当前位置:首页 >> >>

6Sigma-DMAIC Problem Solving Tools


Data Collection

Communication

Brainstorming

DMAIC Problem Solving Tools
Solving Issues

Team work

DMAIC PROCESS
DEFINE

Define:
MEASURE

Measure: Analyze:

Select Critical To Customer (CTx) characteristics and performance standards. Create/validate measurement system to be used. Identify sources of variation from performance objectives. Discover process relations and establish new procedures. Implement process controls. Verify process controls are institutionalized Incorporate project lessons at other business units

ANALYZE

Improve: Control: End: Replicate:

IMPROVE

CONTROL

REPLICATE

END

6 SIGMA QUALITY ACTION DECK FOR ROMEO ENGINE PLANT
When solving an issue, complete each phase of D M A I C

D=Define.

Review Meters. Define your problem statement. Verify priority with Manager/SFE. Review DCP, QPS & FMEA. Refer to the SGA Problem Solving Tool Book for varoius problem Red=less than 50% of solving tools to use during this phase. Re-define your problem statement phase is complete. M=Measure. Develop and implement a data collection plan. Review gage R & R's, transducer verifications ect.

A=Analyze. I=Improve.

Analyze data that has been collected and place into appropriate chart forms. Potential Root Cause will be identified in this phase. Implement process improvement and validate that the team has reached its goal. **After the Improve phase is complete, verify that the DCP, QPS, FMEA & visual aids have been updated per the PCR.

Yellow=more than 50% of the phase is complete. Green=Phase is complete.

C=Control.

Monitor & adjust on going proformance. COMPLETE ALL FIELDS AND MAINTAIN A FOLDER FOR EACH ISSUE FOR EVIDENCE TRACKING. Last Reviewed Date: Line: Phase Date Opened Target Date Comp. Date

Issue# Station #

Target Measurable (JPH, FTT, etc)

Target / Actual Improvement (ex. 1 JPH)

Problem Statement

Root Cause

Corrective Action/Plan

Person Resp

D M A

I

C

Uncontrolled Document - For Ref. Only

GREEN BELT PROBLEM PRIORITY RESOLUTION
Department:__________ Operation:________ Station:________ Start Date:__________
Team Members ______________________ ______________________ ______________________ ______________________ Roles/Responsibilities __________________________ ___________________________ ___________________________ ___________________________ Team Members ____________________ ____________________ ____________________ ____________________ Roles / Responsibilities ________________________ ________________________ ________________________ ________________________

Define Phase:

Refine the understanding of the problem to be addressed and select critical to quality characteristics. What can be improved (Problem) ? _____________________________________________________________________________________ _____________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________ What is Critical to Quality (CTQ) ?______________________________________________________________________________________ _____________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________

Measure Phase: Establish techniques for collecting data about the current performance of processes identified in the Define Phase.

What Data needs to be collected ?__ ____________________________________________________________________________________ _____________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________ Validate the Measurement System (R&R)_________________________________________________________________________________ _____________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________ _____________________________________________________________________________________________________________________

Analyze Phase:

Target improvement opportunities by taking a closer look at the data and identify source(s) of variation: _____________________________________________________________________________________________________________________ ____________________________________________________________________________________________________________________ ____________________________________________________________________________________________________________________ ____________________________________________________________________________________________________________________ Identify Root Cause(s):_______________________________________________________________________________________________ ____________________________________________________________________________________________________________________ ____________________________________________________________________________________________________________________ ____________________________________________________________________________________________________________________ ____________________________________________________________________________________________________________________ Ask the 5 Whys:______________________________________________________________________________________________________ ____________________________________________________________________________________________________________________ ____________________________________________________________________________________________________________________ ____________________________________________________________________________________________________________________ ____________________________________________________________________________________________________________________

Improve Phase: Develop implement and validate improvement alternatives that will achieve desired performance levels as defined by CTQs Resolution / Action Plan: Task Responsibility Target Date 1._________________________________________________________________________________________________________________ 2._________________________________________________________________________________________________________________ 3._________________________________________________________________________________________________________________ 4__________________________________________________________________________________________________________________ 5__________________________________________________________________________________________________________________ 6__________________________________________________________________________________________________________________ 7__________________________________________________________________________________________________________________ 8__________________________________________________________________________________________________________________ 9__________________________________________________________________________________________________________________ 10_________________________________________________________________________________________________________________ Control Phase: Process improvements and monitor ongoing performance in order to sustain the gains achieved in the Improvement Phase.

