Six Sigma Session For Production And Project Team By Lt Col Vikram Bakshi
SIX SIGMA SESSIONFOR PRODUCTION AND PROJECT TEAM ( COVERING BROADLY YELLOW & GREEN BELT TOPICS ) BY LT COL ( RETD) VIKRAM BAKSHI (BLACK BELT)
Topics• Introduction – Quality tools and Six Sigma Mistaking Proofing• Process Chart / CTQ /CTP/ DEFECT determination/DPMO/Sigma Calculations• Six Sigma Methodologies & Tools – How to make a Flow Chart – Check Sheet of data collection – Cause and effect Diagram (Fish Bone) – Pareto Chart – Histogram – Scatter Diagram – Control Chart• Value Stream Process• Failure Mode and Effect Analysis• How YOU can use Six Sigma: DMAIC Process
What do you mean by “Quality Control” • Fitness for Use • Grade/Quality• The component are said Characteristics to be in good quality if • Appearance they work well in the /performance life / equipment for which they reliability/taste/ are meant to be odor/maintainability/time • VOC Give so much to customer as he values • Who is the customer> VOC> Quality characteristics> CTQs
What are the KPI of any Organization • But Customer is Not• Functional Targets Interested – Profitability • Quality> Product and – Sales Turn over Support Quality – Market Share • Cost – Share Prices • Delivery Schedule
Traditional Management• Time, cost, quality-- • Focus is on achieving choose any two Functional goal of• QA function to Sales/ Design/ Purchase/ independently check up HR/ Production/ Logistics/ on and "control" Accounts/ Administration production• Remove and rework • VALUE IS TO THE defective product through ORGANIZATION inspection P/F• Cost of quality, optimal inspection and rejection
SIX SIGMA • FOCUS IS ON QCD • - It is a management• VALUE IS TO CUSTOMER methodology that helps • ENSURES LONG RUN organization to improve profitability GOOD WILL AND LOYALTY OF CUSTOMER • Disciplined Data Collection and statistical analysis to • ULTIMATE GOAL achieve IS ROCE – Reduce defects RETURN ON CAPITAL – Eliminate Wasteful practices INVESTMENTS – Delight internal and external customer by fulfilling present and future needs• TOTAL SAVING FOR ANY BLACK BELT PROJECT IS 20 LAKHS PER ANNUM
CTP: What are the actions required to achieve the CTQ• What does my customer need from • What is the measurement information to be captured? What is the metric? our process • How is the information going to be collected? Does a form need to• How is our process performance be developed? from the customer perspective • Can it come from an existing source?• How does my customer measure my • When will it be collected? How process granular does the team need to be?• How does my customer view my • Is continuous information needed? process • Will a week-end or month-end number suffice (a point in time number versus continuous)?• What can we do better• How would my customer like for our process to perform
Difference between QA and TQM/6 Sigma. PHASE QA TQM/SIX SIGMA TQM 6 SIGMA SCOPE Prodn All Processes Quality circle is All processes middle mgmt and tech personnel Area of On Line Before the Action Beginning Change triggered Internal and External by internal System Appraisal Prevention customer Function Inspection Plan Do Audit Tech Tech processes Analysis Check processes + Business issues like cost /cycle times/ attrition Status Manager GM rates/meeting delivery schedulesISO9000 SIX SIGMA No Bottom line Profitability is mainWhat to do goal How to do No time Line Well defined Time linesQA is base TQM/TPS is the base Continual Break through Improvement ImprovementDocumentation first Improvement firstImplementation later Documentation laterLeads to Certification Lead to Profit improvements
Lean Tools and ApproachesSix Sigma and Lean concepts and methods areoften combined into Lean Six Sigma.• Both are driven by customer requirements.• Both try to eliminate waste, reduce costs, speed things up, and improve quality.• Both focus on real dollar savings.• Both rely on a systematic methodology.
PROCESS TOOLS Define Measure Analyze Improve Control Define the purpose Understand the Identify and confirm Pilot solutions, Maintain the And scope. Collect Current process Root Causes w/Data evaluate resulting data Gain, standardize Background info. •Data Collection/ •Hypothesis tests •Pugh Matrix •Standardization •Charter Sampling Plan •C&E Diagrams •FMEA •Training •SIPOC •Run Time/Control •Run/Control Charts •New Paretos •Control Plans •VOC Charts •Histograms •New Process Sigma •Run/Control Charts6σ •Affinity Diagram •Frequency Plots •Stratified Frequency •Ideal Flowcharts •Process Audits •CTQ Defined •Scatter Plots Plots •Pilot Studies •Communication •Oper. Definitions •Test for Normality •Regression •Commitment Scale Plans •Pareto •Transformations •DOE •Tree Diagrams •Ownership •Communication •Process Sigma •Validated Root •Gantt Charts transition Plan •Flowcharts Cause List •Planning Grids •Multiply •Gantt Chart •Histograms •Affinity Diagram Solutions •Value Stream Mapping •Autonomation •Operational Analysis • Management •Takt Time •JIT •Six Big Losses Deployment System •Eight Wastes •Visual Stds./Controls •OEE •Vendor ScorecardsLean •Visual Factory •Kanban •SMED •Real-time Dashboards •Error-Proofing •Standardization •TPM •Simple Graphical •Workload Balancing •5S •One-Piece Flow Analysis •Change Control •Future State •Kaizen •Go to Gemba ALL TOPICS OF LEAN AND SIX SIGMA CAN BE COVERED IN FUTURE SESSIONS FOR IMPLEMENTATIONS 12
• How Do we achieve CORRECTION Corrective Preventive Prevention Action Action• RCO > Reduced chance of Rectifying the First find root Extend learning Occurrence mistake that has occurred cause Build POKA ( Fish bone) YOKE• ( Pen Drive) Damage Build POKA Think control YOKA to proactiely to prevent build occurrence of prevention in• ICD> Increase chance of root cause processes detection before things go wrong• (Car seat belt alarm)• (Railway signal)• (CCTV) Six sigma is all about building Prevention Processes If processes does not allow mistake No need to inspect mistake
The Six Sigma Evolutionary Timeline 1818: Gauss uses the normal curve 1924: Walter A. Shewhart introduces to explore the mathematics of error the control chart and the distinction of analysis for measurement, probability special vs. common cause variation as analysis, and hypothesis testing. contributors to process problems. 1736: French 1896: Italian sociologist Vilfredo mathematician Alfredo Pareto introduces the 80/20 Abraham de rule and the Pareto distribution in Moivre publishes Cours d’Economie Politique. an article introducing the normal curve. 1949: U. S. DOD issues Military Procedure MIL-P-1629, Procedures 1960: Kaoru Ishikawa for Performing a Failure Mode Effects introduces his now famous and Criticality Analysis. cause-and-effect diagram.1941: Alex Osborn, head of 1970s: Dr. Noriaki KanoBBDO Advertising, fathers a introduces his two-dimensionalwidely-adopted set of rules for quality model and the three“brainstorming”. types of quality. 1986: Bill Smith, a senior engineer and scientist introduces 1995: Jack Welch the concept of Six Sigma at launches Six Sigma at GE. Motorola 1994: Larry Bossidy launches Six Sigma at Allied Signal.
