details about how the data will be collected. This plan can then be handed over to the team members to execute the data collection plan and provide to the rest of the team members. What Operational Definition Who When Where How Much Cycle Time Time from order entry to order fulfilled in days John Beginning of every month SAP records 10% random sample Cost Amount of revenue in $ for each financial quarter Mary Every quarter Financial DB All records % Defects Number of defects divided by number of production pieces (including rework) Smith Every shift Scrap buckets 1 hour every shift change

Sample Size Calculation What is sample size? Sample size is the number of completed responses your survey receives. It’s called a sample because it only represents part of the group of people (or target population) whose opinions or behavior you care about. For example, one way of sampling is to use a “ random sample ,” where respondents are chosen entirely by chance from the population at large. With this definition in mind, let’s dive into the following topics: The different ways to interpret your sample’s results The formula used to calculate sample size Why having an appropriate sample size for a survey matters How the significance of sample size varies across survey types Understanding sample sizes 54

55 Here are three key terms you’ll need to understand to calculate your sample size and give it context: Population size: The total number of people in the group you are trying to study. If you were taking a random sample of people across the U.S., then your population size would be about 317 million. Similarly, if you are surveying your company, the size of the population is the total number of employees. Margin of error: A margin of error tells you how many percentage points your results will differ from the real population value. For example, a 95% confidence interval with a 4 percent margin of error means that your statistic will be within 4 percentage points of the real population value 95% of the time. Sampling confidence level: A percentage that reveals how confident you can be that the population would select an answer within a certain range. For example, a 95% confidence level means that you can be 95% certain the results lie between x and y numbers. Sample Size Calculation

Sample Size Calculation 56

Center or Central Tendency 57 Measure of Central Tendency of Data – 3 Ms :- M ean : Mean or Average is signified by X bar M edian : This is the value which has equal number of values above it and below it, when arranged in ascending order. M ode : This is the value which occurs with the highest frequency.

Objectives of Variation 58 Understand that variation in the data represents the voice of the process Recognize two general types of variations – common cause and special cause – and the implementation of the different causes. Use appropriate tools to study variation Understand process variation to identify and control/eliminate the primary sources or causes of the variation. Data type dictates how much variation we will see.

Two Types of Variation 59 Type of Variation Characteristics Common Cause Always Present Expected Predictable Normal Random Special Cause Not always Present Unexpected Unpredictable Not Normal Not Random Chance

Spread or Dispersion 60 The measures of variation in data are – Range [R] : It is the difference between the maximum and minimum values; Range (R) = Max Value – Min Value Standard Deviation: Standard deviation is calculated as Square root of Variance. It is the sum of the square of difference of each data point from the mean. So if each data point deviates to a large extent from the mean, its square and the total of all the squares will be large. Hence when SST is higher it indicates higher variation. SST is thus an important concept to measure variation. Variance : The difference of each data point from the average. Population Variance is the sum of the square of the difference of each data point from the average divided by the no. of data points.

Importance of Standard Deviation 61 Why is Standard Deviation an important Measure – Let us take an Exercise – There are two Batsmen A & B Given are 20 data points of the runs scored by these 2 batsmen. Each data point is the run scored in one day series. Mr A Mr B Batting Avg 53.9 Batting Avg 50.8 Std Dev 64.44 Std Dev 7.76

Understanding Data Types

Normal Six Sigma Curve 63 Normal distribution, is a probability distribution that is symmetric about the mean, showing that data near the mean are more frequent in occurrence than data far from the mean. In graphical form, the normal distribution appears as a " bell curve ".

Normality Checking 64 Check the Normality of Data : Minitab : Stat > Basic Statistics > Graphical Summary Brand A : p-value is 0.007 so the Data is Not Normal. Brand B : p-value is 0.236 so the Data is Normal.

Histogram and Bar Graph Appropriate for Continuous data Larger sample size (> about 50) Can be used to What is the central location and spread of data View shape of the distribution (is it skewed) Check if the distribution is normally distributed Check if data is unimodal or multimodal Are there gaps or outliers in the data Appropriate for Attribute data Binary or ordinal Can be used to Compare relative proportions

Minitab Minitab software : (https://www.minitab.com/en-us/downloads/) Minitab software will automatically give one month free trial if you have not used it before. Minitab will be used for all statistical calculations

Introduction to Minitab Minitab is designed for statistical analysis and can be easy to use once you know where to get help We will be using Minitab software throughout the learning process Approach to use Minitab: Manipulate your data in Excel and then copy and paste the data into Minitab. Use Minitab to analyze your data and then copy and paste the analysis to Power Point. Make sure to save your analysis so that you can revisit them later. Main sections: Menu bar (access different tools & functionality) Session window (all your text output goes here) Worksheet (enter your data here in columns) Minitab is the standard tool for data analysis in the Six Sigma world

Minitab Exercise on Probability Distribution 68

Binomial Probability Distribution: Exercise 69 If the data is binary, then probabilities are estimated using Binomial Distribution On an average, 2 % of the transactions processed in a process are defective. On a particular day, out of 400 transactions audited 21 turned out to be defectives. Is it an indication that the process performance deteriorated?

Calculation of Binomial Probabilities using Minitab 70 Step 1: Copy the defective data to Minitab Worksheet as shown below:

Calculation of Binomial Probabilities using Minitab 71 Step 2: Go to C a l c > P r o b ab ili ty Di st r i b u t i on s > B i no m i a l

Calculation of Binomial Probabilities using Minitab 72 Step 3: Select Cumulative probability, Enter Number of trials, Probability of success, Input Column, Option Storage and click ‘OK’ button A cumulative probability refers to the probability that the value of a random variable falls within a specified range. Frequently, cumulative probabilities refer to the probability that a random variable is less than or equal to a specified value.

Calculation of Binomial Probabilities using Minitab 73 Step 4: Minitab will calculate Binomial Probabilities as display in Optional Storage Column as shown below

Binomial Probability Distribution: Exercise 74 Number Audited (n) Defectives (d) Chance of getting d or less defects (%) 400 0.03 400 2 1. 31 400 4 9.73 400 6 31.09 400 8 59.26 400 10 81.79 400 12 93.81 400 14 98.38 400 16 99.66 400 18 99.94 400 20 99.99

Binomial Distribution: Exercise 75 Let p = 2 % = 0.02 Number of Transactions Audited = 400 From Binomial Distribution, Probability of getting less than 20 defectives in 400 transactions = 0.9999 Hence Probability of getting 20 or more defectives = 1 – 0.9999= 0.0001 i.e. if the process is operating at 2 % defectives: T he chance of getting 21 defectives out of 400 is almost 0 , Process performance has deteriorated.

Poisson Probability Distribution: 76 If the data is Count, then probabilities are estimated using Poisson Distribution Poisson Distribution: Exercise The average number of repeat calls per day in a voice process is 20. On a particular day , there were 25 repeat calls. Is there any problem with the process that day?

Calculation of Poisson Probabilities using Minitab 77 Step 1: Copy the different values of repeat calls to Minitab worksheet as shown below:

Calculation of Poisson Probabilities using Minitab 78 Step 2: Go to C a l c > P r ob a b ili ty Di st r i bu t i o n s > P o i ss o n

Calculation of Poisson Probabilities using Minitab 79 Step 3: Choose Cumulative probability, Enter Mean, Input column & Optional storage as shown below and click “OK” button. A cumulative probability refers to the probability that the value of a random variable falls within a specified range. Frequently, cumulative probabilities refer to the probability that a random variable is less than or equal to a specified value.

Calculation of Poisson Probabilities using Minitab 80 Step 4: Minitab will display the probabilities in the Optional storage column as shown below Note: % Chance = Probability x 100

Poisson Probability Distribution: Exercise 81 Average number of repeat calls per day = 20 Average Repeat Calls Repeat Calls (d) Chance of getting d defects or less (%) 20 0.00 20 5 0.01 20 10 1.08 20 15 15.65 20 20 55.91 20 24 84.32

Poisson Probability Distribution: Exercise 82 Average number of repeat calls per day = 20 Probability of getting less than 25 repeat calls = 0.84 Hence Probability of getting 25 or more repeat calls = 1 - 0.84= 0.16 = 16 % I f the process is operating at 20 repeat calls per d ay : T he chance of getting 25 repeat calls is 16 % . 16 % is large enough to c onclude that there is nothing wrong in the Process.

Measurement Systems Analysis (MSA) Repeatability : It is the variation in measurements obtained with one measurement instrument when used several times by same person while measuring the identical characteristics of the same part. This is also called as Equipment Variation. Reproducibility : It is the variation in the average of measurements made by different people using the same measurement instrument while measuring the same characteristics on same part. This is also known as Appraiser Variation. The MSA study for continuous and discrete data is quite similar. We take 10 to 20 samples for a study, provide them to two or three appraisers for the first trial, and then rerun the study. The main difference is in the fact that the appraisers use a gauge to measure the part in continuous data. For discrete data, however, it is left to the knowledge of the appraisers whether the transaction is defective.

Analyzing the Gage R&R Results 84 Attribute Gage R&R reveals two important findings – percentage of repeatability and percentage of reproducibility. Ideally, both percentages should be 100 percent, but generally, the rule of thumb is anything >=80 percent is quite adequate. Continuous Gage R&R study (Stat > Quality Tools > Gage Study ) indicates whether the inspectors are consistent in their measurements of the same part (repeatability) and whether the variation between inspectors is consistent (reproducibility). R&R <= 10% - Measurement System acceptable R&R 10% to 30% - May be acceptable. Make decision based on importance of characteristics, customer input etc. R&R over 30% - Not acceptable. Find problem, remove root cause.