What will be Monitored ? ___________________________________________________________________________________________ How will it be Monitored ?____________________________________________________________________________________________ ___________________________________________________________________________________________________________________ Who will Monitor ? ________________________________________________________________________________________________
For Reference Only - Uncontrolled Document

A3 - Problem Solving Worksheet
Date:___________________ Team Members:______________________________________________________ _______________________________________

Root Cause / Pareto / Control Charts
:

Problem Statement:
:

Measurement System Analysis / Gage R&R:
:

DOE - New Process Map - Pilot Brainstorming / Fishbone
:

:

Process Mapping - Line Walks:
: :

DCP/FMEA Updates, PM Task, Error Proofing

Uncontrolled Document

Question Log
Project________________
Ques. # Date Question Who to Ask Answer Date

For Reference Only

Project Closure Checklist
Black Belt/Green Belt: Project Champion: Deployment Champion: Process Owner: Master Black Belt: Team Members: Problem Statement: CTQs : Defects to be Reduced: Beginning DPMO/Sigma Level: Final DPMO/Sigma Level: Customer Satisfaction Impact: Value of Customer Satisfaction: Finance Signature Required:

Project Type: DMAIC DFSS

Define
Start Date:

Measure
Start Date:

Analyze
Start Date:

Improve
Start Date:

Control
Start Date:

Develop Problem Statement ID Project Scope ID Customers/CTQs Develop High Level Process Map Estimate Customer Satisfaction/COPQ Impact ID KPOV Metrics (Ys) Create Project Timeline Update Project Management System Project Champion Approval
End Date:

Brainstorm KPIVs (Xs) Develop Data Collection Plan Analyze and Validate Measurement System Perform or Review FMEA Collect Data Establish Baseline DPMO/Sigma Level ID Project Objectives Update Project Management System Project Champion Approval
End Date:

Micro Level Process Map Benchmark Quantify KPIVs & Improvement Opportunities Root Cause Analysis Define Performance Objectives Update Project Management System Project Champion Approval

Diagnose KPIV Performance Establish KPIV Performance Objectives ID Alternative Solutions Determine Optimal Solution Update FMEA Cost/Benefit Analysis Develop & Implement Pilot Plan Validate Improvements Update Project Management System Project Champion Approval

Apply Mistake Proofing Principles (If Applicable) Develop Control Metrics (SPC) on KPIVs Develop Monitor Plan (SPM) on KPOV Develop Long Term MSA Plan Establish or Update SOPs Establish or Update Training Plan Validate Control by Process Owner Establish BB Audit Plan with Time Line Update Project Management System Project Champion Approval
End Date:

End Date:

End Date:

JAR 7-12-00

D

M

A

I

C

The goal of the Define phase is to define the project's purpose and scope and obtain background information about the process and it's customers. Identify the CTQ'S (Critical to Quality) Characteristics.

DEFINE

PURPOSE: ?What is the problem? ?How big is it? (JPH, Downtime, # of faults) ?Where is it? ?Who is your customer? ?Internal or External? ?What is important to your customer? CT Trees TOOLS: (Not all need to be used) ? Meters or MMS ? Pareto Charts ? Run Charts ? CT Trees (CTQ, CTC, CTD) ? Fishbone ? C&E Matrix ? 5 Why’s
For Reference Only

? SIPOC ? Process Flow Mapping ? Y=f(X) ? Six Sigma Storyboard/Presentation ? Is / Is Not ? Kano Model

