White Belt Training · 2016-04-26 · White Belt Training Quality and Continuous Improvement...
Transcript of White Belt Training · 2016-04-26 · White Belt Training Quality and Continuous Improvement...
1/11/2016
1
White Belt TrainingQuality and Continuous Improvement
Concepts and Practices
Introductions
• Name
• Company & Role
• Expectations from class
1/11/2016
2
Today’s Agenda
• Lean Six Sigma – definition and justification
• Fundamentals of Quality
• Data Collection
• Team Fundamentals
• Roles and Responsibilities
• Plan, Do, Check, Act
• 8 wastes
• Mistake Proofing
• 5 why
• Standard Work
• 5S
• Kan Ban
• Six Sigma – DMAIC• Fishbone
• XY Matrix
• Measles Chart
• Basic Stats
Module # 1 Agenda
• The Case for LSS?
• Quality – How is it Defined and Measured
• Where Do We Start – Data / Metrics / Variation / Sampling
• Value of Six-Sigma
• DMAIC Methodology
1/11/2016
3
What is LSS?
• LSS = Lean / Six-Sigma
• Lean – Focuses on efficiency and eliminating waste in process.
• Derived from Toyota Production Methods
• Six-Sigma – Focuses on quality management and eliminating defects.
• Started at Motorola, refined at GE
Each methodology is unique in complexity and deployment.
Each is focused on continuous process improvement and utilizes a variety of quality and improvement tools to help drive and sustain business success.
Each is continually seeking perfection!
The Case for LSS
• Continuous improvement is integral to success in all types of organizations.
• In many areas of business and industry, today’s workforce needs to be equipped in the basic tools of quality assurance, waste elimination, problem solving, critical thinking, advancing an understanding & improving customer relations, and project based team work.
IF YOU’RE NOT IMPROVING -- YOU’RE STANDING STILL!
1/11/2016
4
What is LSS?
• Taken Together…
• LSS – relies on a collaborative team effort to improve performance by systematically removing process waste and inefficiency.
• LSS – should be viewed as an enterprise wide deployment effort coached and mentored through the belt based training system.
• The belt personnel are designated as white belts, yellow belts, green belts, black belts and master black belts, each maintaining different degrees of training and experience.
Fully deployed both methodologies work in concert to drive process improvement & $UCCE$$
MAKING YOUR PROCESSES BETTER, FASTER & CHEAPER
What is Quality?
• Two types of “quality”
Qualitative - “The works of DaVinci are beautiful.”
qualitative tends to be subjective
Quantitative – “My Maserati does 185”
quantitative is objective
• Posited for industry per quality gurus:• Joseph Juran - “Fitness for use.”
• Philip Crosby - “Conformance to requirements.”
1/11/2016
5
Who determines the need?
• In all cases, the customer determines what quality “is.”
• Quality can have “value” as a consideration• While we usually “think” of “quality as the very best”, we tend to “live” with “quality is the best
at the level we can afford”
Critical To Quality
• CTQs are the internal critical quality parameters that relate to the wants and needs of the customer. • They are not the same as CTCs (Critical to Customer), and the two are often confused.
• CTCs are what is important to the customer
• CTQs are what’s important to the quality of the process or service to ensure the things that are important to the customer.
http://www.isixsigma.com/dictionary/critical-to-quality-ctq/
1/11/2016
6
Stated Another Way…
• Critical to Quality is an input• The right color paint
• Correct and consistent pressure in paint system
• Well functioning, clean paint spray nozzles
• Critical to Customer is an output• Consistent, even, shiny coat of metallic flake paint on their vehicle
Does Quality = Inspection?
• No! This is a misperception. You can’t inspect quality into the product!
• Inspection may be necessary in some instances; however:• Inspection does not make the product better – it only identifies bad product to remove
• Inspection is costly – it takes time and effort to inspect product
• If inspection is done by someone other than the person who made the product or provided the service, the focus is in the wrong place
• People should have an interest in their work, and be proud of good work
• Inspection by another person can promote a punitive system
1/11/2016
7
Systems that Rely on Inspection
• This style of quality management is called “Quality Control”• Usually there are many inspectors
• It becomes more difficult to fix problems or find them (because the inspection may occur long after the fact)
• Has a higher level of defects in the process
Quality Assurance
• People are responsible for their own work
• Very few inspectors – quality helps and supports people doing their own inspections, and audits the process, not the part or service
• Greater sense of ownership and pride of work
• Problems found and resolved more quickly
1/11/2016
8
Cost of Quality, Continued…
• Three areas of cost:
• Prevention costs (good design/quality built in it)
• Appraisal costs (a Quality Management System (QMS), with tools and techniques to measure/monitor products and/or services)
• Failure costs (warrantees, replacement etc.)
Prevention costs the least!Failure costs the most!
Cost Of Poor Quality, (COPQ)
VISIBLE
LESS VISIBLE
1/11/2016
9
COPQ Taken Together…COST OF QUALITY
(COQ)
COST OF POOR QUALITY
COST OF CONFORMANCE
Internal Failure,Non-conformance,Internal Deviation,
Use AS-IS
External Failure,Non-conformance
caught at CustomerField Failures,
Returns
Appraisal CostsPlanning and Prevention Costs,
Good Design,Quality Built In
So Where Do We Start?
• We need to know what to improve, first.
• You can’t know what to improve unless you measure it or have data available.
• In order to measure it, you have to establish metrics or develop an effective Data Collection Plan (DCP).
REGARDLESS OF METHODOLOGY SELECTED…
– LEAN / SIX-SIGMA / QUALITY MANAGEMENT TOOLS –
IT IS AT THIS POINT WE BEGIN TO MAKE A CONSCIENCE DECISION TO…
“IMPROVE”
1/11/2016
10
Planning For Metrics / Data Collection
• The simply but critical questions you should begin to ask…
• WHAT – questions do we need to answer?
• HOW – will we recognize and communicate the answers?
• WHAT – data analysis tools are we going to use and HOW will the results be communicated?
• WHAT – type of metrics/data do we need?