Minitab Exercise on Gage R & R 85

Exercise: Gage R & R– Continuous Data 86 G i v e n t h e da ta b e l o w f o r r ea d i n g b y 3 a p p r a i s e r s o n 6 c a ll s w i th 2 t r a il s . D etermine whether the measurement system is acceptable . Call ID Appraiser A Appraiser B Appraiser C 1 2 1 2 1 2 1 65 60 55 55 50 55 2 100 100 100 95 100 100 3 85 80 80 75 80 80 4 85 95 80 75 80 80 5 55 45 40 40 45 50 6 100 100 100 100 100 100

Step 4: Minitab will give the following Output Source Var Comp % Contribution Total Gage R & R 17.434 4.06 Repeatability 7.338 1.71 Reproducibility 10.096 2.35 Part-To-Part 4 1 1 . 5 6 8 95.94 Total 42 9 . 2 100 Source StdDev (SD) (6 * SD) (%SV) Total Gage R & R 4.1754 25.052 2 . 1 6 Repeatability 2.7088 16.253 1 3 . 6 Reproducibility 3.1774 19.065 1 5 . 3 4 Part-To-Part 20.2871 1 2 1 . 7 2 3 9 7 . 9 5 Total Variation 20.7154 1 2 4 . 2 7 4 100 If < = 1 %. Gage acceptable Else if > 30 %, Gage not acceptable 87 Exercise: Gage R & R– Continuous Data

Six Sigma Table 88

Capability Index for 2-Sided Specifications Cpk (Actual Process Capability) Determine Cpk for USL Determine Cpk for LSL Cpk is the minimum of the two values The value of Cpk is the minimum of two process capability indices. One process capability is Cpu , which is the process capability based on the upper specification limit. The other is Cpl , which is the process capability based on the lower specification limit. Algebraically, Cpk is defined as shown in the figure. · Cpl measures how close the process mean is running to the lower specification limit · Cpu measures how close the process mean is running to the upper specification limit · Cpk equals the lesser of CPU and CPL.

Capability Index – Some Other Terms CP Means Potential Capability and is calculated as : Ppk is for Long Term and ideally for Full Population, which talks about Process Performance.

Six Sigma Tools: DPU, DPMO and PPM Defects per Million Opportunities (DPMO) – a ratio of the number of defects (flaws) in 1 million opportunities when an item can contain more than one defect. To calculate DPMO, you need to know the total number of defect opportunities. For Exercise, a form contains 15 fields of information. If 10 forms are sampled and 26 defects are found in the sample, the DPMO is:

Determining Sigma Level for Data 92

Minitab Exercise on Process Capability 93

Exercise : Process Capability for Continuous Data 94 The cycle time (in Minutes) of each transaction in a day on both the shifts was collected. The SLA for the cycle time of the process is 60min. Calculate the process capability. Cycle Time Shift 50 1 51 1 50 1 55 1 56 1 52 1 48 1 52 1 51 1 49 1 52 1 50 1 56 1 52 1 52 1 Cycle Time Shift 60 2 55 2 49 2 53 2 46 2 51 2 50 2 41 2 51 2 58 2 54 2 57 2 50 2 49 2 41 2

Select Stat > Quality Tools > Capability Analysis > Normal Capability Test 95

Select Stat > Quality Tools > Capability Analysis Capability Test 96

Defectives Process Capability (Binomial Capability) 97 The number of defective tractors produced have been collected for a period of 1 year (52 weeks of data). The data is stored in columns: Tractors Produced and Tractors Defective. The maximum percentage defective we are willing to accept is 5%. Determine the capability of this process. Defectives data follow a binomial distribution

Yield Yield is defined as the number of units coming out of a process divided by the number of units going into that process over a specified period of time. Only good units with no rework or scrap are counted as coming out of an individual process .

Yield (Multiple Step Process) Step No. Quantity Bills In Defectives Defective Bills Corrected OK Bills Out Step Yield Classic Yield 1 100 5 4 99 0.99 2 100 3 2 99 0.99 3 100 2 1 99 0.99 0.97 Step 2 Step 1 Step 3 Let’s look at a process of generating bills that has 3 steps. Data for each step in the process is shown below So final rejection is = 3. Final yield shows higher value. Although, final yield measures process performance, It does not expose HIDDEN FACTORY!

First Pass Yield (FPY) First pass yield is the probability that all CTQ opportunities produced in a step in the process will conform to their respective standards. Exercise : Total Bills checked = 100 Defects Total No. Wrong Name Spelling 6 Wrong Address 3 Wrong Quantity 1 Taxation Terms not clear 4 Total Defective – 3 % Rejection – 3% Final Step Yield – 97% Total Defects – 14 First Pass Yield – 86%

Rolled Throughput Yield (RTY) Since Classic Yield, which is unit driven, is high and if only these measures are reported, it creates false impression of good quality level and efficiency. However, Rolled Throughput Yield, which is defect driven, is an accurate measure of performance which exposes Defects, COPQ, cycle time and work in process inventory. Step 2 FPY 90% Step 1 FPY 86% Step 3 FPY 95% RTY = 0.86 * 0.90 * 0.95 = 0.73

102 Module 4 Analyze Phase

Brainstorming 103

Brainstorming Session Identify root causes or major value drivers Plan for session (use multiple sessions if required, clearly define the scope) Invite the right people (limit the number to 10) Provide written invitations to participants (at least a few days in advance) Establish ground rules (for an effective session) Capture every idea to eliminate or reduce the root causes Establish a time limit (typically 30-60 minutes) Organize and categorize after the session Ensure contribution from all team members

Multi-Voting Use this technique to narrow down the list of solutions to a manageable number Clarify with the team what the team is voting on – make sure all team members understand each solution Give each team member a number of votes (approx. half the number of items on the list) Each person assigns their vote to a given item (either done in parallel or sequential) After one round of voting, drop items with few or no votes from the list If the top list of ideas are not clear, do a second round of voting, assigning fewer votes per person Idea 1: Online Monitoring Idea 2: Enforce SOP Idea 3: Pokayoke Device Idea 4: Buddy System Idea 5: Offline Training Idea 6: Outsource Sol.

After the Brainstorming 106 SR. NO. RELEVANCE FREQUENCY RATING 1 High High A 2 High Medium B 3 Medium High 4 Medium Medium C 5 High Low 6 Low High 7 All other combinations D

Cause & Effect Analysis

Why Why Analysis In a 5Why Analysis, 5 is not a magic number, in some cases it could take more than 5 Why’s and in some cases less than 5 Whys

Question: Why are you throwing sawdust on the floor? Answer: Because the floor is slippery and unsafe. 2. Question: Why is the floor slippery and unsafe? Answer: Because there is oil on it. 3.Question: Why is there oil on it? Answer: Because this machine is dripping. 4.Question: Why is it dripping? Answer: Because oil is leaking from this coupling. 5.Question: Why is it leaking? Answer: Because the rubber seal inside the coupling is worn out. Why Why Analysis – Example 1

1. Question: Why did the machine stop? Answer: Because the fuse blew due to an overload. 2. Question: Why was there an overload? Answer: Because the bearing lubrication was inadequate. 3. Question: Why was lubrication was inadequate? Answer: Because the lubrication pump wasn’t working right. 4. Question: Why wasn’t the lubrication pump working right? Answer: Because the pump axle was worn out. 5. Question: Why was it worn out? Answer: Because sludge got in. Why Why Analysis – Example 2

Pareto Analysis

Hypothesis Methodology Hypothesis testing methodology Formulate a question dealing with a practical problem Translate it into a statistical hypothesis to be evaluated Determine a statistical decision Translate it back into a practical solution Hypothesis tests require very few samples to make objective judgments about a population parameter When conducting a hypothesis test there is always a risk of making a wrong decision However, we can control the degree of risk by taking a few precautions to make the hypothesis methods powerful tools to make good decisions based on data.

Hypothesis Testing 113 Two sets of data are under comparison, statistically they are same – Null Hypothesis H0. They are different – Alternate Hypothesis Ha. To draw a conclusion whether group(s) of data similar We start by making a statement that ‘Groups are Similar’. Based on calculations, we either accept or reject the above statement. So, first step towards Hypothesis Testing is to find evidence against the statement. The Null Hypothesis assumes that all groups are equal. This is like a person standing trial is innocent, until proven guilty. You are the opponent’s lawyer; You must prove that he is guilty.

Minitab Exercise on Hypothesis Testing (Parametric) 114

Exercise : Problem Description In Illinois, there is a packing plant named Fabric Manufacturers. In it, a machine packs cartons with jars. It is supposed that a new machine will pack faster on the average than the machine currently used. To test this hypothesis, the times it takes to pack ten cartons are recorded (in seconds). Assume a difference of 1 second as practically important difference. + more records…

2-Sample t Test Click on Assistant > Hypothesis Tests > 2-Sample T Specify that each sample is in its own column Specify the two columns ( PackTime_Old ) and ( PackTime_New ) Specify that we are interested in the alternative hypothesis of old time is greater than the new time Select an alpha risk of 0.05 (95% confidence) Select a practical difference value of 1. Click on the OK button Ho: Mean pack times are the same. Ha: Pack time of newer machine is less than the older machine

2-Sample t Test Sample size for our analysis is sufficient (N = 10) P value of 0.002 indicates that we reject the null hypothesis. Confidence interval indicates that the difference in pack times lies between 0.53 to 1.65 seconds.

Exercise : Problem Description In the same organization, in Illinois, we want to find out if the Standard Deviation of the Pack Time Old Machine is different from the Standard Deviation of the Pack Time New. + more records…

Standard deviation Test Click on Assistant > Hypothesis Tests > 2-Sample Standard Deviation Specify that each sample is in its own column Specify the two columns ( PackTime_Old ) and ( PackTime_New ) Specify that we are interested in the alternative hypothesis of standard deviation of the old machine is different from the standard deviation of the new machine Select an alpha risk of 0.05 (95% confidence) Leave the % reduction blank. Click on the OK button Ho: Mean pack times are the same. Ha: Pack time of newer machine is less than the older machine

2-Sample Standard Deviation P value of 0.789 indicates that we accept the null hypothesis. Confidence interval indicates that the two confidence intervals for the standard deviations overlap => that the population standard deviation could be the same!

Exercise : Problem Description Unique Healthcare has discovered a drug to reduce the headache. We would like to test the efficacy of this drug – so we perform a controlled study. There are 20 participants in this study. Half of them get the drug and the other half get a placebo for the first one month of the study period. The next month, the first half who got the drug now get the placebo and the other half get the drug. The difference in the reported times for it takes to relieve the headache with the placebo and with the drug are shown in the table. Determine if the drug is better than the placebo in relieving the headaches by at least 5 minutes.

Paired t Test Click on Assistant > Hypothesis Tests > paired T Specify the two columns DS_Placebo DS_Drug Specify that we are interested in the alternative hypothesis of placebo taking longer to cure headache vs. the drug Select an alpha risk of 0.05 (95% confidence) Select a practical difference value of 5 minutes. Click on the OK button Ho: Times for placebo and drug are the same. Ha: Headache is cured faster for drug vs. placebo

Paired t Test Sample size for our analysis is sufficient (N = 20) P value of 0.001 indicates that we reject the null hypothesis. Confidence interval indicates that the difference in times is between 5.7 and 11.2 minutes.