How do Others Perform? IRS Tax Advice (phone in) 1,000,000 1% of Hospitalized Patients Injured byDefects per Million Negligence 100,000 Doctor Prescription Writing 10,000 Airline Baggage Handling Average 1,000 Company Deaths caused by anesthesia during surgery 100 Domestic Airline Best-in-Class Fatality Rate 1 2 3 4 5 6 7 Sigma Scale of Measure 17
Understanding Six Sigma Sigma DPMO STATISTICALLY STATISTICALLY 2 308,537 Six Sigma refers to aa Most US Six Sigma refers to 3 66,807 process that produces Businesses process that produces 4 6,210 only 3.4 defects per only 3.4 defects per million opportunities. million opportunities. 5 233 6 3.4 GoalBusiness An overall strategy that encompasses yourStrategy organization’s quality philosophy. It sets the vision for achieving Six Sigma levels in key processes.Tools And A set of statistical tools and a disciplinedTactics methodology used by specially trained individuals to improve processes by reducing variation and defects. 18
QUALITY GOAL OF SIX SIGMAAs defined by Motorloa Less than 3.4 defects per million opportunities for error ie .00034 % ie and accuracy of 99.99966% • Define Your Product Unit• VOC >CTQ to satisfy the – A unit is something that is customer satisfaction or quantified by customer requirements – Measurable and observable output of your business process• Define your Product /Service • Define your product Opportunities defects of Error – Does not meet Customer ( – Total Number of Chances per unit Internal and external) to have a defect . specification or requirements – Each opportunity must be – Causes customer independent o other opportunities dissatisfaction – Does not fulfil functional or physical requirements
Customer Needs vs. Customer CTQ’s• Customer needs are the data collected from customers that gives information • CTQ about what they need or want from your process. Customer needs are – Stated Needs: often high level, vague, and non- Specified by customer specific Policies/tender/LOI “I need a quick response!” – Unstated Needs: “I need accurate information!” Are not specified (labor need legal• CTQ’s are customer needs requirements) translated into critical process requirements that are specific and measurable. • Expectations: Delighters of the• A fully developed CTQ has three elements: Y metric, target, customer specification/tolerance limits 21
Getting to the CTQ’sTranslating a customer need into a fully developed CTQ Time from inquiry to Example: resolution (Y metric) Quick Response CTQ 5 minutes or less (Target) Not greater than 60 minutes (specification / tolerance limit) 22
Define – Customer Requirements What are the CTQs? What motivates the customer? Voice of the Customer Key Customer Issue Critical to Quality SECONDARY RESEARCHListening Posts Industry Intel Market Data Industry Benchmarking Customer Correspondence Customer Service PRIMARY RESEARCH Surveys Surveys OTM Obser- Focus Groups vations
Examples• Area: Call Center • Area: Book Publisher• Customer Complaint: I wait • Customer Complaint: I cant stand consistently too long to speak to to an any typing error in books I executive purchase• CTQ Name: Executive Responsiveness • : Typograpghic Quality• CTQ Measure: Time on Hold seconds) • CTQ name CTQ measure:• CTQ Specification: Less than 60 Number of typography mistakes seconds from call connection to • CTQ Specification: Zero automated response system typographic mistakes• Defect : Call with holds time equal and greater than 60 seconds• Unit : Call • Defect: Any typogrpahic mistake• Opportunity : 1 per call • Unit : Word• Sigma ? • Opportunity: No of letter per word
DPU AND DPMO• Take the data collected based on the data • Calculate the Defects Per Million collection plan, using an appropriate sample Opportunities using the following size. formula:• Calculate the number of defects in the sample based on the Operational Definitions. • DPMO = Defects x 1 Million divided by No of Units• Define what an opportunity is. Be cautious processed at that review point x in defining too many opportunities; this will Opportunities for error in that unit artificially inflate the Sigma Level• Suppose that a project is focused on a DPMO= 60 x1000000 divided by 250 x billing process. The team wants to 1 have correct bills sent to the customer. They have defined one opportunity for this process - either the bill is correct or not. All =240 000 of the bills produced are the same in terms of complexity. The team took a sample of 250 bills and found 60 defects • Using a conversion table, the team found the Sigma Level to be about 2.2. They used this information to baseline the current process performance.