• WHERE – in the process can we get the metrics/data we need?
• WHO – in the process can give us the metrics/data?
• HOW – can we collect the metrics/data from the individuals with minimum effort and error?
• WHEN – is the metrics/data to be provided?
• HOW – much will it cost to collect the metrics/data?
• WHAT – additional information do we need for future analysis?
What is a Metric / Data?
• A Measurement of Performance. It can be:
• Product or Service Related – % non-conforming product, % dropped calls.
• Monetarily Related – gross sales last quarter.
• Ability to Meet Performance Standard – attainment to schedule, comparison to a standard.
• Some metrics track “real time”
• What did we produce in the last hour.
• Some track to historical performance
DO YOU CURRENTLY HAVE
METRICS IN YOUR PROCESSES?
1/11/2016
11
Basic Idea Behind Metrics / Data Collection
“You can’t control what you don’t measure.”
“You can’t improve without a standard.”
• If you don’t measure, you can’t compare performance:
• To a known standard (meds delivered to patients within a 2 hour window)
• To an expectation of performance (increase sales $ next month by 10%)
• To itself to see if it’s improving or worsening (weekly on time delivery)
Intentionally Blank
1/11/2016
12
Team Fundamentals
Some Benefits of Teams
https://knowhownonprofit.org/people/teams/about-teams-and-types-of-team/importance
1/11/2016
13
FORMING
STORMING
NORMING
PERFORMING
ADJOURNING
TEAM STAGESDAY 1
FORMINGDAY 2
STORMINGDAY 3
NORMINGDAY 4
PERFORMINGDAY 5
ADJOURNING
CHARACTERISTICS• Questioning• Socializing• Displaying
eagerness• Focusing on group
identity & purpose• Sticking to safe
topics
CHARACTERISTICS• Resistance• Lack of
participation• Conflict• Competition• High emotions• Moving towards
group norms
CHARACTERISTICS• Reconciliation• Relief – lower
anxiety• Members are
engaged & supportive
• Developing cohesion
CHARACTERISTICS• Demonstrations of
interdependence• Healthy system• Ability as a team to
effectively produce• Balance of task and
process orientation
CHARACTERISTICS• Shift from task to
process• Sadness• Recognition of
team & individual effort
STRATEGIES• Take “lead”
individual contacts• Clear expectations
& consistent instructions
• Quick response
STRATEGIES• Normalize• Encourage
leadership
STRATEGIES• Recognize
individual & group efforts
• Provide learning opportunities and feedback
• Monitor energy in the group
STRATEGIES• Celebrate• Guide from the
“side” – minimal intervention
• Encourage group decision making & problem solving
• Provide opportunities to share learning across team
STRATEGIES• Acknowledge
change• Provide
opportunity for summative team evaluations
• Provide opportunity for acknowledgement
Conflict, Cont’d
Positive Conflict
• Leads to mutual respect
• Sharing of ideas
• Blended solutions
• True best idea implemented as people see beyond personal impact or ego
Negative Conflict
• People retreat to their corners
• Become further entrenched in paradigms
• Growing hostility
• Functional silos
• No change made
1/11/2016
14
Conflict Management Strategies
Style Party #1 Party #2 Use When
•accommodatingI lose you win issue is small to you, and granting other is conciliatory
•avoidingno winners or losers timing is not right, need more info, saving face
•collaboratingwin win foundation of trust and teamwork
•compromisingbig pic win, pers loss when consensus is needed to move forward
•competingI win you lose in crisis, when you are absolutely right and can't budge
Consensus
• The Principles of Consensus Decision Making
•Consensus decision making is a process used by groups seeking to generate widespread levels of participation and agreement.1 Not a majority vote
http://consensusdecisionmaking.org/Articles/Basics%20of%20Consensus%20Decision%20Making.html
The key take-away for consensus is that people are willing to give up personal or departmental goals for greater organizational goals. They are willing to agree on and support a course of action even if they would prefer a different course of action.
Can you live/support with the change even though it wouldn’t be your first choice?
1/11/2016
15
•Inclusive: As many stakeholders as possible are involved ingroup discussions.
•Participatory: All participants are allowed a chance to contribute to the discussion.
•Collaborative: The group constructs proposals with input from all interested group members. Any individual authorship of a proposal is subsumed as the group modifies it to include the concerns of all group members.
• Agreement Seeking: The goal is to generate as much agreement as possible. Regardless of how much agreement is required to finalize a decision, a group using a consensus process makes a concerted attempt to reach full agreement.
•Cooperative: Participants are encouraged to keep the good of the whole group in mind. Each individual’s preferences should be voiced so that the group can incorporate all concerns into an emerging proposal. Individual preferences should not, however, obstructively impede the progress of the group.
http://consensusdecisionmaking.org/Articles/Basics%20of%20Consensus%20Decision%20Making.html
Consensus Building
Let’s go out to eat example
• Example – four people in an office are going out to lunch
• Ted’s favorite is Mexican, but he is flexible
• Joan is a vegan; otherwise, as long as there is a vegan plate on the menu, she is fine.
• Paul wants pizza, burgers or BBQ
• Andi is fine with anything that is not too expensive
• The group discusses their desires, potential compromises and non negotiables; they decide that veggie pizza will work for all
1/11/2016
16
Team Structure and Hierarchy
• Regardless of what type of team structure you have, whether it’s a belt system following Six Sigma or another discipline, consider the following:• Leadership at the top of the organization must drive the change for it to be successful
• Provide a clear vision
• Provide a reason for change
• Provide a clear goal
• Provide a strategy, such as six sigma or lean deployment
• Empower employees
Other Levels of the Organization
• All levels of the organization must be involved for it to be effective
• This does not mean that everyone at once have to be involved; it means that even working in one department, the new employee to the senior exec will be involved in some fashion
• People should not be afraid to make mistakes
• People need to be proud of their workmanship
• People in the process are stakeholders
1/11/2016
17
Six Sigma Lean
Problem Statement
Six Sigma + Lean
Waste Elimination Standard Work Flow Customer PULL
SPEED &AGILITY
STABILITY & ACCURACY
Six Sigma
Variation Reduction Scrap / Rework Elimination Process Optimization Process Control
Lean
1/11/2016
18
Lean vs. Six Sigma
* Not all-inclusive, just showing major tools
White Belt
http://image.slidesharecdn.com/six-sigma-management-slides4268/95/six-sigma-management-slides-13-728.jpg?cb=1271394957
1/11/2016
19
What’s my Role as a White Belt?