Exercise : Problem Description In We-All Infotech, the current process of software development resulted in 3 defects per 1000 lines of code. An improvement was made to the process and the new process resulted in 2 defects in 800 lines of code. Can we say statistically that we have made an improvement to the process?

2-Sample % Defective Click on Assistant > Hypothesis Tests > 2-Sample % Defective Specify the sample statistics: Number of items tested and number of defectives for the old software Number of items tested and number of defectives for the new software Specify the alternative hypothesis of % defective for the old software greater than the % defective for the new software. Select an alpha risk of 0.05 (95% confidence) Click on the OK button Ho: % Defective (old) = % Defective (new). Ha: % Defective (old) > % Defective (new)

2-Sample % Defective Sample size is sufficient to detect a difference of 0.47 at 60% power P value of 0.6 indicates that we accept the null hypothesis

Exercise : Problem Description The number of defectives observed in a process on one of the days is 54 out of 1000 items inspected. Can we statistically conclude that the % of defectives is greater than 5%? In this regard just some decimal increment would not be considered as a significant increment.

1-Sample % Defective Click on Assistant > Hypothesis Tests > 1-Sample % Defective Specify the number of items inspected (1000) and the observed number of defectives (54) Specify the target % defective (5) Specify the alternative hypothesis of % defective greater than 5% Select an alpha risk of 0.05 (95% confidence) Click on the OK button Ho: % Defective = 5%. Ha: % Defective > 5%

1-Sample % Defective Sample size is sufficient to detect a difference of 2.18 at 90% power P value of 0.3 indicates that we accept the null hypothesis 90% Confidence interval indicates that the % defective lies between 4.3 and 6.7%

Correlation Analysis 130 Correlation is a statistical measure which determines the co-relationship or association of two quantities. It is used to represent the linear relationship between two variables. In Correlation, there is no difference between dependent and independent variables i.e. correlation between x and y is similar to y and x. Correlation indicates the strength of association between variables. Correlation aims at finding a numerical value that expresses the relationship between variables. Correlation Ascertainment : 1 Indicates Strong Positive Correlation -1 indicates Strong Negative Correlation 0 indicates No Correlation

Scatter Diagrams 2 Dimensional XY plots Used to show relationship between independent (x) and dependent (y) variables A graph that shows how two variables are related to one another Data can be used in a regression analysis to establish equation for the relationship No Correlation Positive Correlation Negative Correlation

Correlation Analysis If I want to check if TV Installation Time is having a Correlation with TV Order Processing Time. In this case, we would find out that TV Installation Time is having a relation with Order Processing Time and Order Processing Time is also having the same relation with TV Installation Time. 132

Correlation Analysis • Click on Stats > Basic Statistics > Correlation • Select the columns for which you want to compute the correlation analysis • Installation Time • Order Processing Time • Select the checkbox to display P values • Click on the OK button • From the analysis, we can see that the correlation coefficient is 0.426 (weak) but the P value is statistically significant (P < alpha) 133

Regression Analysis 134 For Regression, we check for P Value < 0.05 or not. If P value is < 0.05, we could conclude that Regression between Y and X is good. Also R Square Value to be checked if it is > 90% or not.

Minitab Exercise on Regression Analysis 135

Exercise : Problem Description 136 The manager of Princep Motors wants to investigate how the plant’s electricity usage depends on the plant’s production. The data for each month of the year is shown. The electrical usage is in million KWH and the production is measured in million dollar units of the cars produced that month. Car Production Electricity Usage 4.51 2.48 3.58 2.26 4.31 2.47 5.06 2.77 5.64 2.99 + more records…

Regression Analysis Click on Assistant > Regression > Simple Regression Select Y column as Electricity Usage Select X column as Car Production Select the Linear Model Select Alpha level of 0.05 Click on the OK button

Regression Analysis

Regression > Understanding Minitab Output 139 Lucky Foods is analyzing factors that affect the percentage of crumbled potato chips per container before shipping (response variable). You are conducting the regression analysis and include the percentage of potato relative to other ingredients and the cooking temperature (Celsius) as your two predictors. The following is a table of the results. If the p-value is smaller than the alpha-level, the association is statistically significant. R^(adj) indicates that 63.61% of the variation in y is explained by variation in chosen X’s

Gap Analysis The purpose of gap analysis is to ensure that we have sufficient number of root causes (X’s) in order to address the problem at hand and achieve the improvement we are after. When you perform a hypothesis test, you need to answer two questions: is it statistically significant and practically important. If it is not, then you don’t continue with that cause. Keep looking for causes that satisfy both the constraints. If you can’t meet your improvement target that you promised in your D-phase, then continue to look for more root causes until you can get the improvements that you are after. If none of your root causes can achieve your target, then you either need to expand your search for causes, use a different approach to solve (like DFSS) or change your improvement target! Cause & Effect Diagram Most Likely Causes Hypothesis Test Is this statistically significant and practically important? Can we meet our improvement target? Done No Discard Yes No Yes

141 Module 5 Improve Phase

Brainstorming One of the most used methods for generation of project ideas. Brainstorming uses the power of synergy from discussion of ideas between a cross-functional team (team members build on others’ ideas). Two types of brainstorming Freeform: Each team member contributes ideas for a given problem independently (writing on sticky notes and sticking on a whiteboard) Structured: Ask each team member for ideas and write it on a whiteboard. Use multiple rounds if required. Team members can pass if they don’t have any additional ideas. Rules for effective brainstorming No criticism of ideas Go for large quantity of ideas Build on each other’s ideas Encourage wild and exaggerated ideas The more ideas the better

Generate solution ideas The reduction or elimination of the root causes from the last phase is the basis for the solutions which the team will generate. First step is to generate as many solutions as possible (more the better), later we will go ahead and pick the best solutions to implement. Start with the most important root cause that has the biggest impact on the problem statement and come up with all the ideas to address this issue. Continue this process until all the root causes are addressed. Quantity is important at this stage -> Even crazy ideas are fine! Root Cause 1 Root Cause 2 Root Cause 3 Idea 1 Idea 2 Idea N

Six thinking Hats Six Thinking Hats is a simple, effective parallel thinking process that helps people be more productive, focused, and mindfully involved. You and your team members can learn how to separate thinking into six clear functions and roles. Each thinking role is identified with a coloured symbolic "thinking hat." By mentally wearing and switching "hats," you can easily focus or redirect thoughts, the conversation, or the meeting. Six thinking hats helps teams stay focused on creative problem solving by avoiding negativity and group arguments

Improvement and Implementation plan Improvement Plan Most Important Plan Implementation plan helps answer the who, what, where, when, why, and how questions for the implementation as a whole. It can be used to communicate with key stakeholders on implementation plan and generate buy-in. Key elements of the plan: Updated business case (if changed) Communication plan (if required) Resource requirements (if required) Project plan (important element) Overview of pilots (if required) Use your implementation plan to improve the chance of success of your implementation. The implementation plan describes how we will implement the solution

Sample Improvement Plan

Sample Implementation Plan

Piloting the solution Piloting is the trial implementation of all or part of the proposed solution on a reduced scale The basic methodology behind a pilot is the PDCA cycle Pilots give us the opportunity to define cause-effect relationships of root cause and solution Experience the solution without committing the entire organization Better understand the impacts and obtain feedback from stakeholders Test the validity of the solution and increase organizational buy-in Validate and refine cost and benefit estimates Piloting the solution reduces risk by starting on a small-scale, correcting gaps, and up-scaling

6 Identifications of FMEA Identify Risk Identify Risk Priority Number Identify Mitigation Plan/Action Identify Responsible Person Identify Target Date of Closure Identify Re-evaluation of Risk QUANTIFYING THE RISK Risk Priority Number Severity of effects X Occurrence (Probability of failure) X Detection Likely-hood Risk Priority Number (RPN)

Sample FMEA Form Supplier: Potential Failure Mode and Effects Analysis Supplier Code : (Process FMEA) System : Sub System : Component : Drawing No.: Process Responsibility : Rev : FMEA Number : Model Year / Vehicle(s): Revision No.: Key Date : Rev Date : Page: OF Core team : Part No.: FMEA Date: Prepared by : Item Function Potential Failure Mode Potential Effect(s) of Failure SEVER CLASS Potential cause(s) Mechanism (s) of Failure OCCUR Current Process Controls Prevention Current Process Controls Detection DETEC PRN Recomm -ended Action(s) Responsibility & Target Completion Date Action Results Action Taken SEV OCC DET PRN

Before/After Scenarios Revised Process Capability Revised SIPOC

Steps in Designing An Experiment Define the Problem The mileage of my New Car is not up to advertised standards. I want to improve my car’s mileage State the hypothesis Some combination of Speed, Gas Octane and Tyre Pressure will provide me with the Optimum Gas Mileage. Identify the Dependent (Y) and independent (X) variables Dependent Variable = Gas Mileage (Y) Independent Variables = Tyre Pressure Octane Speed 152

Steps in Designing An Experiment 4. Determine the Test Levels for each independent variable Independent Variables ( Xs ) Factors Level (-) Level (+) Tyre Pressure 30 35 Octane 87 92 Speed (mph) 55 65 5. Calculate the number of trials to test all combinations of test levels Number of Test Levels (t0) Number of Factors = 2 to Power(3) = 8 Trials 153

Steps in Designing An Experiment 6. Construct the Experimental Trials Table Tyre Pressure Octane Speed Gas Milage - - - + - - - + - + + - - - + + - + - + + + + + 154

Steps in Designing An Experiment 7. Run the Experiment Tyre Pressure Octane Speed Gas Milage - (30) - (87) - (55) 26 + (35) - (87) - (55) 27 - (30) +(92) - (55) 30 + (35) + (92) - (55) 33 - (30) - (87) + (65) 18 + (35) - (87) + (65) 21 - (30) + (92) + (65) 19 + (35) + (92) + (65) 22 What effects (Main and Interaction) seem to be significant? Main effect is the effect of one X on the Y Interaction effects are the combined effects of two or more Xs on the Y 155

Steps in Designing An Experiment 8. Summarize the Data : 156 Tyre Pressure Octane Speed Gas Milage ABC A B C -30 -87 -55 26 - +35 -87 -55 27 + -30 +92 -55 30 + +35 +92 -55 33 - -30 -87 +65 18 + +35 -87 +65 21 - -30 +92 +65 19 - +35 +92 +65 22 + Factors + - Main Effect Interaction Effect Tyre 25.75 23.25 2.50 Octane 26.00 23.00 3.00 Speed 20.00 29.00 -9.00 ABC 24.25 24.75 -0.5 * At this stage always check the Absolute Values