D

M

A

I

C

Tool Box Structure
? Each phase has an introduction statement, a purpose and a list of tools for that phase. ? Each of the tools has a level of difficulty associated with it:
Easy (should be familiar with now or can learn in one day) Intermediate (may need some help from Green Belt coordinator or BB Mentor) Advanced (may need assistance from BB mentor or Master BB)
For Reference Only

D

M

A

I

C

Define Phase (1 of 2)
TOOL
Meters / MMS

DESCRIPTION
Enables your team to stratify data and identify constraints Helps your team focus on the problems that are causing the most trouble. This helps to identify the areas where your efforts will have the biggest payback. Enables your team to study baseline process performance to identify trends or patterns over time. Critical to Quality Characteristics are features by which customers evaluate the quality of your product or service. The CTQ Tree enables your team to describe your customer's needs and the corresponding measurable characteristics. If a product or service does not meet a CTQ, it is considered defective. Allows the team to identify, explore, and graphically display all of the possible causes related to the problem. Used to identify the Process Inputs that have the greatest impact on the customer wants.

Pareto Chart Run Chart

CTQ Tree (Critical to Quality) Fishbone C&E Matrix
For Reference Only

D

M

A

I

C

Define Phase (2 of 2)
TOOL
5 Why’s

DESCRIPTION
Help get to the root cause quickly. Suppliers, Inputs, Process, Outputs, and Customers (SIPOC) analysis helps keep the product on track and helps promote successful results. Map out: 1. What we THINK it is. 2. What is REALLY is. 3. What is should be. Allows your team to structure the relationship among the Y's (the CTQ's) the y's (the process outputs directly affecting the Y's), and the x's (the casual factors directly affecting the Y's. Global 8D tool that helps better identify the problem by listing what the problem IS and what it IS NOT. Use to track progress throughout the project. Helps your team understand your customers' requirements, which are sorted into three categories: Must Be, More is Better, and Delighters.

SIPOC Process Mapping

Y=f(X) Formula Is / Is Not Six Sigma Storyboard

Kano Model
For Reference Only

D

M

A

I

C

Meters, Quality Worx, or MMS
? Use the systems within your departments to identify your constraints. ? Stratify the data from Meters or MMS using tools such as Pareto’s.

For Reference Only

D

M

A

I

C

Allows the Team to focus efforts on the problems that offer the greatest potential for improvement. ? The Pareto principle suggests that most effects come from 80% relatively few causes.
% Rejects

Pareto Chart

?

In quantitative terms: 80% of the problems come from 20% of the causes (machines, raw materials, operators etc.)

? ?

Therefore effort aimed at the right 20% can solve 80% of the problems. Double (back to back) Pareto charts can be used to compare 'before and after' situations.

10%

3% C

3% D

2% E

2% F

A

B

For Reference Only

D

M

A

I

C

Allows the Team to study data obtained from a checksheet or other datacollection source to analyze trends or patterns.
50% 40% 30% 20% 10% 0%
Jan Feb Mar Apr May Jun

Run Chart

Improvement Action A Improvement Action B

? Run charts display process performance over time. ? In a run chart, events, shown on the y axis, are graphed against a time period on the x axis. ? Run charts can also be used to track improvements that have been put into place, checking to determine their success. ? Also, an average line can be added to a run chart to clarify movement of the data away from the average.
For Reference Only

D

M

A

I

C

Allows the Team to identify CTQ characteristics or features, by which the customer evaluates your product or service.
Researched information returned quickly. Answers Given by Reps. Reps can answer questions asked by customer without further research. Customers greeted by name. Good Customer Service Friendly Reps

CT Trees (CTQ, CTC, CTD)

Knowledgeable Reps

Customers not interupted.

Time on hold is minimal. Short Wait Customers transferred immediately to the person who can help them.