• Valued team member• Understand basic concepts (elaborate...basic quality, etc.)
• Familiar with some simple tools (fishbone, XY Matrix)
• Understand how to function effectively as a team member
• Recognizes that collaboration and consensus are the drivers to acceptance of new ideas
• Understands different types of data and data collection, can support the team in data collection
• Advocate of change – often a liaison to their home dept
1/11/2016
20
PDCAA simple technique to start in Continuous Improvement or in a smaller organization
Plan-Do-Check-Act (PDCA)• As a model for continuous improvement.
• When starting a new improvement project.
• When developing a new or improved design of a process, product or service.
• When defining a repetitive work process.
• When planning data collection and analysis in order to verify and prioritize problems or root causes.
• When implementing any change.
1/11/2016
21
Plan-Do-Check-Act (PDCA)• PLAN – so, we’ve set a strategic goal.
• Now, what questions do we have to consider to get there?
• How will the goals be met?
• What data collection do you need to consider?
• This is necessary to plan as we’ll need it in the check stage.
• DO – actual execution. Small scale is best to start.
• One department…
• One process…
• Single item.
Plan-Do-Check-Act (PDCA)• CHECK – is where benchmarking comes into play, and why the metrics
we spoke about earlier are important.
• Did we make a statistically significant and practical difference?
• There are a variety of comparison tools you can use to test if you are on the right track.
• ACT – ASQ finds the act phase as comparable to “standardize.”
• This is very true when the outcome is positive at the desired level.
• Act can also be the stage at which you go back to PLAN because you didn’t achieve the intended goal.
• Even if a positive outcome is achieved, remember that this is a cycle.
1/11/2016
22
Let’s go down the lean path first…
8 Wastes
1/11/2016
23
Waste in the Process
• 8 waste acronym – DOWNTIME
• Defects
• Overproduction
• Waiting
• Non value add activity
• Transportation
• Inventory
• Motion
• Employee Under-utilization
Another Acronym
• TIM WOOD
• Transportation
• Inventory
• Motion
• Waiting
• Over-production
• Over-processing
• Defects
1/11/2016
24
Defects
• The other very obvious waste
• Does not conform to customer requirements therefore:
• Is scrap
• Requires rework
• Remember, having to fill out paperwork to disposition the mess and the rework (if that’s the solution) make for over-processing waste
Over-Production
• Over-production is producing more than is needed by the customer• All of the wastes of inventory, plus:
• Over-production can cut into other allocated resource time
• Over-production can deplete materials allocated to other jobs
1/11/2016
25
Waiting
• Waiting is one of the more obvious wastes. It occurs whenever we are waiting for any resource:• Material
• Tools or equipment
• People
• information
• Another process to finish before we can start ours (can be ahead or behind)
Non-Value Added Activity
• Repetitive steps or unnecessary steps• Having a customer fill out multiple forms that each have common fields (repetitive).
• Someone writes data on a sheet that is taken to someone else to type it into a DB (let’s discuss).
• Shot blasting the rust off of a part, setting it outside where it gets rusty, then having to shot blast it a second time.
• Multiple choices for phone menus, leading to a single choice, only to have Customer Service Rep ask what you need help with.
1/11/2016
26
Transportation
• Transportation involves any movement involved in bringing materials or tools to the desired work area.• Some transportation is “necessary” but always treat it as a waste so that alternatives can be
considered.
• Real life example of transportation elimination: some suppliers introduce materials right onto the shop floor at point of use, rather than warehousing it. This takes advantage of “Just in Time” manufacturing principles and often utilizes a “kan-ban” system.
In the Past…
Cutting
Milling Drilling
Heat Treat
Departments were laid out by functionality. All of the drills were in one department, all of the mills elsewhere, etc.
We gave the WIP a big ride around the factory!
1/11/2016
27
Now…
Cellular manufacturing. One of each machine in the process, operators work within the cell. Drastically cuts down on transportation – cuts down on errors, too, as operator B can interact with operator A in a much quicker period of time.
Inventory
• Why is inventory considered a waste?• Tying up capital in unsold parts or unsalable part (Work In Process or WIP)
• Taking up floor space
• Makes for greater amounts of waste due to damage and spoilage
• Idea is to have the minimum amount for reasonable usage
• Even “consumables” should be considered inventory
• True story: one hospital had a storage room containing approximately a two year supply of gloves. Much of the space could have been utilized for other much needed supplies, which were now located in an area further away
1/11/2016
28
Motion
• Motion is different from transportation in that it focuses on the person• Particularly related to ergonomics
• Lifting
• Body twisting
• Bending
• Repetitive motion
• Travel to retrieve items
Employee Involvement
• Toyota identifies “unused employee potential” as a waste.
• This is very true – although it may be harder or impractical to point out in your process map.
1/11/2016
29
Paper AirplaneLooking for waste
What Waste Do You See?
• Try to describe each of the 7 wastes that you see
1/11/2016
30
Kaizen
• Kaizen is from the Japanese – two words meaning “make better” or “continuously improve”• Kaizen by itself most often means incremental improvements
• Kaizen program of employee contributions
• Smaller but necessary and good application to immediate work area
• Considered evolutionary change, not revolutionary
Kaizen, Cont’d
• For kaizen to work well, it must:• Be supported by management
• Recognize employees for kaizen contributions
• Make kaizen implementations part of a metric
• Needs to be documented
• Standard work as applicable
• Training to all affected employees
• Be run through PDCA cycle
• Change is not a one time event
• Evaluate change and then evolve to find next level of improvement
1/11/2016
31
Kaizen Blitz, or Rapid Change Event
• Kaizen blitz is a concentrated activity that takes place over several days.• It tends to be more than incremental change
• It is sometimes “revolutionary” not evolutionary change
• It may involve a new way of doing things, thus requiring an overhaul of:
• The workspace or production floor
• The standardized work
Non-Linear Change
• This is clearly “revolutionary” change, but doesn’t necessarily have to be under the auspices of kaizen blitz
• Best example – photography going from conventional to digital
• Paper based record keeping becoming computer based
• Non linear takes more planning, time for acceptance (consensus) and communication – since you’re venturing into uncharted territory
1/11/2016
32
Mistake Proofing
Mistake proofing or error proofing is a technique to eliminate human induced errors or misses in inspection.