Steps in Designing An Experiment 9. Draw Conclusions and Make Recommendations : Speed has an important effect on Gas Mileage Drive at 55 to get the Best Gas Mileage Tyre Pressure does not significantly affect Gas Mileage Set Tyre Pressure for best Tyre Wear Results Octane does not significantly affect Gas Mileage Buy the Octane that optimizes cost of engine cleanliness. There is no sufficient Interaction Effect between any factor(s). 157

Difference between Six Sigma and Kaizen

159 Module 6 Control Phase

Control Plan Control Plan / Improvement sustenance plan: Implement all actions mentioned in ‘Recommendation Plan’ Continue to monitor project and do data analysis for next 2 months 160

Control methods Each solution needs a control method to ensure that the output stays in control. The control method is added to the Control Plan that will be given to the process owner. Control methods Mistake-proofing Verbal instructions Standard operating procedures (SOP’s) Statistical process control (SPC) Employ the control methods that ensures the most effective control with the least effort – sometimes a combination is required. Control strategies differ in implementation effort and in effectiveness

What is mistake proofing ( Poka Yoke ) Poka-Yoke is a simple concept that means mistake proofing in Japanese. Poka-Yoke is any automatic device or method that either makes it impossible for an error to occur or makes the error immediately obvious once it has occurred Developed by Dr. Shigeo Shingo to achieve zero defects. Poka-Yoke does not necessarily involve use of expensive equipment or time-consuming process but uses the team’s ingenuity to avoid errors in the first place. This is a better form of control than SPC since mistakes don’t happen in the first place, there is no need to constantly monitor the process and take corrective action. Poka Yoke is the best form of control

Standard Operating Procedures (SOP) SOP consist of written instructions on how to setup and run a process. It can contain detailed process maps and/or standard operating conditions. They should be used in any process improvement effort. However, they are not sufficient : They may not be understood by the users They may be understood but not really followed They may not reflect current way of working People may change and don’t know about SOPs SOP’s are required but are not very effective form of control

How to create and use SOP SOPs should be created by the people who do the work and should reflect the best way of working of the entire team Use picture and storyboards rather than text to describe the process There should be a process for updating SOP’s as mistakes are corrected and improvements are suggested SOPs should be reviewed and approved by management. SOP’s should be verified for accuracy and edited SOP’s should be located wherever the process is conducted. It should possibly be updated on an annual (or regular basis) Clear ownership of the SOP (creation and revision) should be established Always include training and a way to check if the SOP is understood (such as periodic tests)

Statistical Process Control (SPC) SPC is a statistical approach to monitor and control any process and improve process performance over time SPC uses control charts for that purpose. The control charts contain two control limits (LCL and UCL). If the data points fall within the control limits, we conclude that the process is in control. If the data points fall outside, we investigate for special causes and take corrective action if required. SPC can be applied in any process in any area (manufacturing or service) SPC is better than documentation since we constantly monitor and know immediately if the process is not in control. SPC is used to track and control process stability

Procedure to implement Control Charts Identify what to measure & control Define sample size and frequency Determine rational subgroups (only common cause variations within group) Select the appropriate control chart Take initial set of data from the process & determine control limits Verify stability and if stable lock the control limits (if in control) Develop an out-of-control action plan (OCAP) Train the people involved Continue to monitor the process using control charts

Control Limits vs Specification Limits Control limits are derived from natural process variability, or the natural tolerance limits of a process Specification limits are determined externally, for Exercise by customers or designers There is no mathematical or statistical relationship between the control limits and the specification limits

Minitab Exercise on Statistical Process Control 168

Exercise : Problem Description American Paints wants to determine if the paint production process is in control. Quality of the paint production is critical to ensure highest level of customer satisfaction. In order to monitor the process, sampling is used and two metrics are measured and reported every 4 hours, the concentration levels (mol/L) and yield (%). Determine if the process is in control. + more records…

Control Charts With Assistant Click on Assistant > Control Charts > I-MR Chart Specify the Data Column As Paint Conc. Specify that we want to estimate the control limits from the data. Click on the OK button Without Assistant Click on Stats > Control Charts > Variable Charts for Individuals > I-MR Chart Since the data is continuous and the subgroup size is 1, we use an I-MR Control Chart

I-MR Control Chart All the data points are within the control limits (LCL and UCL) => process is stable and in-control. If any points were outside the control limits, we would have to investigate for special causes The MR Chart helps us in assessing the stability of the process caused by the variation between consecutive individual data points.

Exercise : Problem Description Prime Vision Tech wants to check if the number of defects made on a TV production line is in control. All TVs manufactured in a day are tested at the end of the line and the number of records are stored. There are multiple types of defects possible in the TV. Data was collected for a period of one month. + more records…

Control Charts With Assistant Click on Assistant > Control Charts > U Chart Specify the TV Defects as the defects column Specify the TV Produced as the subgroup sizes column Specify that we want to estimate the control limits from the data. Click on the OK button Without Assistant Click on Stats > Control Charts > Attribute Charts > U Chart Since the data is attribute (defects data), we use the U-Chart

U Control Chart All the data points are within the control limits (LCL and UCL) => process is stable and in-control. Note that U Char uses the Poisson distribution to estimate the control limits. If any points were outside the control limits, we would have to investigate for special causes

Exercise : Problem Description Handsome Tractors Ltd wants to check if the number of defective tractors manufactured by a company is in control. Each week we inspect the number or tractors manufactured and the number of defectives produced that week. A tractor is considered defective if it does not pass the critical items inspection test at the assembly inspection area. + more records…

Control Charts With Assistant Click on Assistant > Control Charts > P Chart Specify the Tractors Defective as the defects column Specify the Tractors Produced as the subgroup sizes column Specify that we want to estimate the control limits from the data. Click on the OK button Without Assistant Click on Stats > Control Charts > Attribute Charts > P Chart Since the data is attribute (defectives data), we use the P-Chart

P Control Chart All the data points are within the control limits (LCL and UCL) => process is stable and in-control. Note that P Char uses the Binomial distribution to estimate the control limits. If any points were outside the control limits, we would have to investigate for special causes

178 Module 7 Lean Management

Lean Six Sigma 179

Lean Manufacturing Lean Manufacturing is a system for maximizing product value for the customer while minimizing waste without sacrificing productivity. Lean manufacturing as we know it today has its roots in the Toyota Production System (TPS), but before it was known as TPS, they simply called it just-in-time manufacturing. There were 3 things the Toyota Production System attempted to prevent: Muda – Everything in your manufacturing process that creates waste or causes constraints on creating a valuable product. Mura – Everything that creates inconsistent and inefficient work flows. Muri – All tasks or loads that put too much stress on your employees or machines. 180

Lean Manufacturing In Muda, there are 8 wastes you should work to eliminate: Defects, Overproduction Waiting Not utilizing talent Transportation Inventory excess Motion waste Excess processing What is Jidoka? Jidoka is a principle implemented in lean manufacturing where machines automatically stop working upon detecting an abnormal condition and operators try fixing the defect to prevent recurrence of the issue. 181

Lean Generic Principles There were also 5 principles that every Lean manufacturing system adhered to: Value – A company delivers the most valuable product to the customer. Value Stream – Map out the steps and processes required to manufacture those valuable products. Flow – Undergo the process of ensuring all of your value-adding steps flow smoothly without interruptions, delays, or bottlenecks. Pull – Products are built on a “just-in-time” basis so that materials aren’t stock piled and customers receive their orders within weeks, instead of months. Perfection – Make Lean thinking and process improvement a core part of your company culture. 182

Based on Japanese words that begin with ‘S’, the 6S Philosophy focuses on effective workplace organization and standardized work procedures. This is the focus of every Lean Initiative. 6S simplifies our work environment, reduces waste and non-value activity while improving quality, efficiency and safety. Let's understand these – Sort – ( Seiri ) the first S focuses on eliminating unnecessary items from the workplace. An effective visual method to identify these unneeded items is called tagging. Sorting is an excellent way to free up valuable floor space and eliminate such things as broken tools, obsolete jigs and fixtures, scrap and excess raw material. The Sort process also helps prevent the JIC job mentality (Just In Case.) In an IT organization, periodic deletion or archival of unwanted documents or mails; appropriate naming conventions & folder organization of documents can make our workplace effective. 183 Lean 6S Philosophy

Set In Order (Seiton) is the second of the 6Ss and focuses on efficient and effective storage methods. "A place for everything and everything in its place." Exercise of effective Seiton would be Configuration Mgmt system which helps arrange Configuration items for easy retrieval & version control Shine: (Seiso) Once you have eliminated the clutter and junk that has been clogging your work areas and identified and located the necessary items, the next step is to thoroughly clean the work area. Daily follow-up cleaning is necessary in order to sustain this improvement. People take pride in a clean and clutter-free work area and the Shine step will help create ownership in the equipment and facility. Standardize: (Seiketsu) Once the first three 6S’s have been implemented, you should concentrate on standardizing best practice in your work area. Allow your employees to participate in the development of such standards. They are a valuable but often overlooked source of information regarding their work. Think of what McDonalds, Pizza Hut, UPS, Blockbuster and the United States Military would be without effective work standards. 184 Lean 6S Philosophy

Sustain: (Shitsuke) This is by far the most difficult S to implement and achieve i.e. Discipline. Human nature is to resist change. More than a few organizations have found themselves with a dirty cluttered workspace a few months following their attempt to implement 6S. The tendency is to return to the status quo and the comfort zone of the "old way" of doing things. Sustain focuses on defining a new status quo and standard of work place organization. Safety ( Anzen ) : Safety First. Once fully implemented, the 6S process can increase morale, create positive impressions on customers, and increase efficiency and organization. Not only will employees feel better about where they work, the effect on continuous improvement can lead to less waste, better quality and faster lead times. Any of which will make your organization more profitable and competitive in the market place. 185 Lean 6S Philosophy

Lean and Six sigma complement each other perfectly. While Six Sigma employs tools, Lean applies principles. Six sigma tools can be used independent of one another, whereas lean principles are best used together. Lean focuses on eliminating waste and Six Sigma is concerned with eliminating variation. Both have the common goal of making a process more efficient and effective. We all have to remember that the Six Sigma and/or Lean tour is not an One-Night Journey but would evolve for time to come and finally come out as a great success. For this it needs support from the Sr. Management and confidence in people, so that the result can be achieved. 186 Relation of Lean with Six Sigma

Dedicate Black Belt and Leadership Team full time Select the Best people for these roles Select the projects that are most important to the business Get and publicize results Leaders must articulate what Lean Six Sigma means clearly, simply and frequently. Involve employees in all aspects of transformational change Recognize and reward positive change behaviours and skills Integrate leadership training Launch Six Sigma by leveraging existing organizational strengths Plan and enforce a focus on transactional processes Reinforce existing corporate values ( eg , integrity, customer focus etc) during implementation 187 Institutionalizing Lean Six Sigma

Integrate Lean Six Sigma into existing business unit strategic planning sessions, operational reviews and management team meetings Create accountability through visibility Proactively plan for Lean Six Sigma communication events Share best practices and lessons learned across the business Enforce a common language 188 Institutionalizing Lean Six Sigma (Contd.)