Use the table to convert DPMO into the Sigma Level
Another example• A purchase order has 20 • DPMO= DPU X 1000000 Opportunities for error. DIVIDED BY The person who enters OPPORTUNITIES FOR the order makes one ERROR IN THAT UNIT defect on an average. What is Sigma level for • = 1X 1000000 this process DIVIDED BY 20• DPU= No of defects divided by No of units = = 50,000 1 divided by1=1 SIGMA LEVEL= 3.1
CALCULATION SIGMA LEVEL FOR THE ENTIRE ORGANIZATIONSub No of No of Opportuni DPU DPMO SIGMAprocess units defects ty for LEVEL errorSales Order 20 5 w/o POFin GRN Bills 10 25 MISTAKE SHR 23 8 DELAYEDProductio 100 25n MISTAKE STOTAL OPP DPU
SPC• Capability Indices • Cp = USL – LSL divided by 6s• Process capability is the ability of the process to meet the • Cp is the potential capability requirements set for that process. indicating how well a process could be if it were centered on• One way to determine process target. This is not necessarily its capability is to calculate actual performance because it• capability indices. does not consider the location of the process, only the spread.• Capability indices are used for • It doesnt take into account the continuous data and are closeness of the estimated• unit less statistics or metrics. process mean to the specification limits.• There are many capability indices but the two most commonly used are Cp and Cpk (or Pp and Ppk).
DPU AND DPMO• Take the data collected based on the • Calculate the Defects Per Million data collection plan, using an Opportunities using the following appropriate sample size. formula:• Calculate the number of defects in the sample based on the Operational • DPMO = Defects x 1 Million Definitions. divided by Units x Opportunities• Define what an opportunity is. Be cautious in defining too many DPMO= 60 x1000000 divided by 250 x opportunities; this will artificially inflate 1 the Sigma Level =240 000• Suppose that a project is focused on a billing process. The team wants to • Using a conversion table, the team have correct bills sent to the customer. found the Sigma Level to be about 2.2. They have defined one opportunity for They used this information to baseline this process - either the bill is correct the current process performance. or not. All of the bills produced are the same in terms of complexity. The team took a sample of 250 bills and found 60 defects
PROCESS CAPABILITY• A control chart statistically determine upper and • How do we calculate lower limits either side of a process average . IN CONTROL X bar Mean and• Process width = UCL - LCL UCL= X bar + 3 sigma sigma also called as standard LCL= Xbar – 3 Sigma deviation :WILL COVER LATER PW= 6 SIGMA IN PPT IN HISTOGRAM PORTION Specification Width is given by customer UCL AND LCL must remain within USL and LSL = USL - LSL Cp= SW divided by PW = SW divided by 6 sigma PROCESS IS SAID TO BE CAPABLE ONLY IF Cp > 1 Cpk and Control Charts Not being covered WILL COVER IN DETAIL IN NEXT SESSION INCLUDING Cpk
99% Accuracy Practical Meaning of “99% Good”• 20,000 lost articles of mail per hour• Unsafe drinking water almost 15 minutes each day• 5,000 incorrect surgical operations per week• 2 short or long landings at most major airports each day• 200,000 wrong drug prescriptions each year• No electricity for almost 7 hours each month 99% Is NOT Good Enough Anymore
6 – The Measurement 99.0% = 3.85 99.9996% = 6 Is there really a big difference between 99.0% & 99.9996%?
What’s in a name?• Sigma is the Greek letter representing the standard deviation of a population of data.• Sigma is a measure of variation (the data spread) σ µ
Facts• All work is a process• All processes have variation and waste• Variation causes defects waste causes loss• Processes can be improved by understanding the nature of the variation and waste
What does variation mean? 20• Variation means that a 15 process does not produce 10 the same result (the “Y”) 5 every time. 0 -5• Some variation will exist in -10 all processes.• Variation directly affects customer experiences. Customers do not feel averages!
Measuring Process Performance The pizza delivery example. . .• Customers want their pizza delivered fast!• Guarantee = “30 minutes or less”• What if we measured performance and found an average delivery time of 23.5 minutes? – On-time performance is great, right? – Our customers must be happy with us, right?
How often are we delivering on time? Answer: Look at the variation! 30 min. or less s 0 10 20 x 30 40 50• Managing by the average doesn’t tell the whole story. The average and the variation together show what’s happening.
Reduce Variation to Improve Performance How many standard or less 30 min. deviations can you “fit” within s customer expectations? 0 10 20 x 30 40 50• Sigma level measures how often we meet (or fail to meet) the requirement(s) of our customer(s).