General Hard to Measure
For Reference Only

Specific Easy to Measure

D

M

A

I

C

Allows the Team to identify, explore, and graphically display all of the possible causes related to the problem.
Measure Method Machine Problem Statement People ? ? ? ? ? Environment Materials

(Cause and Effect Diagram/Ishakawa Diagram)

Fishbone

The cause-and-effect diagram is a method for analyzing process dispersion. The diagram's purpose is to relate causes and effects. Effect = problem to be resolved, opportunity to be grasped, result to be achieved. Excellent for capturing team brainstorming output and for filling in from the 'wide picture'. Helps organize and relate factors, providing a sequential view. Click here for an example.

For Reference Only

Fishbone Cause Effect Diagram

The Fishbone is a brainstorming tool intended to display connections between a problem and the possible causes. General areas of concern are usually Method, Materials, Machines, Measure, Environment, & People, however, you can add or subtract these areas as needed for your project.
Work Process- How work is done
Method

Equipment/Tools used
Machine

Parts/Components used
Materials

Cause Cause

Sub Cause Sub Cause For each branch, ask, what the possible cause could be

Sub Cause

Everyone should agree on the effect or the problem statement before you begin.

For each cause, ask, Is there a sub cause?

Sub Cause

Problem statement
Cause Cause

People

Environment

Measure

Human Aspect -Training, Knowledge, Human error etc.

Buildings, Logistics, Space (Lighting, Air Quality, Temp

Data Collection (Meters data)/Calibration

1. Before beginning the fishbone, clearly identify the effect or the problem. This will be your Problem Statement and should be placed in the box on the right. (Head of the fish) 2. Identify the possible causes. - Breakdown into sub causes. Ask why does this happen. 3. Build the diagram. Place each cause under the appropriate title. Each sub cause under the appropriate cause. (The causes and sub causes will become the bones of the fish.) 4. Analyze the Diagram- Solutions may be identified at this time, discuss the cost, effectiveness, and feasibility of solutions.

For Reference Only

D

M

A

I

C

Cause & Effect Matrix
Used to identify the Process Inputs that have the greatest impact on the customer wants. Click here to
1. List outputs (Customer wants) 2. Rate outputs by importance to the customer using a rating scale of 1-3-9 (1 being low importance 3 being medium importance 9 high importance) 3. List inputs down the side of the Matrix. This step uses the Process map inputs directly 4. Correlate inputs to outputs. (Rate by importance) 5. Cross Multiply and Prioritize
(3x1)+(1x9)+(9x9)+(3x9) = 120
open Excel file below.

Rating of Importance to Customer

3
1 Verification of order acceptance

1
2

9
3

3
4 5 6

2
7 8 9 10 What the Customer wants 11 12 13 14

Delivery with in forty five days

24-hour notification of delivery

Online vehicle trcking

What in our process can effect the output

3
Process Inputs 1 Order System

1

The input with the highest number will be the one you focus on.
Total

9 3 1 1

3 9 9 3

1 9 9 3

1 1 9 9

42

2 Plant

3 Rail Carrier

4

5

102 120 60

6. Focus on the input that has the Total Highest Number

4 Mix center

5

For Reference Only

Cause Effect Matrix Rating of Importance to Customer 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Process Inputs

Total

1

2

3

4

5 Rating Legend: 9 = HIGH Importance 3 = MEDIUM Importance 1 = Low Importance
For Reference Only

D

M

A

I

C

5 Whys
Help you get to root cause quickly Start with your focused problem, then ask “WHY” 5 Times. BENEFITS ?This is good tool to use after the Fishbone has been completed. ? Use inputs from the fishbone to drill down to the possible root cause(s). ?Helps to determine relationships between different causes. ?Easiest tool to complete without getting into statistics. ?When is 5 Whys Most Useful? When problems involve human factors or interactions.
Click here for an example
For Reference Only

5 WHYs (Repeated Whys)

Customers complain about waiting too long to get connected to Problem Statement: staff when they call during lunch hours. 1. Ask Why? Why does this problem happen? Backup operators take longer to connect callers.

2.Why? Why does it take backup operators longer to connect callers? Backup operators don't know the job as well as the regular operators/ receptionists do.