189 Case Studies

190

191

192 Module 8 Project Closure

Project Closure Actions 193 Ensure Good Controls Do not close the project until you have deployed appropriate control methods to ensure sustenance of the solutions and process. Control of the solutions is probably one of the hardest parts of the DMAIC project. Controller Sign-off Capture all the expected savings from the project and have a close-out meeting with the financial controller and get their sign-off on any savings claimed on the project. Process Owner Handoff Complete any training of the people who will be running the process from this point forward (process owner & his/her team) Have a final closure meeting with the process owner. Discuss improvements made, critical X’s, control plans, control methods. The solution must be understood and accepted by the process owner.

Project Closure Actions 194 Leverage Project Investigate all the other areas of the company where the same or similar solutions could be useful and share the report and findings with the appropriate stakeholders Document the project findings in a central repository that can be searched by other project managers in case they want to leverage your findings. Lessons Learned Capture the learnings from the project (what went well and what could be improved) and share with other stakeholders Communicate lessons learned and project findings with extended audiences using newsletter, email, website update. Record the Project Archive and save all your data, analysis, reports into one central location that can be accessed by other people in the company if required

Project Closure Actions 195 Sponsor Closure Have a formal meeting with the project sponsor and share all the project findings and reports and get a formal sign-off of the project closure. If required, setup a meeting with the Six Sigma community (MBB, Champion) to share project findings for any certification requirements. Celebration Thank your team members for their contribution and recognize their efforts and contributions using rewards and recognition as appropriate.

A Sample Project Template 196

Project Title 197

Define Phase Identification of Voice of the Customer (VOC) Identification of Critical To Quality (CTQ) Identification of Specification Limits Relation Between VOC > CTQ > USL and LSL Project Charter SIPOC Analysis Tollgate Process/Checklist/Review 198

Define Phase – Sample Project Charter 199

Define Phase – Sample SIPOC 200

Measure Phase Detailed Process Mapping Data Collection Plan Process Capability Measurement System Analysis Probability Distribution Tollgate Process/Checklist/Review 201

Measure Phase 202

Analyze Phase Brainstorming Results Cause & Effect Analysis Why Why Analysis Pareto Analysis Hypothesis Testing Correlation Analysis Regression Analysis Tollgate Process/Checklist/Review 203

Analyze Phase 204

Improve Phase Improvement Plan Implementation Plan Risk Management Plan Solution Selection Matrix *** 7 Management Tools *** Porter’s Five Forces *** Design of Experiments Before and After Scenario Tollgate Process 205 *** for BB

Control Phase Control Plan and Improvement Sustenance Plan Control Methods Statistical Process Control (SPC) Project Learning Reporting Project Closure Formalities Final Closure with Vote of Thanks Tollgate Process/Checklist/Review 206

Lean Lean Manufacturing Checks and Implementation Lean Generic Principles Checks and Implementation Lean 6S Philosophy Checks and Implementation Identify the 3 Lean Gs *** 207 *** for BB

208 Green Belt Project Assessment

The Scenario : “ I don’t like to wait a long time on the phone to get my questions answered ” – BPO/KPO Service related Concern towards Response. Currently the End Customers are calling the Backoffice functions with questions but getting connected to a representative itself is taking almost >=15 minutes on an average which are making them very uncomfortable. The SLA that has been always stated is <=5 minutes. After that the original question is getting resolved in their usual normal time. This is making the Customers extremely dissatisfied. Project Assessment # 1 Define Phase Customer Type, CTQ and Specification Limits Project Charter : Problem Statement and Goal Statement. Measure Phase Original Process Capability with Minitab Analyze Phase Cause & Effect Root Cause Analysis (5) Improve Phase Improvement Plan (5) Revised Process Capability with Minitab Control Phase Improvement Sustenance Plan (5)

210 Green Belt Modular Questions

Question Why is data important in Six Sigma process? Six Sigma can be strictly applied only when data is available Without data decision making process is difficult Dr. Harry defined Six Sigma with data, so we use it Statistics need data and Six Sigma needs statistics 211 Answer : B

Question VOC can be achieved through various means. It is important to get a clarity on how our business is run and gather requirements. VOC is related to CTQ and USL & LSL? True False 212 Answer : A

Question Why calculate Standard Deviation for Six Sigma? To calculate Variations from the Mean To calculate Range To calculate Problems in the Process All the above 213 Answer : A

Question Central Tendency is? Standard Deviation & Mean Mean, Median & Mode Mean, Median & Standard Deviation Mean & Range 214 Answer : B

What is a process ? It is a set of activities done to construct a building A collection of steps done in an order, to result in a product or result A product/ result-oriented approach It is a temporary endeavor to produce unique result Question 215 Answer : B

SIPOC stands for? Supplier, Input, Product, Output and Customer Sales, Internal, Policy, Outcome and Customer Supplier, Input, Process, Outline and Customer Supplier, Input, Process, Output and Customer Question 216 Answer : D

Major components of process is ? Supplier Customer Events Start & End points Question 217 Answer : B

Who is termed as a process owner? Sponsor A Functional Manager Green Belt Master Black Belt Question 218 Answer : A

The Analyze phase of the DMAIC the process has the characteristic of: identifying the problem statement and creating a project charter identifying the voice of the customer establishing relationships between the x's and y’s collecting data and checking if the current process is capable of meeting customer specifications Question 219 Answer : C

Which are the Twin Statements and are must to be included in Project Charter? VOC & CTQ Problem and Goal Statement Scope In and Scope Out Question 220 Answer : B

The Full Data and part of the Data are called. Population and Sample Continuous and Discrete Sequential and Random Binary and Sequential Question 221 Answer : A

Gage R & R talks about. Repeatability and Reproducibility Adjustment to Data Measurement Fault Finding towards variation of data Calculation towards data accuracy Question 222 Answer : A

In the Root Cause Management in Analyze Phase, important components are Brainstorming, Cause & Effect and Why Why Cause & Effect and Why Why Brainstorming Technique and Hypothesis Testing Correlation and Regression Analysis Question 223 Answer : A

Hypothesis Testing we start with identification of Null Hypothesis and conclude whether to reject or accept Null Hypothesis. True False Question 224 Answer : A

Brainstorming Session in Improve Phase works on. Explore Root Causes Explore Solution Ideas Explore dependencies Explore relationships between data sets Question 225 Answer : B

Control Phase deals with majorly two things. Control Plan and Control Methods Before and After process capabilities Improvement and Implementation Plans Reducing variations from the process Question 226 Answer : A

Process Capability for Continuous and Discrete could be determined by. Cpk and Dpmo methods respectively Cp and Dpu methods respectively Standard Deviation and Mean methods respectively Mode and Median methods respectively Question 227 Answer : A

Six Sigma is about Quality, while Lean is about: Utilizing “belts” Improving processes using statistical techniques A thorough analysis of complex problems Increasing speed and reducing waste Question 228 Answer : D

Continuous Improvement using Six Sigma methodologies require the use of the process. DMADV BMAIV IDOV DMAIC Question 229 Answer : D

Bar Graph works for. Continuous Data Discrete Data Both Continuous and Discrete Data Normal and Non Parametric Data Question 230 Answer : B

In order to calculate Dpmo for Discrete Process Capability, the following ingredients are necessary. Defects, Units, Opportunities Defects, Head count of people Units, Opportunities, USL and LSL Opportunities, CTQ, VOC Question 231 Answer : A

Probability distribution talks about opportunity of Chance of certain incident happening based on a given situation. True False Question 232 Answer : A

In Correlation Analysis. We find out the mutual and co relationship between two sets of data and if positive, negative or no correlation exists We find out the dependency of one data on another We find out if there is significant difference between one data set and a target We find out if two data sets are different from each other Question 233 Answer : A

Regression Analysis talks about. Dependency of X data on Y Data How much Y Data changes with X Data changes Relationship between X and Y data How much X Data changes with Y Data variability Both A and B Question 234 Answer : E

For Regression, we check for P Value and also R Square Values for our result. True False Question 235 Answer : A

What is the key concept of the Six Sigma core philosophy? Six Sigma is a business improvement approach that seeks to perform the “6S"s Six Sigma is a business improvement approach that seeks to find and eliminate causes of defects and errors Six Sigma is a cost-cutting approach that seeks to identify and eliminate unnecessary workers Six Sigma is a revenue-generating an approach that seeks to find ways to produce in greater quantity Question 236 Answer : B

For Normality Testing we consider P value for a data to be normal only if. P Value > 0.05 P Value <= 0.05 P Value >=95% P Value > 0.75 Question 237 Answer : A

The Sigma Shift of 1.5 Sigma is basically the difference between. Zst and Zlt Cpk and Ppk Cpk and Dmpo Cpk and Cp Question 238 Answer : A

In a Hypothesis Testing, we try to identify. Significant difference between data sets Mutual relationship between data sets Dependencies between data sets None of the above Question 239 Answer : A

In Hypothesis Testing, we start off with identification of. Alternate Hypothesis Alpha Error Beta Error Null Hypothesis Question 240 Answer : D

Control limits are derived from natural process variability, or the natural tolerance limits of a process True False Question 241 Answer : A

Between Specification Limits and Control Limits. There is only mathematical relationship between them There is only statistical relationship between them There is no statistical or mathematical relationship between them Both are very much related to each other Question 242 Answer : C

Control Charts could be drawn for. Both normal and non normal data Population and Sample data Continuous and Discrete data Only data which has specification limits Question 243 Answer : C

P and U Control Charts are for. Continuous Data Normal Data Non Parametric Data Discrete Data Question 244 Answer : D

X Bar R Chart and I MR Chart are for. Continuous Data Normal Data Non Parametric Data Discrete Data Question 245 Answer : A

246 Thanks For attending Lean Six Sigma Green Belt Training

247 Lean Six Sigma Black Belt Coverage ….

248 Module 1 Concepts

Training perspective Project perspective Role perspective Green Belt and Black Belt 249

Balanced Scorecard 257 The Balanced Scorecard (BSC) w as originally developed by Dr. rt Kaplan of Harvard University and Dr. David Norton as a framework for measuring organizational performance using a more BALANCED set of performance measures. Traditionally companies used only short-term financial performance as measure of success. The “balanced scorecard” added additional non-financial strategic measures to the mix in order to better focus on long-term success. The system has evolved over the years and is now considered a fully integrated strategic management system. It is suggested to have up to 20 measures across all of these metrics, so that the management can focus on what is important. The chosen metrics should reflect the strategy of the company ( e.g. a new company may want to focus on revenue from new products / customers acquisition while a mature company may focus on revenue. Each metric is tracked usually on a monthly basis (At least) and reviewed by management. If a metric is in Red (behind target), It can be one of the potential Lean Six Sigma project.