• The error proofing device becomes your “inspector” for the characteristic it’s evaluating
• Eliminates human error based on Shingo’s 10 errors (next slide)
• Is usually simple in design (preferred)
Sources of Error, Per Shingo
1/11/2016
33
The sources of Defects
• Processing
• Parts• Right part, wrong process Right patient, wrong med?
• Wrong part, correct process Knee replacement, wrong knee?
• Right part, but missed a process Sponge not removed during surgery?
• Adjustment errors Adjusting medication dose?
• Set-up errors Power level on equipment?
Defects can be detected
• Before they happen (impending)• Prevents altogether or tells of impending failure
• During the making of the defect• Detection during the cycle, so only one of few parts made
• After the fact• Prevents from going to next operation
1/11/2016
34
POKA-YOKE - MISTAKE PROOFING FUNCTION
• What is it?
• Ideally, poka-yokes ensure that proper conditions exist before actually executing a process step,
preventing defects from occurring in the first place.
• Where this is not possible, poka-yokes perform a detective function, eliminating defects in the
process as early as possible.
• Why is it important?
• Poka-yoke helps people and processes work right the first time.
• Poka-yoke refers to techniques that make it impossible to make mistakes.
• These techniques can drive defects out of products and processes and substantially improve quality and reliability.
POKA-YOKE - MISTAKE PROOFING FUNCTION
• Used as an extension of other tools.
• More often than not deploying mistake proofing measures is an extension of a Pareto Analysis
and/or a Failure Mode effect Analysis (FMEA).
• Toyota has an average of 12 mistake-proofing devices at each workstation and a goal of
implementing each mistake-proofing device for under $150.
• When to use it?
• Poka-yoke can be used wherever something can go wrong or an error can be made.
• It is a technique, a tool that can be applied to any type of process be it in manufacturing or the service industry.
1/11/2016
35
POKA-YOKE - MISTAKE PROOFING FUNCTION
• Errors are many types…
1. PROCESSING ERROR:Process operation missed or not performed per the standard operating procedure.
2. SETUP ERROR: Using the wrong tooling or setting machine adjustments incorrectly.
3. MISSING PART:Not all parts included in the assembly, welding, or other processes.
4. IMPROPER PART/ITEM:Wrong part used in the process.
5. OPERATIONS ERROR:Carrying out an operation incorrectly; having the incorrect version of the specification.
6. MEASUREMENT ERROR:Errors in machine adjustment, test measurement or dimensions of a part coming in from a supplier.
POKA-YOKE - MISTAKE PROOFING FUNCTION
• Step by step process in applying Mistake Proofing:
1. Identify the operation or process - based on a Pareto Analysis and/or FMEA.
2. Analyze the 5-whys and understand the ways a process can fail.
3. Decide the right Mistake Proofing approach, such as using a…
• shut out type (preventing an error being made), or an…
• attention type (highlighting that an error has been made) poka-yoke…
• take a more comprehensive approach instead of merely thinking of poka-yokes as limit switches, or automatic shutoffs…
• a poka-yoke can be electrical, mechanical, procedural, visual, human or any other form that prevents incorrect execution of a process step.
1/11/2016
36
POKA-YOKE - MISTAKE PROOFING FUNCTION4. Determine whether a…
• contact - use of shape, size or other physical attributes for detection…
• constant number - error triggered if a certain number of actions are not made…
• sequence method - use of a checklist to ensure completing all process steps is appropriate.
5. Trial the method and see if it works.
6. Train the operator, review performance and measure success.
The combination of errors and defects are often connected.
If we classify the two, we can apply the correct type of Poka-Yoke.
POKA-YOKE - MISTAKE PROOFING FUNCTION• The mistake proof device becomes your “inspector” for the characteristic it’s
evaluating.
• Its design is to eliminate human error.
• There are 5 basic types of mistake proofing.• Guide Pins
• Error Detection and Alarms
• Limit Switches
• Counters
• Checklists
1/11/2016
37
POKA-YOKE - MISTAKE PROOFING FUNCTION
GO / NO GO GAGEGUIDE PINS
What prevented you from putting the disk in upside down or backwards.
Green wont fit in the hole – to small.Red end fits in hole to large.Green fits – Red does not… Just right.
Cam prevent oversized / undersized from entering the process. Can also prevent mis-orientation.
LIMIT SWITCH
Simple Limit Switch connected to a drill head. The drill wont engage until the work is pushed up against the gate.
POKA-YOKE - MISTAKE PROOFING FUNCTION
PARTS COUNTER / SCALE
Best for small parts in high quantities. Good fro parts containers. Must “tare” the container weight.
AUTOMATIC PARTS COUNTER USING LIMIT SWITCH
Best on continuous line where parts pass single file.
25
1 2 3 4
1 2 3 5
1/11/2016
38
POKA-YOKE - MISTAKE PROOFING FUNCTION
• Alarms:• Temperature too high or low
• Dwell time too long or short
• Out of position
• Machine down (this is a failure – it’s not just for parts only)
• Speed too fast or slow
• Pressure too high or low
• Example:• Low pressure alarm on grind line – might indicate clogged filter.
• Can prevent impending failure from happening (no air pressure) by allowing operator to clean filter.
• Reduces extra work associated with reduced air pressure.
5 Why
• Thought is that asking “why?” 5 times will usually take you to the root cause level.
• Can be more or less than 5; 5 is a general rule.