3.Why? Why don't the backup operators know the job as well? There is no special training or job aids to make up for the gap in experience, and they receive all their training on the job. *The first 3 Whys are progressively deeper, more specific,actionable causes. The team should focus on the third why. 4.Why? Why don't they have special training or job aids? In the past, the organization has not recognized these needs.

5.Why? Why hasn't the organization recognized this need? The organization has no system to identify training needs.

*The forth and fifth whys are broader causes and are outside the scope of the problem-solving team and should be provived to management as data about systems issues that are deep causes of problems. *Where should you stop? What is the level at which you should take action? It depends on the scope of authority of the person or team investigating the issue. An immediate fix for the example given would be to provide at least SOME training for the current backup operators. * Remember to record all of your ideas. Some of these ideas can be used with other tools or possibly help with another project. BENEFITS *This is good tool to use after the Fishbone has been completed. Use inputs from the fishbone to drill down to the possible root cause(s). *Helps to determine relationships between different causes. *Easiest tool to complete without getting into statistics. When is 5 Whys Most Useful? When problems involve human factors or interactions.

For Reference Only

5 WHYs (Repeated Whys)

Problem Statement: 1. Why Does This Happen?

2.Why?

3.Why?

4.Why?

5.Why?

* Remember to record all of your ideas. Some of these ideas can be used with other tools or possibly help with another project.

For Reference Only

D

M

A

I

C

S I P O C – High Level Process Map

A SIPOC helps identify the key elements of your process and define the boundaries and scope of your project.

SUPPLIER

INPUTS

PROCESS

OUTPUT

CUSTOMERS

Romeo 2V/HP & Kellogg Cranks

Lube

Apply Loctite to Rear Face of Block Install Rear Seal Retainer Lube Crank Flange Install Rear Seal to Crank Test Rear Seal for Leaks Install Oil Slinger

NO LEAKS

Dealers

Cranks Intermet

Loctite Rear Seal Retainers Rear Seals

Car Buyers

Chicago Rawhide

Romeo Al Blk Romeo CI Blk

Blocks

Slinger

For Reference Only

D

M

A

I

C

Process Flow Mapping
? After completing a high level look at the process (SIPOC), walk the process using your flow charts from your DCP or create one if one does not exist. (Click here to see an example.) ? You may also use a Job Observation form at this point to ensure the task being done matches the existing process map. ? There are 3 types of process maps:
– What we THINK it is. – What it REALLY is. (Discover through line walks and job observation - includes hidden factory and waste) – What it SHOULD be. (Removes hidden factory and nonvalue added work.)

For Reference Only

Process Mapping
white
Process step or Operation (Flow Chart) ~A Process Map or Flow chart is a picture that shows all of the steps of a process or procedure. ~A Process Map is used to define and analyze processes by building a step-by-step picture for analysis, discussion, or communication purposes. ~A Process Map is also used to help define, standardize, or find areas for improvement in a process. ~A Process Map should show the entire process from start to finish,and everything in between showing inputs, pathways and circuits, actions or decision making points. ~The process is mapped using a set of symbols that are unique in color and shape, each identifying a different step in the process. ~Include Hidden Factories- Hidden Factory refers to fixing output that fails to pass the process. Repair bays are a good example. 1. 2. 3. 4. 5. Decision (Yes or No) Test Stations Repair Bays 6. Define the Process Boundaries – (Beginning to End) (Don’t forget the Hidden Factories) Define the Process Steps. Sort the steps into the order of occurrence. Place the steps into the Flow chart symbols and create the chart. Evaluate the chart for completeness, efficiency and possible problems. Don’t forget to identify your work. (Title, Date, and Names of Team Members)

Yellow
Quality check, Inspection Or measurement Quality audits in Repair Bays

Red
Delay

Yellow
Transport or movement Of material Or transmission of Information Shipping of engines