Balanced Scorecard 258 * Adapted from Robert S. Kalpan and David P. Norton, Using the Balanced Scorecard as a Strategic Management System”, Harvard Busin

Balanced Scorecard 259

Change Management 260

Change Management 261

Kotter's 8-Step Change Model 262

Kotter's 8-Step Change Model 263

264 Module 3 Measure Phase

Minitab Exercise on Gage R & R 265

Exercise: Gage R & R– Continuous Data 266 G i v e n t h e da ta b e l o w f o r r ea d i n g b y 3 a p p r a i s e r s o n 6 c a ll s w i th 2 t r a il s . D etermine whether the measurement system is acceptable . Call ID Appraiser A Appraiser B Appraiser C 1 2 1 2 1 2 1 65 60 55 55 50 55 2 100 100 100 95 100 100 3 85 80 80 75 80 80 4 85 95 80 75 80 80 5 55 45 40 40 45 50 6 100 100 100 100 100 100

Step 1: Copy the data to Minitab worksheet as shown below Exercise: Gage R & R– Continuous Data 267

Step 2: Choose Gage R&R Study (Crossed) from Stat Menu as shown below: Exercise: Gage R & R– Continuous Data 268

Step 3: Enter Part Numbers, Operators & Measurement Data. Choose Xbar and R as shown below Click “OK” button 269 Exercise: Gage R & R– Continuous Data

Step 4: Minitab will give the following Output Source Var Comp % Contribution Total Gage R & R 17.434 4.06 Repeatability 7.338 1.71 Reproducibility 10.096 2.35 Part-To-Part 4 1 1 . 5 6 8 95.94 Total 42 9 . 2 100 Source StdDev (SD) (6 * SD) (%SV) Total Gage R & R 4.1754 25.052 2 . 1 6 Repeatability 2.7088 16.253 1 3 . 6 Reproducibility 3.1774 19.065 1 5 . 3 4 Part-To-Part 20.2871 1 2 1 . 7 2 3 9 7 . 9 5 Total Variation 20.7154 1 2 4 . 2 7 4 100 If < 1 %. Gage acceptable Else if > 30 %, Gage not acceptable 270 Exercise: Gage R & R– Continuous Data

An Project Manager considered 10 transaction and chosen 2 appraisers at random for Gage R&R study. The transactions were evaluated on “Correct” or “Incorrect” basis. For all the 10 transactions actual results (Standard) are also available with TL. 2 appraisers processed each transaction twice within gap of one week. The results are as follows. Study the Gage R&R. Gage R&R for Discrete Data 271 TR A N S A CT ION NUMBER STANDARD RADHA KRISHNA TRIAL 1 TRIAL 2 TRIAL 1 TRIAL 2 1 CORRECT CORRECT CORRECT CORRECT INCORRECT 2 INCORRECT CORRECT INCORRECT INCORRECT INCORRECT 3 INCORRECT INCORRECT INCORRECT INCORRECT INCORRECT 4 CORRECT INCORRECT INCORRECT CORRECT CORRECT 5 INCORRECT INCORRECT INCORRECT INCORRECT CORRECT 6 CORRECT CORRECT CORRECT CORRECT CORRECT 7 CORRECT CORRECT CORRECT CORRECT CORRECT 8 CORRECT INCORRECT CORRECT CORRECT CORRECT 9 INCORRECT INCORRECT CORRECT INCORRECT INCORRECT 10 CORRECT CORRECT CORRECT CORRECT CORRECT

Enter the data in Minitab worksheet. Test1 and test 2 results of same appraiser should be at one place, as shown Gage R&R for Discrete Data 272

Select Stat > Quality Tools > Attribute Agreement Analysis Gage R&R for Discrete Data- Analyze Results 273

Click on columns as shown E n t e r 2, 2 , Radha, Krishna Click on columns as shown Gage R&R for Discrete Data- Analyze Results 274

Percent Repeatable by appraiser (it should be >=80%) Gage R&R for Discrete Data 275

Repeatability Vs Standard 276 Gage R&R for Discrete Data

Percentage Reproducibility For all appraisers (it should be >=80%) Percentage Reproducibility For all appraisers Vs Standard (it should be >=80%) Since R&R is less than 80%, root causes to be identified and corrective actions to be taken. Re-conduct study to assess the improvement. Gage R&R for Discrete Data 277

Six Sigma Table 278

Defectives Process Capability (Binomial Capability) 279 The number of defective tractors produced have been collected for a period of 1 year (52 weeks of data). The data is stored in columns: Tractors Produced and Tractors Defective. The maximum percentage defective we are willing to accept is 5%. Determine the capability of this process. Defectives data follow a binomial distribution

Defectives Process Capability (Binomial Capability) 280 With Minitab Assistant Click on Assistant > Capability Analysis > Binomial Capability Enter the Tractors Defective as the defective column Specify the Tractors Produced as the subgroup sizes column Specify the maximum %defective as 5% Process Sigma Level is around 3.70 and DPMO is 135 73 Defectives data follow a binomial distribution

Determining Sigma Level for Data 281

282 Module 4 Analyze Phase

Exercise : Problem Description Hardening of the arteries is a major contributor to ill health. It progresses by the accumulation of plaque on the interior walls of arteries that obstructs the blood flow. This obstruction causes pressure differences within the blood flow and eventually the artery collapses with possibly fatal consequences to the patients. Lifespan Pharma has built a model to investigate the collapsibility of the arteries. The R&D Team has investigated three different levels of stenosis and determined the flowrates in ml/s when artery collapses. Determine if these differences are statistically significant. + more records…

One-Way ANOVA Click on Assistant > Hypothesis Tests > One-Way ANOVA Specify that the Y data for each X are in separate columns and select the columns (Stenonis1-Stenosis3) Select an alpha risk of 0.05 (95% confidence) Leave the difference blank Click on the OK button Ho: Flow rates are the same for all Stenosis. Ha: At least one flow rate is different

One-Way ANOVA Sample size is sufficient to detect a difference of 3.4 or greater P value of less than 0.001 indicates that we reject the null hypothesis. Looking at the confidence intervals, it is clear they do not overlap. Hence, the stenosis flowrates are different.

Two-Way ANOVA Analysis with Example For example, suppose a botanist wants to explore how sunlight exposure and watering frequency affect plant growth. She plants 40 seeds and lets them grow for two months under different conditions for sunlight exposure and watering frequency. After two months, she records the height of each plant. In this case, we have the following variables: Response variable: plant growth Factors: sunlight exposure, watering frequency And we would like to answer the following questions: Does sunlight exposure affect plant growth? Does watering frequency affect plant growth? Is there an interaction effect between sunlight exposure and watering frequency? (e.g. the effect that sunlight exposure has on the plants is dependent on watering frequency) We would use a two-way ANOVA for this analysis because we have two factors. If instead we wanted to know how only watering frequency affected plant growth, we would use a one-way ANOVA since we would only be working with one factor.

Exercise : Problem Description Riverscape Garments wants to determine differences in shopping behavior of men vs. women during normal periods, holiday period, and sale periods. A survey was conducted on how many men and women made purchases during each of these periods. Determine if there is any difference in the shopping behavior of the two genders.

Chi-Square Test for Association Click on Assistant > Hypothesis Tests > Chi-Square Test for Association Specify that you will enter data in a table below and select that outcomes are in columns Y name is Gender with 2 outcomes X name is Period with 3 values Enter the details of the observations in the table. Select an alpha risk of 0.05 (95% confidence) Click on the OK button Ho: proportion for all groups are equal. Ha: proportions are not the same across groups

Chi-Square Test for Association P value of 0.005 (Low) indicates that we reject the null hypothesis.

Minitab Exercise on Hypothesis Testing (Non-Parametric) 290

Non Parametric > 2 population > Mann-Whitney Test 291 One of American State Municipality uses two brands of paint for painting stripes on roads. A National H ighway official wants to know whether the durability of the two brands of paint are different. For each paint, the official records the number of months the paint persists on the highway. Can you help the highway official with the analysis ? + more records…

Non Parametric : 2 population : Mann- Whitney Test 292 Check the Normality of Data : Minitab : Stat > Basic Statistics > Graphical Summary Brand A : p-value is 0.007 so the Data is Not Normal. Brand B : p-value is 0.236 so the Data is Normal.

Non Parametric : 2 population : Mann-Whitney Test 293 Check the Normality of Data : Minitab : Stat > Nonparametrics > Mann-Whitney Interpret the results : The null hypothesis states that the difference in the median number of months that the paint persists between the two brands is 0. Because the p-value is 0.009, which is less than the significance level of 0.05, so we reject the null hypothesis . We can also conclude that the difference in the median number of months the paints persists between the two brands is not 0. The 95.5 Percent CI indicates that the population median of Brand B is likely to be greater than Brand A Ho : Medians are the same Ha : Medians are not the same

Non Parametric : More than 2 data sets : Kruskal-Wallis Test 294 A Nuclear S cientist wants to determine whether temperature changes in the ocean near a nuclear power plant affect the growth of fish. The scientist randomly divides 25 newly hatched fish into four groups and places each group into a separate, simulated ocean environment. The simulated environments are identical except fo r temperature. Six months later, the scientist measures the weights of the fish. Please conduct the below Statistical Tests and suggest if nuclear powe r p lant affects the growth of fish : Kruskal-Wallis Test

Non Parametric > More than 2 data sets > Kruskal-Wallis Test 295 p- value is > 0.05 . Accept Null Hypothesis.