Another Example Roller Bearing Manufacturing Diameter is a CTQ (Critical To Quality Parameter) Nominal diameter = 2.5mm Minimum Spec = 2.25mm Maximum Spec = 2.75mm 41
Example (Cont.) Lower UpperSpecification Specification Limit Limit Nominal Diameter Customer is expecting 2.5 mm 2.5 mm But will allowNo Less Than No More Than some variation within the Spec 2.25 mm 2.75 mm range. 42
Example (Cont.) Manufactured Roller Bearing Diameter Actual Micrometer Measurements 43
Example (Cont.) Manufactured Roller Bearing Diameter Variation ending up as a defect 44
Example (Cont.) Let’s Look at Some Basic Statistics Mean diameter = 2.50 mm Standard Deviation = 0.125 mm On Average it’s OK It’s a Variation issue 45
Example (Cont.) Reducing Variation is Clearly the Key to Improving Process Capability 2 46
Example (Cont.) Reducing Variation is Clearly the Key to Improving Process Capability 3 47
Example (Cont.) Reducing Variation is Clearly the Key to Improving Process Capability 4 48
Example (Cont.) Reducing Variation is Clearly the Key to Improving Process Capability 5 49
Example (Cont.) Reducing Variation is Clearly the Key to Improving Process Capability 6 50
I think now we know what we mean by SIX SIGMA LETS AGAIN REWINDProcess Defect • The Objective of 6 Sigma is to achievecapability DPMO level of 3.4 defects or errors out of Count in(Sigma every million defect opportunities. %Level) • This translates into 99.9997% 690000 perfection1 69 308000 30.82 (2 Times) • While everybody talks about customer satisfaction and world class quality, here we are transferring Quality into 66800 6.683 (5 Times) Quantifiable language 0.62 Six Sigma Approach focuses on: 62104 (10 Times) Customer needs Data-driven improvements 230 0.023 The inputs of the process5 (30 Times) And this results in: 0.00034 Reducing or eliminating 3.46 (80 defects Reducing process variation Times) Increasing process capability
Managing Up the Sigma ScaleSigma % Good % Bad DPMO 1 30.9% 69.1% 691,462 2 69.1% 30.9% 308,538 3 93.3% 6.7% 66,807 4 99.38% 0.62% 6,210 5 99.977% 0.023% 233 6 99.9997% 0.00034% 3.4
Table Comparison of 99.9% Vs 99.9997%Activity/Operation Error Nos with 99.9 % Quality Error Nos with 99.9997 %QualityDelivery of 300,000 300 mis deliveries 1 Mis deliveryLetters60 500 Crashes <2 CRASHESClearance of 200,000 200 wrong clearance < 1 wrongcheques in a week clearanceHandling of 60,000 60 bags lost in a day < 2 BAGS LOST INnos bagagages by A WEEKairline in a day
Six Sigma is. . .• A performance goal, representing 3.4 defects for every million opportunities • Who is the Customer to make one.• A series of tools and methods used to improve or design products, • What are the Quality processes, and/or services. Characteristics• A statistical measure indicating the required by him number of standard deviations within customer expectations.• A disciplined, fact-based approach to • CTQ managing a business and its processes. • Actions required to• A means to promote greater awareness of customer needs, achieve the CTQs performance measurement, and business improvement.
Examples of the Sigma ScaleIn a world at 3 sigma. . . In a world at 6 sigma. . .• There are 964 U.S. flight • 1 U.S. flight is cancelled every cancellations per day. 3 weeks.• The police make 7 false • There are fewer than 4 false arrests every 4 minutes. arrests per month.• In MA, 5,390 newborns are • 1 newborn is dropped every 4 dropped each year. years in MA.• In one hour, 47,283 • It would take more than international long distance 2 years to see the same calls are accidentally number of dropped disconnected. international calls.
PART THREE ANYBODY FOR A SMOKE BREAK OR CONTINUE!• Six Sigma Methodologies & Tools – How to make a Flow Chart – Check Sheet of data collection – Cause and effect Diagram (Fish Bone) – Pareto Chart – Histogram – Scatter Diagram – Control Chart
Process Maps• Another important set of deliverables of the Define phase are process• maps.• The maps should be based on the actual state of the process, or "asis“ maps.• They should not show the desired state at this point in the project.
High Level Process MapsDevelop high level process maps to include next-level subprocess overviews,identify process linkages and gaps to ensure that the teams are aligned.Deliverables:a. Sub Team Processes Mapped to Critical Stepsb. Team Leaders Met and Reviewed & Aligned Maps
SIPOC. • Establish a name for the process.• The first stage would be to create a SIPOC. SIPOC stands for • Define the starting point and the ending point Suppliers, Inputs, Process, of the process to be improved. These should Outputs and Customers. already be listed in the scope section of the team Charter.• The SIPOC is a very top-level • List the key outputs of the process. Usually, view of the process to be this list includes up to three or four main improved. outputs even though the process may produce more.• By starting the mapping process • Define who receives those outputs, i.e. the at this level, it allows the team to customers. These customers may be internal (part of the business) or external. quickly develop a common understanding of the process to • State the top-level process steps of the improve and the key customers process. Keep the list to four to eight main and suppliers steps. These steps do not contain any decision points or feedback loops. • List the inputs to process. Stick with one to four main inputs. • Define who supplies the inputs to the process.
SIRPORC REQUIREMENT CUSTOMERSOURCE REQUIREMENTSUPPLIER INPUTS OF PROCESS OUTPUT CUTOMER OF THE SUPPLIER PROCESS AREA OF PROCESS WHERE WE EXPECT TO FOCUS OUR INITIAL MEASUREMENTS SUBPROCESS
Process UnderstandingSIPOC – Suppliers, Inputs, Process, Outputs, CustomersYou obtain inputs from suppliers, add value through your process, and providean output that meets or exceeds your customers requirements.