• If your 5 why is done well, you should be able to start at the last (5th) answer and work your way back up stating, “Therefore, the cause of this why is the previous statement”, continuing back to the first why.
• Never assume an answer in the question…as in, “Why don’t the employees follow the inspection method?” Unless that’s the answer to the previous level question.
• You can have more than one branch, if there are several potential “whys.”
• Let’s look at an example.
1/11/2016
39
Service Industry 5 Why
http://www.velaction.com/5-whys/
Part is out of specification
Why is part out of specification?
Hole is undersized
Why is hole undersized?
The wrong drill was used
Why was the wrong drill used?
3/8”drill was placed in ½” tool holder on turret
Why was 3/8”drill placed in ½” tool holder on turret?
Special job – set-up operator changed tool but didn’t replace
Why didn’t operator replace?
Didn’t mark sheet with reminder flag
1/11/2016
40
a 50-year-old woman was hospitalized after taking Flomax, used to treat the symptoms of an enlarged prostate, instead of Volmax, used to relieve bronchospasm
http://www.medicinenet.com/script/main/art.asp?articlekey=53208
Standard Work
• Essence of standard work• Consistency – do it the same way every time!
• Best practice – this is developed with input from the team
• Process optimization
• Variability reduction
• Safe and most efficient method for the worker
• Standard work provides a benchmark so that you know when something is not in order
• Not as a punitive standard! Help your team member get back to standard and discover cause of problem
1/11/2016
41
Standard Work
• Standard work provides a reference for auditors
• Should be used as the training document for new workers
• Make as visual as possible, succinct• Pictures aid people, especially those who are English as a Second Language (ESL) or have
difficulty reading
https://www.youtube.com/watch?v=qhGwyMK4ux8
Video
Addressed in Standard Work
Elements
• Safety
• Takt time
• Ergonomics
• Parts flow
• Maintenance
• Routines
• Quality or inspection points
Operations
• Cycle time
• Work sequence
• Standard inventory
1/11/2016
42
http://markdhaworth.com/stand.html
The Recipe Analogy
• A recipe is basically a standardized work.• Tells you when you need to do an operation
• Before starting, preheat your oven to 350 degrees
• Tells you the equipment you need to use
• Using an electric blender…
• Tells you the sequence in which the operations are performed
• Mix your dry ingredients first, then fold in egg mixture
• Tells you the settings for your equipment
• Beat on high. Set oven to 350 degrees
1/11/2016
43
Standardized Work and Training
• The only way to make standardized work successful is to have all personnel who carry it out:• Be well trained in the standard work
• Observed actually demonstrating the process after training for a check of understanding
• Carry out the standard work every time
• No exceptions
• Can recommend new procedure under kaizen, which can be tested under controlled conditions
• Have the standard work available for review
• Have periodic retraining
• Have back-up personnel trained in the event primary operator is out
5S
• 5S is a fundamental tool for continuous improvement
• Focus is on keeping an orderly workplace• Improves productivity
• Improves safety
• Increases morale
• Although housekeeping is a large component, it is more than a housekeeping program
• Effective for offices as well as manufacturing, hospitals, and other facilities
1/11/2016
44
Sort
• Sort is the first of the 5 S’s
• Concentration is on Sorting out the equipment or items which are:• No longer in use
• Need repair
• Are broken beyond repair
• Items should be moved to a “red tag” area for review
http://www.kaizenworld.com/_Media/5s-explanation_med_hr.jpeg
Straighten
• The catchphrase for this S is, “A place for everything, everything in its place.”
• Every single item is evaluated for:• How often used
• Weight
• size
• Based on these factors, placed in best spot for usage and ergonomics
• Becomes a dedicated spot, either labeled or shadow board• In manufacturing, includes raw materials, pallets and other consumables
1/11/2016
45
Examples
http://www.slideshare.net/kkraebbe/teaching-lean-workshop-shs-asq-poole-hinton-kraebber-2010-02-25
http://5sbestpractices.ning.com/photo/tool-organization
Sweep and Shine
• Very straightforward title – but there are underlying reasons for the “deep clean”• If a machine is dirty, you often can’t see it leak oil
• A machine leaking oil is a problem
• Clean it so you can see the leak, then fix the leak!
• Dirty filters cause loss of effectiveness, or not work at all
• Bad for people if air filter, bad for machine if keeping dust out of motor
• Clean the filter or replace!
1/11/2016
46
Standardize
• In this step, now we want to make schedules for the activity we’ve done:• How often do we clean/replace filters?
• Who takes on this function?
• When will it be done? Typically at end of work day (15 minutes of concentrated cleaning activity)
• Scheduling board is a good way to do this – everyone can see
• Cleaning stations with everything laid out (using the 2nd S)
Sustain
• In order to keep the gains made in the other phases, we now use an audit system to ensure that all of the standardized tasks are being done.
• This is very important in the ideology of continuous improvement – you can’t continuously improve unless you sustain what you have.
1/11/2016
47
Controlling the flow of inventory as a waste elimination method
• If over-production and inventory are two big culprits in waste, it makes sense to control them
• There are two techniques in particular the can be helpful• Kan ban (covered)
• Value Stream Mapping (VSM) – not covered, but feel free to look up on your own
The Key to Controlling Inventory – Pull Systems
• As much as possible, create what the customer calls for only when they need it• This is called a “pull” system
• Differentiates itself from a “push” system which is based on forecasts and creates stockpiles of inventory
1/11/2016
48
Great, Simple Example
http://s2.hubimg.com/u/3901061_f260.jpg
Kan Ban
http://www.toyota-global.com/company/vision_philosophy/toyota_production_system/just-in-time.html
1/11/2016
49
Follow up paper airplane exercise
Segue into Six-Sigma
• What about the size variation on the paper?• While lean eliminated the 6 wastes, six sigma might be a better strategy for the 7th waste,
defects
• This is about reducing variation
1/11/2016
50
DMAIC
• Six Sigma relies on a project management technique known as DMAIC.