Advanced Regression Analysis - Logistic Regression 296 Linear Regression and Logistic Regression, both the models are used for predictions. The Differences between Linear Regression and Logistic Regression Linear and Logistic regression are the most basic form of regression which are commonly used. The essential difference between these two is that Logistic regression is used when the dependent variable is binary in nature. In contrast, Linear regression is used when the dependent variable is continuous and nature of the regression line is linear.

Advanced Regression Analysis - Logistic Regression > Types 297 There are 3 Types of Logistic Regression Binary Logistic Regression – For Binary Response variable Ordinal Logistic Regression –For response with three or more outcomes that have an order Nominal Logistic Regression - For response with three or more outcomes that do not have an order Binary Logistic Regression Binary Logistic Regression is used to display the relationship between one continuous predictor and a binary response. A binary response has two outcomes, such as pass and fail. Example : A medical researcher wants to examine the relationship between the age that a patient suffers a myocardial infarction and whether the patient has a favourable or an unfavourable outcome after treatment. Because the response is a binary variable and there is only one predictor, the researcher uses a simple binary logistic regression.

Advanced Regression Analysis - Logistic Regression > Types 298 Ordinal logistic regression Ordinal Logistic Regression is used to model the relationship between a set of predictors and an ordinal response. An ordinal response has three or more outcomes that have an order , such as low, medium, and high. You can include interaction and polynomial terms, nest terms within other terms, and fit different link functions. Example: A field biologist studies the survival time of salamanders and wants to determine whether survival is related to region and the level of water toxicity. The biologist divides survival time into three categories: less than 15 days, 16 to 30 days, and over 30 days. Nominal logistic regression Nominal Logistic Regression is used to model the relationship between a set of predictors and a nominal response. A nominal response has three or more outcomes that do not have an order , such as a scratch, dent, and tear. Example : A school administrator want to investigate the variables that affect a student's preference for certain classes. The administrator uses nominal logistic regression to determine whether a student's age and the teaching method for a class is related to class preference.

Advanced Regression Analysis - Logistic Regression > Example 299 A marketing consultant for a Revive Popcorn company investigates the effectiveness of a TV advertisement for a new Popcorn product. The consultant shows the advertisement in a specific community for one week. Then the consultant randomly samples adults as they leave a local supermarket to ask whether they saw the advertisements and bought the new Popcorn. The consultant also asks adults what their annual household income is. Because the response is binary, the consultant uses binary logistic regression to determine how the advertisement and income are related to whether or not the adults sampled bought the popcorn. + more records…

300 Module 5 Improve Phase

The Seven Management tools belong to Operations Research & Japan’s Total Quality Management Philosophy and found their use first in Second World war. They are a set of tools and techniques used for planning and managing any type of operations effectively. The main purpose of these tools was to guide the managers in planning, analysis and decision making. These tools were invented separately by different people for various purposes but were organized and clubbed together to achieve efficient planning and management of operations. The list below discusses the Seven Management and Planning tools: Affinity Diagram: It is used to organize many ideas or decision criteria into groups based on their underlying relationships and affinity (likeness). 7 Management and Planning Tools

Relationship Diagram: This is also called as Interrelationship diagram. It is used to signify the strength of relationship between two processes or entities. Tree Diagram: This helps in understanding the process level-by-level by breaking down complex processes to the minute level of detail. Prioritization Matrix: This is a tool that helps in prioritizing an option over others, given a set of decision-making criteria. 7 Management and Planning Tools

Matrix Diagram: This is another tool that helps in establishing relationship between variables. Process Decision Program Chart (PDPC): This tool is based on the Decision tree but has other additional features like mapping the failure mode, risk involved, effect of failure with each decision node. Activity Network Diagram (AND): This is also called as Arrow Diagram and is a tool used by Project Management professionals to map their activities and sequential tasks in a visual format to understand and optimize the project duration. 7 Management and Planning Tools

Theory of Constraints (TOC) 304 Identify the system’s constraints. A system constraint limits the firm from achieving its performance and goals. Thus, constraints must be identified and prioritized for impact. Decide how to exploit the system’s constraints. How to manage the constraint to achieve full potential of the organization? Subordinate everything else to the above decisions. Constraints may have a limit, so look for ways to reduce the effects of a constraint, improve Capability or look to expand the capacities of the constraints. Elevate the system’s constraints. Try to eliminate the problem of the constraints. Strive to keep improving the system. Back to step1. After the constraint has been broken, go back to step one and look for new constraints.

Beginning in 1946 and still evolving, TRIZ was developed by the Soviet I nventor Genrich Altshuller and his colleagues. TRIZ in Russian = Teoriya Resheniya Izobretatelskikh Zadatch or in English, the theory of Inventive Problem Solving According to TRIZ, universal principles of creativity form the basis of innovation. TRIZ identifies and codifies these principles, and uses them to make the creative process more predictable. I t is said that traditional inventing is “Trial and error” resulting in much wasted time, effort and resources. Another approach is using common sense, logic and Basic Sciences in problem solving. Technical evolution and invention have certain patterns. One should be knowledgeable with them to solve technical problems. TRIZ ( Theory of Inventive Techniques) 305

TRIZ ( Theory of Inventive Techniques)

SCAMPER Technique 307 Literally, the word SCAMPER means “to run playfully about as a child.” But Scamper is also a creative brainstorming technique to help push thinking. It’s a tool that helps people to generate ideas, forcing them to look from different perspectives.

SCAMPER Technique 308  The SCAMPER technique uses a set of directed questions which you answer in order come up with new ideas.  SCAMPER is an acronym which stands for questions relating to the following: S Substitute Think about substituting part of your product/process for something else. By looking for something to substitute you can often come up with new ideas. C Combine Think about combining two or more parts of your probortunity to achieve a different product/process or to enhance synergy. A Adapt Think about which parts of the product/process could be adapted to remove the probortunity or think how you could change the nature of the product/process. M Modify / Magnify / Minify Think about changing part or all of the current situation, or to distort it in an unusual way. By forcing yourself to come up with new ways of working, you are often prompted into an alternative product/process.

SCAMPER Technique 309 P Put to another Use Think of how you might be able to put your current solution/ product/process to other purposes, or think of what you could reuse from somewhere else in order to solve your own probortunity. You might think of another way of solving your own probortunity or finding another market for your product. E Eliminate Think of what might happen if you eliminated various parts of the product/process/probortunity and consider what you might do in that situation. This often leads you to consider different ways of tackling the probortunity. R Re-arrange / Reverse Think of what you would do if part of your probortunity/product/process worked in reverse or done in a different order. What would you do if you had to do it in reverse? You can use this to see your probortunity from different angles and come up with new ideas.

SMED (Single-Minute Exchange of Dies) How long does it take to change the tire??? 310

Solution Selection Matrix The solution selection matrix uses facts and data to select solutions rather than opinions and assumptions. It is more of a structured approach to select your best solution(s). It may be difficult to quantify and use this approach if there are a lot of solutions, so first use multi-voting and/or pairwise ranking and narrow down the list to a manageable number before using this approach. Use the following criteria to select your solutions Sigma – the effect the solution on the sigma goal Timing – the time required to implement the solution Cost – the financial impacts associated with each solution Other – factors not considered above (morale, risk etc.) Rate each solution based on the above criteria and use the overall score to help select your solution and sell this solution to your management. Criteria Low (0) High (10) Sigma Impact Proposed solution does not impact the primary metric (0% impact) Proposed solution has a very large impact on primary metric (>100% impact) Time Impact Proposed solution takes too long to implement (> 1 year) Proposed solution can be quickly implemented (< 1 week) Cost Impact Proposed solution is very expensive to implement (> $10K) Proposed solution can be implemented with no additional costs (~ $0K) Other Impacts Proposed solution has a very high level of risk (>= 9) Proposed solution has very little risk (<= 1) Discuss with your team and come up with your own decision criteria to evaluate solution you are addressing. Each company will have its own range of allowable costs & time to implement.

Minitab Exercise on Design of Experiments 312

Design of Experiments - Exercise 313

Design of Experiments - Exercise 314

Design of Experiments - Exercise 315

Design of Experiments - Exercise 316

Design of Experiments - Exercise 317

Design of Experiments - Exercise 318

Design of Experiments - Exercise 319

Design of Experiments - Exercise 320

Porter’s Five Forces Porter's Five Forces is a simple but powerful tool for understanding the competitiveness of your business environment, and for identifying your strategy's potential profitability. This is useful, because, when you understand the forces in your environment or industry that can affect your profitability, you'll be able to adjust your strategy accordingly. For example, you could take advantage of a strong position or improve a weak one and avoid taking wrong steps in future.

322 Module 6 Control Phase

Exercise : Problem Description Vigilant Manufacturing Company needs to determine if the outside diameter of a PVC pipe that our company manufactures is in control. We use a digital calipers to measure the outside diameter of the pipe on one end and then repeat the measurement at the other end. We then capture the inside diameter at both ends. So, in total we have 4 readings for one pipe. + more records…

Control Charts With Assistant Click on Assistant > Control Charts > Xbar -R Chart Specify the Diameter as the Data Column Specify the Dia Group as the subgroup IDs column Specify that we want to estimate the control limits from the data. Click on the OK button Without Assistant Click on Stats > Control Charts > Variable Charts for Individuals > Xbar -R Chart Since the data is continuous and the subgroup size is 5, we use an Xbar -R Control Chart

Xbar -R Control Chart One data points is below the LCL => process is not stable (need to investigate reasons for this departure). If special causes found, deploy actions to address this special cause to bring process back in control. If any points were outside the control limits, we would have to investigate for special causes

326 Module 7 Lean Management

3G of Lean These are all Japanese words and its simple meaning are : Gemba – The real work place – Where actual work is performed. Gembutsu – The actual product – The actual work done in facility on product. Genjitsu – The actual fact – Find the facts for given situation. The 3-G principles are useful for continuous improvement. Its main objective is to involve managers to visit shop floor, take a look at actual products and gather maximum facts related to it. 327