FLOW CHART • What people are presently doing/ START/ internal and external interconnections FINISH • Relation ship Map how various function and individual interface with each other ACTIVITY/OPERATION • Maps and Flow Chart to make work visible DECISION STAGE • Identify alternative way • Improvement opportunitiesLINK TO NEXT ACTIVITY • Analysis toolA DOCUMENT OR DIAGRAM
TOP DOWN MAP • The steps for creating a top-down chart• However, the SIPOC is probably not are: detailed enough to find opportunities for making the process better. Agree on the start point and end point for the process map.• More detailed mapping is required The next stage of mapping may be a These are already listed in the top-down chart. SIPOC.• The top-down chart takes the • Identify four to eight major steps that information from the SIPOC and adds describe the process from beginning to a second level of information. end.• For each step listed in the SIPOC, the List those steps horizontally across team defines the associated sub- a flipchart page. steps. These are also taken from the• There are still no feedback loops or SIPOC. decision points • Break each major step into three to seven sub-steps. List the substeps under the corresponding major step. •
TOP DOWN SUB PROCESS AFTER WE MAKE SIPOC CHARTReview the map and make corrections as necessary.Rearrange steps, combine sub-steps, or Revise the descriptions of the major steps or sub-steps soAccurately describe the process. Agree on a presentation format for the process map. FROM SIPOC PHASE MAKE YOUR FLOW CHART
NEXT MAKE THE FUNCTIONAL DEPLOYMENT MAP• The third stage of mapping is to create a more detailed "as- is" picture• The functional deployment map displays the steps of a process in sequential order.• The functional deployment process map also illustrates what function performs the process step.
The steps for creating a functional deployment map are:• Review the top-down chart with the team.• List each of the process steps in sequential order in the first column.• Use the horizontal axis across the top to show the location/ responsibility or department for each step performed.• Depict individuals (by job title/position), specific locations, or work functions.• Indicate the steps, activities, and decisions that make up the process under the associated functional column.• Identify the sequential order in which the steps are actually performed.• Use arrows to indicate the direction of the process flow• Review the final map and correct as necessary.
Annotating Process Maps• After the map is complete, additional information can be added depending on the project goals.• For instance, if the goal is to reduce cycle time,the time for each step may be added.• If the goal is to reduce defects, yield information may be added. T• This may help the team identify the areas on which to focus.• However, this type of information may not be readily available and, often, after the data are collected in the Measure phase, the maps are updated
Process Understanding Process Map – should allow people unfamiliar with the process to understand the interaction of causes during the work-flow. Should outline Value Added (VA) steps and non-value add (NVA) steps. F u ll F o r m C ont ro l O pen S ta rt S iz e S o r t s P u ll & S o r t R e c e ip t / D ocs E x t ra c t C k / Vou ch V e r if y P e r f e c t io nR e qua l G ro up No Yes Prep cks, R e m it R u lr s route Prep cks S h ip t o I P Pass 1 Pass 2 vouch V o u c h e rs Key from B a la n c e D ata Cap im a g e No V ou ch OK I nventory Yes Pre p F o ld e r s / F u ll F o r m S h ip t o B ox Q C R e v ie w Cust
Process Understanding Create daily peak Action staff need plan Plan No Yes Can they Call employee Add 30% to To Floor the required make it? (3x) no. No Yes NoOperations Need OJT Make Compare to OJT Re-Tng it? Check off original Billet rpt desired Manually Review returnee Yes Update HR Staff staff & "need No Yes Billet Request Billet Need re to retrain" To Floor -train list Add 40% to Call (3x) Stop! staff needed Create Update Staff No IPS No Billet Rev Do they original Do they No Send Letters Yes Yes Have we No Yes Have we No Yes Interview / Meet Fleet Do they want to billet & want to Call Wait Rank as to desired hired hired New hiring respond? work this call work this List pre-hire "1 2 3" staff enough? enough? criteria peak? uncheck peak? ed What if the HR sends Hire in 1- Yes returnee is Yes Yes req for No No 2 order Start already staffing (3s are HR / working here show up No nos. not Place into Call Recruit on another Do they Do they orienta No No placed) dept 3X program? want to want to tion Stop! Stop! Currently stay on the stay on the send the ltr list list anyways Wait List Yes Yes Yes New & Other Take off Set 14 Take off Set 14 People IPS month IPS month call in system flag (on system flag (on IPS?) IPS?) schedule Yes No Gen Event Roster for Reach rpt in IPS training Show No Call Notify up? 1X HR Yes Training Gen rpt for Ops Kronos Recruit Train No Yes Update Pass? IPS
QUICK WINS• Criteria for Good Quick Wins• After the mapping activity is done, the team should then use the maps to assess if there are some obvious opportunities for Quick Wins.• Quick Wins should be changes that are easy, fast, and cheap to implement, andthat fall within the teams responsibility.• In addition, the team may want• to ensure that the changes are easy to reverse since these ideas may not be validated with data.
SECOND TOOL: CHECK SHEET OF DATA COLLECTION• How many times each value occurred Defect 25Aug 26Aug Total item 2012 2012• Shows frequency of a particular Mold IIII III cracked defect and how often it occurs in a specific location Fiberes II III• Enables Operator to spot the Cracks I IIII problem Pin IIII IIII I• Easily set the priorities Holes•• Part with highest number of defects carries highest priority for Total II I correction
THIRD TOOL : PROCESS APPROACH AND PROCESS MODEL • CONTROL Must Know what people BOUNDARY are presently doing, where are the contributions in relation to Transformations CONTROL customer internal and Using resources and externalINPUT Mgmt Activities OUTPUT Y Relationship Map View X1 X2 Make work VisibleX3…Xn Uncover non valued added To Identify potential breakdowns, Improved opportunities rework loops and source of variation Also used as Anlytcla tool in process
FOURTH TOOL :CAUSE AND EFFECT DIAGRAM (FISH BONE)Fish Bone Diagram - A tool used to solve quality problems bybrainstorming causes and logically organizing them by branches. Alsocalled the Cause & Effect diagram and Ishikawa diagram Provides tool for exploring cause / effect and 5 whys
Cause and Effect Matrix• The Cause and Effect Matrix gives weights to each y indicating the importance of that y.• Man Material Money Machine Method Environment• Then, each x is rated in terms of its correlation to each y. Calculations are made based on the importance and Correlation, and higher scoring xs are the best candidates for data collection.• Rate the degree to which the x affects or is correlated to each y.• Use the following scale: – 0 - no effect or correlation – 1 - small effect or weak correlation – 3 - medium effect or medium correlation – 9 - strong effect or strong correlation.• This scale ensures that the x data that the team thinks has the strongest effect on the y will stand out.• Multiply each rating by the weight and sum across the row, putting the result in last column.• The xs with the highest totals are the ones that the team should try to collect.