• Each of the letters in this acronym stand for a particular phase of the program:• Define
• Measure
• Analyze
• Improve
• Control
DMAIC
• Define - Select the outputs to be improved and therefore measured (known as Y’s)
• Measure - data gathering, making sure our measurement system is adequate
• Analyze - identify the significant factors which control the outputs
• Improve - eliminate or reduce the source of variation
• Control - the plan to prevent slipping back into former modes of operation
1/11/2016
51
Define
• Problem statement
• Prioritize what you’re going to work on
• Set performance goals
Process Mapping
• A process map is key at the start of any improvement project
• Shows you the “as is” state
• Can help you identify areas of waste (lean perspective) or variation (six sigma perspective)
• Immediately identifies “alternate processes” which have to be followed under abnormal conditions (also a common source of variation)
1/11/2016
52
Cont’d
• After finding the opportunity in the “as is” map we draw the “future state” map (this is further along in the process after define our problem statement, find the root cause and propose a new method).• Details how we propose to streamline the process
• We can attach values of
• Time
• money
• By having it mapped out, gives people a visualization
• Can help team or stakeholders to spot potential flaws or missed steps in the future state
Process Map with “swim lanes”
http://www.totheedgecoaching.com/wp-content/uploads/2015/04/Business-Process-Mapping-31.jpg
1/11/2016
53
Pareto
• You also need to rank your problems in order of importance. For example, if your charter wants you to reduce non-conformances on a production line, which non-conformances are the greatest? Or perhaps the most costly?
Pareto Chart
• A simple visual chart to show the relative make-up of studied categories by a variable metric• Type of histogram
• In six sigma, used to rank problems
• Named after Vilfredo Pareto• He observed and reported the 80/20 rule
1/11/2016
54
So Let’s Jump Right In…
• Pareto charts don’t actually identify the root cause, but they direct you to what should be worked on first.
Defects by Occurrence for Period 9/1/08 to 11/1/08
0
5
10
15
20
25
unde
r spec
over
spec
wro
ng th
read
size
burrs
and
roug
h ed
ge
no th
read
wro
ng b
evel a
ngle
missing
hole
type of defect
Fre
qu
en
cy
Let’s go over some of the elements of a good Pareto chart on the next slide.
Cont’d
Defects by Occurrence for Period 9/1/08 to 11/1/08
0
5
10
15
20
25
unde
r spec
over
spec
wro
ng th
read
size
burrs
and
roug
h ed
ge
no th
read
wro
ng b
evel a
ngle
missing
hole
type of defect
Fre
qu
en
cy
This does not
have to be
frequency – it
can also be
time or $
1/11/2016
55
Cont’d
Defects by Occurrence for Period 9/1/08 to 11/1/08
0
5
10
15
20
25
unde
r spec
over
spec
wro
ng th
read
size
burrs
and
roug
h ed
ge
no th
read
wro
ng b
evel a
ngle
missing
hole
type of defect
Fre
qu
en
cy
How Do You InterpretThis Data?
Defects by Occurrence for Period 9/1/08 to 11/1/08
0
5
10
15
20
25
Ove
r spec
wro
ng th
read
size
burrs
and
roug
h ed
ge
no th
read
unde
r spec
wro
ng b
evel a
ngle
missing
hole
type of defect
Fre
qu
en
cy
Defects by Dollar for Period 9/1/08 to 11/1/08
$-
$100
$200
$300
$400
$500
$600
unde
r spec
wro
ng b
evel a
ngle
wro
ng th
read
size
Ove
r spec
no th
read
burrs
and
roug
h ed
ge
missing
hole
type of defect
Do
llar
Am
ou
nt
1/11/2016
56
What About This Data?
Set-Up Delays by Frequency
0
2
4
6
8
10
12
14
Waiting on
Inspection
method
Waiting on
material
Program
change
Waiting on
Crane
Waiting on
Material
Handler
Quality Issue
Type of Delay
Nu
mb
er
of
Ev
en
ts
Note: downtime costs the same, regardless of the type of event.
Set-Up Delays by Time
0
5
10
15
20
25
30
35
40
Program
change
Quality Issue Waiting on
Inspection
method
Waiting on
material
Waiting on
Crane
Waiting on
Material
Handler
Type of Delay
Cu
mu
lati
ve
Tim
e p
er
Ev
en
t
Wrap up on Pareto and Graphs
• Helps us to ID the “worst of the worst”
Can be used to identify frequency, time or money, whichever makes the most sense
As items lessen on the chart, new ones take their place or old ones move to top of list – update frequently!
1/11/2016
57
Process Inputs as Influencers
• Best way to list them all is through brainstorming while doing a fishbone diagram (also known as an Ishikawa diagram)
• This exercise helps you think of all of the potential areas without hampering your thought process
• Let’s learn about the fishbone…
The Parts of the Fishbone
• The question we ask (basically equivalent to the first why in a 5 why) as the problem statement. Then possible causes in the categories of:• Man
• Method
• Machine
• Materials
• Milieu (environment)
• Measurement
1/11/2016
58
Example Fishbone
Man/Manpower
• Possible ideas• Training
• CTQ’s
• Causality
• Inspection methods
• Time constraints
• Perception issues (“I think the lead values numbers over quality”)
• Ergonomics (fatigue issues are sometimes an important consideration)
1/11/2016
59
Shingo and 10 Errors (as shown in mistake proof)
• Forgetfulness
• Misunderstanding
• Identification error
• Trainee error
• Willful error
• Inadvertent error
• Slowness error
• Lack of standard
• Surprise error
• Intentional error
From the book, PokaYoke, Productivity Press
Man/Manpower
• Very Important – never use this category to bash people! Approximately 90% of errors can be traced to management (simply because management owns the system), so be sensitive to this fact. (Deming)
• Question – how often do we wake up in the morning thinking we will go to work to do an inadequate job? We want to succeed!