Yellow

Storage Connect steps Show Direction of Process

Symbols can also Be combined to represent two different actions

Blue

For Reference Only

Team Members_______________ _______________ _______________

Process__________

Date____________

Station # Or Process

Station # Or Process

Station # Or Process

Station # Or Process

Station # Or Process

Station # Or Process

Station # Or Process

YES

TEST Pass/Fail

NO YES
Ship Final Inspection

YES

Repair Bay

NO

NO
Major Repair

YES

Repair

Teardown

NO
Teardown Hidden Factory

For Reference Only

D

M

A

I

C

Output - the Head of the Fish Outcome of the Process

Y =?(X1+X2+…)
Inputs - Bones of the Fish Variables in the Process

For Reference Only

D

M

A

I

C

Y=f(X) Example
Y
LVL FTT =

f (X D-Line FTT+ X C-Line FTT + (…))

Y D-Line FTT = f (X sta 403 NPF + X C/C R/D + X sta 403 Leaks +(…)) Y C/C R/D = f (X Part quality+ X Spindle Align. + XBolts ZT + X Sockets)

For Reference Only

D

M

A

I

C

IS / IS NOT ? This is a tool from the Global 8D system. It helps to better identify the problem by listing what the problem is and what it is not. This will help keep you focused on scoping your problem down to the root cause. ? Click here for electronic version.

For Reference Only

Problem Solving Worksheet IS / IS NOT
1. Problem Statement 2. Problem Discription

IS

IS NOT

Get information

HOW BIG

WHEN

WHERE

WHAT

Uncontrolled Document For Reference Only

D

M

A

I

C

Six Sigma Storyboard / PowerPoint Presentation
? Use to track progress through out project ? Click here to see an example of a Six Sigma Storyboard from the Ford Intranet at the 6-Sigma website - Black Belt Resources page.

For Reference Only

D

M

A

I

C

Allows to Team to identify and prioritize the full range of the customers needs.

Kano Model

Neutral

?

Absent

Characteristic Presence

? The ‘Basic Musts’. Without this, you will never sell the product. E.g. A jet liner that cannot meet airport noise regulations. ? The ‘Dissatisfiers’, the things that cause your customers not to like your product. E.g. a jet airliner that is uncomfortable to ride in.
For Reference Only

Fulfilled

The ‘More is Better’. E.g. a jet airliner that uses a little less fuel than the competition.

Dissatisfaction

Customer Satisfaction

? The ‘Delighters’. These really make your product stand out from the others. Example, a passenger jet that could take off vertically.

Delight

D MEASURE

M

A

I

C

The goal of the Measure phase is to investigate the problem in more detail. You will be looking at the Outputs that affect your CTQ the most and which Inputs affect your Outputs the most. Collect data to create your performance baseline. Use this to track progress. You need to answer the following questions.

PURPOSE: What specifically is the problem? When specifically does it happen? Where specifically does it happen?

?Measurement System Analysis ?Gage R & R (2 options given) / ?Control Charts ?Data Collection (of baseline data) ?Detailed Process Map (includes floor walk) ?Histogram ?Operational Definitions ?Pareto Chart (continue to scope down the project) ?Run chart (looking for process variation/trending) ?Process Sigma / DPMO ?FMEA (can be used here as a place to start looking for clues)
For Reference Only

TOOLS: (Not all need to be used)

D

M

A

I

C

Tool Box Structure
? Each phase has an introduction statement, a purpose and a list of tools for that phase. ? Each of the tools has a level of difficulty associated with it:
Easy (should be familiar with now or can learn in one day) Intermediate (may need some help from Green Belt coordinator or BB Mentor) Advanced (may need assistance from BB mentor or Master BB)

For Reference Only

D

M

A

I

C

Measure Phase (1 of 2)
TOOL
MSA (Measurement Systems Analysis)