FIFTH TOOL :THE PARETO CHART• Another powerful teaming tool is the Pareto chart. The Pareto chart is • The Pareto chart is a bar chart. based on the Pareto principle. The height of the bars indicates the count, or frequency, of• Pareto was an economist in the early occurrence. The bars represent 1900s who discovered that 80% of all one grouping of the data, the wealth was held by 20% of all the people. such as defect type.• This became known as the 80/20 rule • The idea motivating this chart is and it was found to be applicable to that 80% of the count will be due more than the economy. to 20% of the categories.• Eighty percent of the warehouse • The bars are arranged in space is taken up by 20% of the part numbers. Eighty percent of the defects descending order, therefore the are caused by 20% of the defect types dominant group can be determined and it will be the first bar on the left. This chart can be used in a variety of places in a Six
PARETO CHART • We will draw a Pareto Chart • Arrange all data in descending order • Arrange data in percent with No of readings• Pareto Mean 80:20 • Convert Data into Cumulative• After data has been collected we draw pareto to focus on the Vital few. Cumlativ Descending % e• The remaining factors are called Trival Many A 17 34 34• Reasons for Customer Dissatisfaction – – Wrong Items: 17 Delivered Late: 4 B 12 24 58 – Transit Damage: 8 – Late Vehicle : 12 C 8 16 74 – Late Installation: 3 – Absentee of Engineer: 1 D 4 8 82 – Others: 3………………….IN LAST – Failure of items: 2 E 3 6 88 – 17 divided by 50 x 100 = 34 so on F 2 4 92 G 1 2 94 H 3
IDENTFYING VITAL FEW IN PARETO CHART 80:20 100 90 80 70 60 50 40 30 20 10 0 A to D are CTP Vital
Step 2: Measure – Pareto Tool ANOTHER EXAMPLE Customer Service Requests Between Week 31 and Week 38 that Were Delivered Late 99.2% 250 97.2% 100 n = 251 94.4% 90.8% 86.1% 90 200 76.1% 80 70 162 64.5% 150 60# of Requests % of Total 50 100 40 30 50 20 29 25 10 12 9 7 5 2 0 0 DOCS LLR RS NFL O&E START QUESTION RS, DOCS Request Type• MS Excel and QC Tools software used to construct graphs• Tool used to rank data by groups from the group that contains the most data points to the group that contains the fewest data points.• Look for a “large bar” or “young mountain”
SIXTH TOOL No of Defective Defect Defect in PuneHISTOGRAM Boilers In Savli 1 • Displays frequency of 96 90 distribution of 3 continuous /variable 91 81 data 6 82 73 CONSIDER 7 A DATA OF DEFECTS 77 66 OCCURRED IN 9 PRODUCTION IN SAVLI AND PUNE PLANTS 68 61 •Total 50 Nos PRODUCT 7 in each location 62 58 6 HOW WILL YOU DETERMINE 55 56WHICH PLANT IS MORE CONSISTENT WITH LESS RANGE IN DEFECTS 5 47 53 4 38 51 2 23 50
FIRST CALCULATE INTERVAL FROM DATA-DETERMINE TOTAL NUMBER OF PRODUCTS AND THEN SQUARE ROOT IE 50= 7 ROUNDING OFF-96 WHICH IS HIGHEST READING IS DIVIDED BY 7= 14-SO NOW WE HAVE INTERVEAL OF 14 EACH.-DETERMINE READINS AS PER INETRVAL OF BOTH PUNE AND SAVLI PLANTS• 14 to 28> 2 of 23 • 14 to 28> 0• 28 to 42> 5 of 38 • 28 to 42> 0• 42 to 56> 5 of 47 and 6 of 55 = 11 • 42 to 56> 2 of 50, 4 of 51, 5 of 53, 6 of 56• 56 to 72> 7 of 62 and 9 of 68 =16 • = 17• 72 to 84> 7 of 77 and 6 of 82=13 • 56 to 72> 7 of 58, 9 of 61, 7 of• 84 to 98> 3 of 91 and 1 of 96 66 =23 =4 • 72 to 84> 6 of 73, 3 of 81 =9 • 84 to 98> 1 of 90
PLOT NUMBER OF READINGS AS PER Y SCALE IN EACH INTERVAL OF 14 ON X SCALE WIDER THE BASE MORE THE VARIATION AIM IS TO BRING BASE CLOSER TO MEAN 25 savli MEAN Pune 20 3-D Column 3 3-D Column 4 15 3-D Column 5 3-D Column 6 10 3-D Column 7 3-D Column 8 5 3-D Column 9 0 3-D Column 10 28 56 84 3-D Column 11NARROW BASE WIDER BASE
MEASURE OF CENTRAL TENDENCEY• AVERAGE MEAN: 64.18 & 61• MEDIAN: MIDPOINT OF DISTRIBUTION OF DATA : 68 & 61 • CONCLUSION• MODE: RANGE WITH MAX OBSERVATION : 68 &61• RANGE: X max – X min = spread of • A strong Central tendency distribution . 73 & 40. Range is measure of ( Mean , Median, Mode are Variation close to each other ) indicates less variation in process• Another measure of Variation is Standard Deviation also known as Sigma or s + Square root of ( X • If bars are more spread it – Xmean) ( X – X mean) / N means range is higher or• Where X is X1 X2 X3….. higher variation• N is population size• THIS DATA IS ALSO USED TO CALCLULATE Cp•
Step 2: Measure – Histogram Tool ANOTHER EXAMPLE OF HISTOGRAM Customer Service Document Requests Between Week 31 and Week 38 that Were Delivered Late 120 _ x n = 162 (2 outilers of 79 and 64 hours not shown) USL=0.5 hours mean = 7.139 std dev = 8.538 100 96 80# o R q e ts f eus 60 40 20 14 14 10 8 7 4 1 1 1 1 1 2 0 -1.55 0.55 2.65 4.75 6.85 8.95 11.05 13.15 15.25 17.35 19.45 21.55 23.65 25.75 27.85 29.95 # of hours to deliver request MS Excel and QC Tools software used to construct graph Fits data into a frequency distribution Can show upper and lower specification limits and denote data thatfalls outside those limits (the area on which the story should focus) Used for data breakdowns by hours, minutes, dollars, etc.