1/11/2016
60
Method
• Possible ideas• Lack of standardization
• Some follow their “best judgment”
• Conflicting information (English to Metric conversion)
• Method is followed but is incorrect
• Method does not provide all information
• Operation is new, so there really isn’t a method established
Materials
• Possible ideas
• Material does not meet our standards
• We don’t have standards
• Lots of material have variation, or lots are mixed
• Material and process are not very compatible
1/11/2016
61
Machine
• Possible ideas
• Machine is not capable of holding the tolerance
• Machine is not calibrated, or is out of calibration
• Machine is not set at correct settings (speed, temperature, etc.)
Environment
• Possible ideas
• Usually associated with environmental conditions on materials, such as humidity and heat
• Environment can also be environmental factors on workers – lighting, noise
1/11/2016
62
Measurement
• Miscalibrated or uncalibrated measurement equipment
• Equipment does not measure down to the level of precision needed
• Gage or inspector error
• Parallax error on dial style tools
Cont’d
• A question often asked….
• “What if I have trouble figuring out which of two M’s under which I should place the suspected cause?
• Put it in both places – there’s no problem having it on the diagram twice
1/11/2016
63
Exercise
• Fishbone Exercise:
• After walking the class through a fishbone, I divide the class up into two groups –or do a single group for a small class.
• Group #1: Your lawn has brown grass, bare spots, weeds and looks pretty shaggy. What are inputs that contribute to this?
• Group #2: You manage a successful and very busy restaurant. Some customer complaints lately however indicate that the food arrives to the table lukewarm or cool, not hot. Do a fishbone diagram to consider potential causes.
XY Matrix
• XY Matrix can be a continuation of the fishbone, or can be a stand-alone tool
• Ranks the relative strength of the inputs to the outputs
1/11/2016
64
Example
http://asq.org/sixsigma/2007/03/x-y-prioritization-matrix.xls
Example XY Matrix
• I take the group through an exercise showing the relationship of outputs for a beautiful green lawn (lush green grass, no bare spots, even height and no weeds) to the inputs (water, fertilizer, grass seed, mowing, weed killer). Of course, the XY Matrix should be closely tied to the , because these should for the most part be the outputs and inputs evaluated (although others can come up from the fishbone).
• Group #1: You want to have a safe, on time, comfortable and relaxing commute to work each morning. Create an XY Matrix which shows these outputs, figure out the inputs and their relative strength, and calculate to show the rankings of inputs by degree of influence.
• Group #2: You are putting on a presentation to a large audience. You want your audience to experience an informative, engaging, thought provoking and relevant presentation. Create an XY Matrix which shows these outputs, figure out the inputs and their relative strength, and calculate to show the rankings of inputs by degree of influence.
1/11/2016
65
Measles Chart
• Very visual system for displaying location of defects• Done at the time the defect is made/caught
• Eliminates need for a descriptive paragraph in a text report
• Involves the personnel closest to the problem – no “researching” the defect
• Shows a “concentration” of the problem if one exists, making it easier to find the root cause
Concentration Diagram or Location Check SheetMeasles Chart
http://www.syque.com/quality_tools/toolbook/Check/vary.htm
1/11/2016
66
A Concentration Study for Injuries
Measure
• MSA (check our measurement system – is it “robust?”)
• Baseline measurement
• Define or brainstorm X’s
1/11/2016
67
Representation of Data with a Curve
Mean, Median, Mode
• The mean is simply the average. In a normal bell shaped curve, it’s right in the middle.
• The median is the middle point. In a normal bell shaped curve, it’s right in the middle.
• The mode is the most frequently occurring point.
1/11/2016
68
0
1
2
3
4
5
6
4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 16.0 17.0 18.0
Nu
mb
er
Values
Frequency
Mean 10.0
Median 10.0
Mode 10.0
n 29Skewness 0.0Stdev 2.8Min 4.0Max 16.0
Data set is: 4,5,6,7,7,8,8,8,9,9,9,9,10,10,10,10,10,11,11,11,11,12,12,12,13,13,14,15,16So 10 is the middle point of data
0
1
2
3
4
5
6
4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 16.0 17.0 18.0 19.0 20.0 21.0 22.0 23.0 24.0 25.0 26.0 27.0 28.0 29.0 30.0 31.0 32.0 33.0 34.0 35.0 36.0 37.0
Nu
mb
er
Values
Frequency
Mean 10.7
Median 10.0
Mode 10.0
n 29Skewness 3.5Stdev 5.3Min 4.0Max 35.0
1/11/2016
69
When to Use Median
• Median is more appropriate when you have outliers.
• Consider a hypothetical salary survey of people in the room. Here are data points:
• $42K, $46K, $51K and $58K
• The mean salary is $49.25K and the median is $48.5K so both are fine
• But…
Cont’d
• Suppose Bill Gates enters the room, and is now part of our data set?
• $42K, $46K, $51K, $58K and $80 Billion
• $16 billion as the mean salary is not very representative of anyone
• $51K as the median at least represents most of the group, and is not influenced very much by the outlier
1/11/2016
70
When to use Mode
• The most two groups of individuals most susceptible to the flu virus are the very young and very old.
• The data forms a bi-modal distribution.
If you relied on the average age
for the best group to vaccinate,
you would end up with 39 year olds, which would completely miss the targe
Range
• This is the simplest measure of dispersion.
• It is the maximum value minus the minimum value.
• When data is normal, it is a fair appoximator of the spread of data, and it is simple to calculate
1/11/2016
71
Variation
• In the example on the previous slide, what is it we’re looking for? It’s called variation. There are two types:• Common cause variation
• Special cause variation
Common Cause Variation
• In an environment where you are producing goods, common cause variation is typically related to machines, design and the process.
• An oven which has temperature variation, causing variation in heat treat outcome.
• A drill press, which from wear has some issues holding concentricity due to wobble. Hole ranges from nominal to slightly oversized.
• One operator uses high pre-heat on a cutting operation, another does not (standardization/operation issue). Result is slightly different dimensions on cut parts.
1/11/2016
72
Common Cause in a Service Industry
• In an insurance industry, two adjusters are using two separate criteria for claims resulting in differences in payouts.
• A more experienced CSR can look up the information more quickly, which gives a significantly shorter time on hold for customer.