328 Black Belt Project Assessment

Scenario Improving Customer Satisfaction Rating . In XYZ Pvt Ltd, currently few of Customers seems to be not happy with the performance and for last few times (Monthly Ratings) giving a Low Satisfaction Rating of <=9 in a range of 1-10. This is making the Project Head and Internal Management extremely unhappy and efforts are being made in discussion with the Customer to improve the rating subsequently. Some initial discussion has been done with the Customers and efforts are in place to make them satisfied eventually. The expectation is always to get a Customer Satisfaction Rating always between 9-10. Project Assessment # 2 Define Phase Customer Type, CTQ and Specification Limits Project Charter : Problem Statement and Goal Statement. Measure Phase Original Process Capability manually Analyze Phase Cause & Effect Root Cause Analysis (5) Improve Phase Improvement Plan (5) Revised Process Capability manually Control Phase Improvement Sustenance Plan (5)

330 Black Belt Modular Questions

The Analyze phase of the DMAIC the process has the characteristic of: identifying the problem statement and creating a project charter identifying the voice of the customer establishing relationships between the x's and y’s collecting data and checking if the current process is capable of meeting customer specifications Question 331 Answer : C

Measurement System Analysis talks about. Calculation Variation Common Cause Variation Special Cause Variation Measurement Variation Question 332 Answer : D

In Gage R & R. Repeatability means Same Person, Same Data, Same Measurement Unit and how much is the Variation Repeatability means Same Person, Different Data, Same Measurement Unit and how much is the Variation Both are True Both are False Question 333 Answer : A

In case of Continuous Gage R & R. The result shows about Variation % The result shows about Accuracy % The result shows about both Variation and Accuracy % None of the above. Question 334 Answer : A

In case of Discrete Gage R & R. The result shows about Variation % The result shows about Accuracy % The result shows about both Variation and Accuracy % None of the above. All the above Question 335 Answer : B

In Gage R & R generally there are 2 methods of measurement. Agile and Normal Crossed and Nested Parametric and Non-Parametric None of the above Question 336 Answer : B

In Gage R & R. Crossed Measurement refers to measurement of the same data. Nested Measurement refers to measurement of data at different intervals. Both A & B are True Both A & B are False Question 337 Answer : C

What is a process ? It is a set of activities done to construct a building A collection of steps done in an order, to result in a product or result A product/ result-oriented approach It is a temporary endeavor to produce unique result Question 338 Answer : B

In the Root Cause Management in Analyze Phase, important components are Brainstorming, Cause & Effect and Why Why Cause & Effect and Why Why Brainstorming Technique and Hypothesis Testing Correlation and Regression Analysis Question 339 Answer : A

Gage R & R talks about. Repeatability and Reproducibility Adjustment to Data Measurement Fault Finding towards variation of data Calculation towards data accuracy Question 340 Answer : A

Control Phase deals with majorly two things. Control Plan and Control Methods Before and After process capabilities Improvement and Implementation Plans Reducing variations from the process Question 341 Answer : A

Process Capability for Continuous and Discrete could be determined by. Cpk and Dpmo methods respectively Cp and Dpu methods respectively Standard Deviation and Mean methods respectively Mode and Median methods respectively Question 342 Answer : A

Six Sigma is about Quality, while Lean is about: Utilizing “belts” Improving processes using statistical techniques A thorough analysis of complex problems Increasing speed and reducing waste Question 343 Answer : D

In Correlation Analysis. We find out the mutual and co relationship between two sets of data and if positive, negative or no correlation exists We find out the dependency of one data on another We find out if there is significant difference between one data set and a target We find out if two data sets are different from each other Question 344 Answer : A

Why we consider lower of the two Cpu and Cpl while deciding on Cpk value? Lower value is the safest and ideally attenable Higher value is not good Question 345 Answer : A

In order to calculate Dpmo for Discrete Process Capability, the following ingredients are necessary. Defects, Units, Opportunities Defects, Head count of people Units, Opportunities, USL and LSL Opportunities, CTQ, VOC Question 346 Answer : A

Regression Analysis talks about. Dependency of X data on Y Data How much Y Data changes with X Data changes Relationship between X and Y data How much X Data changes with Y Data variability Both A and B Question 347 Answer : E

What is the key concept of the Six Sigma core philosophy? Six Sigma is a business improvement approach that seeks to perform the “6S"s Six Sigma is a business improvement approach that seeks to find and eliminate causes of defects and errors Six Sigma is a cost-cutting approach that seeks to identify and eliminate unnecessary workers Six Sigma is a revenue-generating an approach that seeks to find ways to produce in greater quantity Question 348 Answer : B

The Sigma Shift of 1.5 Sigma is basically the difference between. Zst and Zlt Cpk and Ppk Cpk and Dmpo Cpk and Cp Question 349 Answer : A

In Probability Distribution, Binomial and Poisson Distributions are respectively for. Sample Data and Population Data Binary and Count Data Continuous and Binary Data Normal and Non-Normal Data Question 350 Answer : B

The difference between other types of Hypothesis Testing and One-way- Anova is. Anova deals with more than 2 data sets Anova deals with more than 5 data sets Anova deals with only Parametric Data Sets No Difference Question 351 Answer : A

In 2 way Anova, we deal with Effect on Y data and also interaction effect sets Only Interaction effect between various X Dependencies between Y and X data None of the above Question 352 Answer : A

The objective of DOE is. To verify the ‘Vital Few’ X(s) that impact the quality of Project Y(s) To verify the ‘Vital Ys that are dependent on the Xs . Project Y(s) Both A & B None of the above Question 353 Answer : A

In DOE, it is important to. Determine the test levels for each independent variables Determine the relationship between various test levels Determine more than 2 levels at a time and start the experiment None of the above Question 354 Answer : A

In DOE, the experimental Trials Table shows. The combination of the Levels and the Factors The relation of the Factors between each other The relation of the Levels with each other None of the above Question 355 Answer : A

In DOE, the result that we present are. Main Cause and Effect Main Effect only Main Effect and Interaction Effect Main Effect and Trial Effect Question 356 Answer : C

In DOE, the conclusion actually states. Which Factor has the Main effect on the Problem and also the Interaction Effect between Factors Which Factor has the Main effect and also the Interaction effect Both the above None of the above Question 357 Answer : A

In DOE, A full factorial DOE is practical when fewer than five factors are being investigated. Testing all combinations of factor levels becomes too expensive and time-consuming with five or more factors. True False Question 358 Answer : A

X Bar R Chart and I MR Chart are for. Continuous Data Normal Data Non Parametric Data Discrete Data Question 359 Answer : A

In Hypothesis Testing, we start off with identification of. Alternate Hypothesis Alpha Error Beta Error Null Hypothesis Question 360 Answer : D

Difference between One-way- Anova and other Tests in Hypothesis Testing is. Anova deals with more than 4 data sets Anova deals with more than 2 data sets Anova deals with only non-normal data sets Anova compares one data set with a target. Question 361 Answer : B

Between Specification Limits and Control Limits. There is only mathematical relationship between them There is only statistical relationship between them There is no statistical or mathematical relationship between them Both are very much related to each other Question 362 Answer : C

Control Charts could be drawn for. Both normal and non normal data Population and Sample data Continuous and Discrete data Only data which has specification limits Question 363 Answer : C

At what stage of the process should the presentation of information, analysis, and statistics to the business come? After the data has been gathered. After experimenting has been completed. After all the work is done. During the brainstorming session. All of the above. Question 364 Answer : E

In a Hypothesis Testing, we try to identify. Significant difference between data sets Mutual relationship between data sets Dependencies between data sets None of the above Question 365 Answer : A

Question 366 Answer : A A measurement system analysis is designed to assess the statistical properties of? Gage variation Process performance Process stability Engineering tolerances

Question 367 Answer : B Which of the following activities is value-added? Setup Process Storage Inspection

Question 368 Answer : A In order for a problem to be solved correctly, which of the following must occur first? The problem must be defined Relevant data must be gathered The measurement system must be validated The process must be mapped

Question 369 Answer : D Typically, which of the following actions is NOT used to reduce process cycle time? Analyzing current processes Reducing queue times Setting scheduling priorities Implementing activity-based costing

Question 370 Answer : D Which of the following tools is commonly used in the Measure phase of a project? Affinity diagram Control chart Failure mode and effects analysis Data collection checklist

In DOE, the first step always is to – Define the Problem Relate the Problem with the Effect Calculate the Problem Verify the Problem for its authenticity Question 371 Answer : A

In DOE, stating the hypothesis means. Identifying the causes that is affecting the problem Identifying the factors that is affecting the problem Both the above None of the above Question 372 Answer : B

In DOE, always we have to identify. The Dependant (X) and Independent (Y) The Dependant (Y) and Independent (X) The Cause and Effect of the Problem The relationship between various X Data. Question 373 Answer : B

In Brainstorming Session, you. Identify Root Causes or Major Value Drivers Plan for the session Both A & B are true None of the above Question 374 Answer : A

In a detailed Process Mapping, there are 3 components. Value Added, Non-Value-Added and Value-Enabling Activities. Desired, Non-desired and Not all desired Activities Both A & B are True Both A & B are False Question 375 Answer : A

A Relationship Diagram is used. T o signify the strength of relationship between two processes or entities. To showcase the step by step breakup of each process To showcase the risks related to each planning process None of the above Question 376 Answer : A

377 Thanks For attending Lean Six Sigma Black Belt Training

Presentation on Six Sigma Green Belt IMC

About This Presentation

Slide Content

Tags

Categories

Download

Quick Actions

Statistics

Related Slideshows

Presentation on Six Sigma Green Belt IMC

About This Presentation

Slide Content

Slide 1

Slide 2

Slide 3

Slide 4

Slide 5

Slide 6

Slide 7

Slide 8

Slide 9

Slide 10

Slide 11

Slide 12

Slide 13

Slide 14

Slide 15

Slide 16

Slide 17

Slide 18

Slide 19

Slide 20

Slide 21

Slide 22

Slide 23

Slide 24

Slide 25

Slide 26

Slide 27

Slide 28

Slide 29

Slide 30

Slide 31

Slide 32

Slide 33

Slide 34

Slide 35

Slide 36

Slide 37

Slide 38

Slide 39

Slide 40

Slide 41

Slide 42

Slide 43

Slide 44

Slide 45

Slide 46

Slide 47

Slide 48

Slide 49

Slide 50

Slide 51

Slide 52

Slide 53

Slide 54

Slide 55

Slide 56

Slide 57

Slide 58

Slide 59

Slide 60

Slide 61

Slide 62

Slide 63

Slide 64

Slide 65

Slide 66

Slide 67

Slide 68

Slide 69

Slide 70

Slide 71

Slide 72

Slide 73

Slide 74

Slide 75

Slide 76

Slide 77