OK BEFORE WE GO TO NEXT PARTLETS US AGAIN REWIND WHAT WE HAVE LEARNED
Part four• Value Stream Process• Failure Mode and Effect Analysis • AIM IS TO ADD VALUE TO THE PROCESS IF IT IS LENGTHY AND TIRESOME
REMEMBER THE PROCESS MAP WHAT WE HAD MADE IN SIPOC AND FLOW CHARTNOW OUR NEXT AIM IS TO IMPROVE THE PROCESS CHART BY MAKING OUR SYSTEM A LEAN SIX SIGMA EFFICIENT AND QUICK BUT BY REMOVING ALL NON VALUE ADDED STEPS AND PROCESS IN A BUREAUCRATIC DECISION MAKING THAT DO NOT ADD VALUE BUT ONLY DELAY THE GOAL ACHIEVING
FUNCTIONS FILL IN PROCESS STEPS AND DEPART MENT PERSONS IN SEQUENCE OF ACTIVITIES SUBMITTER Fills
NOW IDENTIFY IN YOUR SYSTEM PROCESS WHAT ARE THE CUSTOMER VALUE-ADDED ACTIVITIES• Customer-value added has all the following characteristics: • A non-value added activity is one that does not fit into the other two categories.• The customer recognizes the value. • Examples of non-value added• It changes the product or service toward something activities are: proofreading, the customer inspection and checking, logging• expects. information, checking calculations, reviewing and• It is done right the first time. approving, moving and set-up, monitoring work, and rework.• An operational-value added activity has all the following characteristics: • The team would focus on the non-value added activities to see if they could be eliminated or• It is required to sustain the workplace ability to minimized. perform customer value-added activities. • An important clarification is that• It is required by contract or other laws and regulations. a non-value added activity doesnt automatically make it an• It is required for health, safety, environmental, or unnecessary activity personnel development reasons.• It is done right the first time.
IDENTFYING VALUE ADDITION ( TIME SPENT IN HOURS)ACTIVITY REQUIRED BY NOT REQUIRED BY NEITHER BENIFTS CUSTOMER CUSTOMER CUSTOMER OR RVA BVA BUSINESS NVAUnderstanding customer requirement 15Checking by customer 5Error corrected 7Prepare Execution plan 10Keep plan aside to attend another activity 20Refer plan for approval by senior 5Implement Plan 20Arrange Invoice 8Await Delivery clearance 15Ship to customer 5Customer Sign 5
IMPROVING THE PROCESS MAP RVA: Real value adding activity to be optimized BVA: Business value adding activity to be minimized NVA:Non value Adding Activity to be eliminated • Map will be Current Base line Map Step 1Draw As is Map • Reflect NVA/ removed/ appraisal/prevention for like failure modes Step 2Draw Should be Map • Compare Maps and Identify actions to move to Should be Level Compare Map • Take action to improve Improve Process processes to Should be Verify, Implement Level Validate Make control Plan
ANOTHER PROCESS IMPROVEMENT TOOL PFMEA• Capture defect in your shop that • Information to have customer does not catch in his operations before FMEA• Surfacing the problem early in • Customer requirements specifications production cycle will enable us to solve • Engineering drawings them wih minimum expenses and time • Warranty information and field failure • Previous FMEA• Why perform a FMEA> To prevent • Process Flow Diagram design /process related problems • Process Capability data during product launch /subsequent operations • Prototype TEST data • Defect failure information• How> By attempting to surface • How customer will use end product problem early before they have a • How customer may abuse the end chance to occur product• Benfit> save time and resources
FMEA WORK SHEET P What can go Potenti Se Critic signi Pote Occure Desi Detecti Abuse by Revis r wrong o PFM al effect in ve rit al fican t ntial caus nce gn verif on R customer ed RPN c e failure mode y es of failur icati on P s s e mech N nais s m t e p Flow chart the process Classify any special characteristics Describe process and function List potential causes for each failure mode List each potential failure mode Estimate likelihood of occurrenceDescribe effect of each type of failure List prevention/detection control Rank severity of failure Rank detection Identify RPN
IMPROVERecommended Responsibility for Action Sev o Detection RevisedAction to improve RPN completing action Actually erity c RPN Taken c u r e n c e List recommended actions to lower RPN List individual /depts for completing actions List actual actions taken Re compute RPN after corrective actions
CONTROL PLAN Requirements Control Equipment Checked By Action planProcess Imposed used if out ofstep control condition occurs Repeat steps for all processes /sub processes/ parts Develop a control plan and a contingency plan
PART SIX HOW WE CAN TAKE A PROJECTTO MAKE OUR ORGANIZATION A LEAN SIX SIGMA
Six Sigma Method DMAIC: To improve any existing product or process Define Measure Analyze Improve ControlWho are the What are the How can wecustomers and most important maintain thewhat are their causes of the improvements?priorities? defects? How is the process How do we performing and remove the how is it causes of the measured? defects? 102