• Less proficient data entry clerks with higher levels of typing errors resulting in non-delivery to intended addressee (transposition of numbers).
Specially Assigned Cause
• Specially assigned cause is due to unusual circumstances. It’s generally harder to find simply because it doesn’t occur that often.
• An new operator, who is not aware of procedure, fills the fountain solution tank in a printing operation with city water, not from the RO valve. The printing press crew fights ink/water balance issues all night.
• Purchasing picks a new supplier of steel based on pricing. The old supplier ran to high side of industry tolerance, new supplier runs low. Fit up in assembly is now off.
• A bent drill bit which should have been discarded is chucked up in the drill. An oversized hole results.
1/11/2016
73
Specially Assigned Cause, Service Industry
• The computer is down, so CSR has to submit manual order tickets, making call times longer.
• Dr.’s handwritten prescription is hard to read, wrong medicine administered (this is extremely serious, so had better be specially assigned cause and not a routine occurrence).
Variation and Conformance to Specifications
• It would be impossible to eliminate all variation, but most of it can be controlled.
• So how much is an acceptable amount?
• Let’s consider this from the perspective of specification limits.
• A specification limit is the upper and lower tolerance that a customer will accept for a product or service.
• For our example, we will consider shoe size.
1/11/2016
74
Cont’d
• In this example, your foot is a perfect size 7.
• Your shoe can vary a little• Anything more than 7-1/4 and the shoe is too loose, causing blisters
• Anything smaller than 6-3/4 and the shoe is too tight, causing discomfort
Size 7
Size 7-1/8
Size 7-1/4
Size 6-7/8
Size 6-3/4
How histogram “creates the curve”
1/11/2016
75
Cont’dSize 6-3/4 Size 7-1/4
Size 7nominal
Upper specification limit
Lower specification limit
Purple bell curve represents variation
Cont’d
• So in the previous slide, the variation is less than our marginal values.
• We’d generally be happy with a pair of these shoes.
1/11/2016
76
New Example
Upper specification limit
Lower specification limit
Size 6-3/4 Size 7-1/4
Size 7nominal
Cont’d
• Now you can see that the process variation extends beyond the specification limits.
• Now as a consumer, there is a definite risk that we won’t accept the product.• There is even a possibility that one shoe is too large, and one is too small!
1/11/2016
77
Another Example
Upper specification limit
lower specification limit
Size 6-3/4 Size 7-1/4
Size 7nominal
What’s going on here?
• The process is not centered (running at nominal).
• As you can see, we are inducing defects in some of the parts from a lack of centering.
• This is a common mistake in many processes.
• Using the exact same graphics, check out the next slide:
1/11/2016
78
After Adjustment
Upper specification limit
Lower specification limit
Size 6-3/4 Size 7-1/4
Size 7nominal
Analyze
• During the analyze phase, there is little as a White Belt that you will be involved in. However, it is good to understand conceptually what higher level belts will be doing with all of the data you’ve collected. Primarily, they will look to see via experimentation (a before improvement and after improvement) if there is:
• a statistically significant difference
• a practical difference
1/11/2016
79
What is a statistically significant difference?
• Consider:
• I take a coin out of my pocket and flip it 10 times. The outcome is five heads.
• I take a second coin out of my pocket, flip it 10 times. The outcome is seven heads.
Analysis
• As you would rightly guess, there is no difference between the coins. It is just random chance that the second coin had a different outcome.
• There are statistical tests used to evaluate whether something is random chance or beyond chance.• We can’t have 100 percent certainty; but we like to test at a level where it is unlikely that we
will be wrong about our prediction.
1/11/2016
80
Practical Difference
• Some groups of before and after improvement are statistically different, but not practically different.
• A new machine can increase production, but the cost far outweighs gain in savings – so this is not practical.
• A new drug is shown (statistically) to extend life expectancy in terminally ill cancer patients by 12 days, but there are side detrimental side effects to the medication – so this would not be considered practical.
Cont’d
• We take a real world problem, model it statistically, and then make a real world decision based on the outcome of the test.
1/11/2016
81
Improving Processes Via Analysis
• So we basically make improvements two ways:
• Shifting the mean (process average)
• Reducing the variation
Improve and Control
• During this stage we take the results from our testing and begin find optimized settings.
• We also begin to standardize the process, and a plan to keep metrics at the new baseline (control).
1/11/2016
82
Standardized Work
• The only way one can maintain quality is to be repeatable and consistent.
• The only way one can make improvements is to have a repeatable and consistent process.
• The best way to be repeatable and consistent is through Standardized work. (already covered)
Checklist in Healthcare (common and in use)
• Patient’s vitals:• Blood pressure
• Pulse
• Circulation
• Etc.
1/11/2016
83
Control
• The control phase prevents our new level of performance from sliding back to former levels of operation.
• The control phase requires set of actions to be taken by specifically named people when levels dip.
• The person responsible for the control plan is the area “manager” in that they are most responsible for the operation. The control plan does not belong to the team.
Control Plan
• The control plan is a set of agreed upon conditions which if are not met will cause an action to take place.
• For example, if there has been an improvement project which keeps the line speed on a coating machine running at 1,200fpm, there may be a control plan that says if the speed of the machine drops below 1150fpm the team will be reconvened. The process owner is responsible fore exercising the control plan.
1/11/2016
84
Monitoring the Input As Well As The Output
This may have been the driver for the project, which is the output.
But notice that the input, time spent doing preventative maintenance, is the predictor.
This is the Y This is the X
Review
• Who determines what quality is?
• Is critical to quality an input or an output?
• Can you “inspect quality into a product?”
• What is the best strategy for agreeing upon an idea as a group?
• Is consensus a majority vote?
• Why is transportation considered a waste? (necessary, but non-value add)
• What is the purpose of a 5 Why?
• Name some of the elements of standard work:
• How does 5S improve safety?
1/11/2016
85
Cont’d
• A kan ban is like a _______________.
• What is DMAIC?
• Where does the problem statement go in a fishbone diagram?
• Give an example of a concentration study/measles chart.
• Why do we monitor the X’s in a control plan?