Six Sigma Tool Box Define Measure Analyze Improve ControlBenchmarking Value Stream Fishbone Modeling SPC Charts Map DiagramsProcess Flow Cause & Effect FMEA Tolerance PerformanceMapping Control MetricsFlow charts Defect Metrics Root Cause Defect Control Multiple Analysis RegressionProject Charter Statistical ANOVA Design Trainas a Team Analysis ChangesSet Up a Plan & Data Collection Cause & Effect PilotingGuidelines for Run Charts, DiagramTeam Time Series Chars, Time Value Charts, Pareto ChartsReview Existing Sampling Scatter Plots Best PracticesDataSIPOC
DEFINE PHASE• Define Phase: The goal of define phase is to define the project scope by understanding background information about the process and its customers• Tools used in define phase as voice of customer, project charter are used decide the scope of project and define boundaries of improvement effort.• It also identifies key stakeholders, time lines, improvement priorities, and improvement targets at the beginning of project.
TOOLS AND METHOD IN DEFINE STAGEDEFINE PHASE STEPS/TOOLSVOC/ QFD/ KANO MODEL/ BENCHMARKING/ CTQ/PROJECTCHARTER/ SIRPORC/DEFINERESPONSIBILTY AUTHDEFINE CUSTOMER REQUIREMNTS VOC CTQ DEFINATIONDEVELOP PROJECT CHARTER PROBLEM STATEMENT GOAL STATEMENT BUSINESS NEED ADDRESSED SCOPE CONSTRAINTS RESOURCES ASSUMPTIONS GUDILINES TEAM COMPOSITION PRELIN PROJECT MILESTONESDEVELOP HIGH LEVEL PROCESS MAP SIPOC/SIRPOCACTION PLAN FOR PROCESS STANDARDIZATION INTERIM CONTAINMENT ACTION
FINAL DEFINE STAGE RESULTS• What are Customer driven CTQ and Process driven CTQ• Sources of Existing Customer data• Assessing Customer requirements and expectations• Recall Vital few customer CTQ• Analyze VOC and its impact on CTQ• Translate Customer needs into CTQ
Measure OverviewWhat is the Measure phase?The Measure phase defines the defects, establishes improvementgoals, determines that the system of measuring defects isrepeatable and reproducible and gathers data about the process.Why is the Measure phase important?The Measure phase ensures that you specifically define thedefects you are going to measure and that your measurementsystem is accurate before you begin to actually measure theprocess. 107
Measure• The primary purpose of • Deliverables from the the measurement phase Measure phase include: is to answer the questions, • Collected data• "How are we doing?" and • Selection of what "How far do we have to measures to use go?“• The team needs to • Baseline of the "current establish a baseline of state“ the current performance level. • Data collection plan
• Develop operational definitions for each CTQ characteristic• Figure out how to measure internal processes affecting each CTQ, KPOV (Key process output variables), KPIV (Input Vars) – Y = F(x)• Figure out what data we need to collect – Easy to collect correctly – Interrupt process as little as possible – Collectors understand why collecting – “gage study” to determine the validity (repeatability and reproducibility) of the measurement procedure for each CTQ• Baseline data – Collect baseline capabilities for each CTQ – Determine the process capability for each CTQ
TOOLS IN MEASURE STAGEDefine Defect, Based on CTQ and process knowledge ,Opportunity, Unit, metric BenchmarkingIdentify CTPs Using Pareto, Correlation Analysis CT MatrixDevelop Data Collection Plan Questionnaires, sampling plansValidate measurements system MSA/Gage R&RCollect data and tabulate data Check sheet and Data sheetCalculate Process capability and Cp Cpk DPMO /Sigma LevelSigma base line calculations
WHAT CAN BE MEASURED• Input Indicators - Measures that • Output Indicators - Focus on the evaluate the degree to which the end result. Measures that evaluate inputs to a process (provided by dimensions of the output - may suppliers) are consistent with what focus on the performance of the the process needs to efficiently business as well as that and effectively convert inputs into associated with the delivery of customer satisfying outputs. products and services to customers.• Examples of input indicators – . # of customer inquires • . Retention rates – . Type of customer inquires • . Total # done, sold, made, etc. – . # of orders • . On-time – . # of positions open • . Complete – . Accuracy of the analysis – . Timeliness
WHAT ALL CAN BE MEASURED AS GOOD AND DEFECTS• Process Indicators - Focus turning opportunities into desired results. Measures that evaluate the efficiency, effectiveness and quality of the transformation processes (i.e., the steps and activities used to convert inputs into customer satisfying outputs.)• Examples of process metrics include: – . Availability – . Time to do something, timeliness – . # of non-standard request – . Yield (first time through) – . # of exceptions (e.g., non standard approvals) – . Quality level (could also be an output metric)