DESCRIPTION
Helps you understand measurement variation. - Includes Gage R&R. Helps you look for patterns over time in process variation, quantify the current capability of your process, and identify when special causes interrupt normal operations. Helps you systematically collect baseline data. Map out: 1. What we THINK it is. 2. What is REALLY is. 3. What is should be. Reveals how often a problem occurs in different settings. A stratified Histogram helps you identify process characteristics that might provide clues about the potential causes of the problems. Precise descriptions that describe how to get a value for each characteristic you are trying to measure.
For Reference Only

Control Charts Data Collection Detailed Process Map

Histograms Operational Definitions

D

M

A

I

C

Measure Phase (2 of 2)
TOOL DESCRIPTION
Displays the relative importance of problems. As in the Define step, it helps you focus your attention and develop a detailed problem statement. Plots data from Check Sheets and other sources and helps you look for patterns over time in process variation.

Pareto Chart

Run Chart

Process Sigma/ DPMO

Calculations that describe the current process capability. Calculating a baseline process sigma level provides a gauge for you to evaluate your progress.

FMEA

Review the FMEA to determine if any of the Potential Failures, already identified, offer any clues for your project.
For Reference Only

D

M

A

I

C

MSA - Measurement System Analysis
The Purpose of performing a Measurement System Analysis is to ensure the information collected is a true representation of what is occurring in the process. A common form of MSA is a Gage R&R. See the next few pages for samples. ? ? ? ? ? ? ? Is the measurement system capable for this study? How big is the measurement error? How much uncertainty should be attached to a measurement when interpreting it? Are the measurements being made with measurement units that are small enough to properly reflect the variation? What are the sources of measurement error? Can we detect process improvement if and when it happens? Is the measurement system stable over time?

For Reference Only

D

M

A

I

C

Gage R&R
A gage R&R is required to ensure that you can trust the data your gage gives you. Your gage needs to be able to do 2 things: 1. Repeat - which simply means that each time you check a particular part, you should get close to the same value. 2. Reproduce - which means that no matter which operator is using the gage, everyone will get the same value. There are (2) Types of Gage R&R’s: Variable- (e.g. Depth and TIR, Dial Gages) Attribute- (e.g. Go/ NO Go or Step Gages)

See the data collection examples on the next pages.

For Reference Only

D

M

A

I

C

Variable Gage R&R Sample Data Collection
Date: ___________________ Part #: _______________________________________ Characteristic to be checked: _____________________

Trial 1 Operator 1:
Part Number
1 3 8 2 7 6 4 5 9 10

Trial 2
Value Part Number
6 10 3 8 1 7 9 4 5 2

Value

?Use random order for each trial to eliminate potential bias during measurement. ?Expand the form as needed to increase the number of parts, trials, or operators. Click here for a link to this sample form.

Operator 2:
Part number
5 7 6 3 10 9 2 1 8 4

Value

Part Number
8 10 5 4 3 1 2 6 7 9

Value

For Reference Only

Gage R and R
Date: ___________________ Part #: _______________________________________ Characteristic to be checked: _____________________

Trial 1 Operator 1:
Part Number
1 3 8 2 7 6 4 5 9 10

Trial 2
Value Part Number
6 10 3 8 1 7 9 4 5 2

Value

Operator 2:
Part number
5 7 6 3 10 9 2 1 8 4

Value

Part Number
8 10 5 4 3 1 2 6 7 9

Value

Operator 3:
Part number
10 1 8 3 2 4 7 9 5 6

Value

Part Number
10 6 9 7 4 8 2 1 3 5

Value

07/30/2004

Page 1 of 1

D

M

A

I

C

Interpreting Variable Gage R&R’s
? After you have collected the data, you need to analyze it. You can do this 1 of 2 ways:
– 1. REP short form for gage R&R’s. – 2. Use Minitab to analyze it for you.

? REP Short Form:

– If you use the short form located on REPDOCS, the % R&R needs to be less than 14%. – If it is greater than 14%, you need to complete the long form.
? This will provide a larger sample size and give more information for evaluation.

? Minitab

– Minitab will automatically do the calculations for you. – It will break it out by trial, by part and by operator.

See next pages for examples.
For Reference Only

D

M

A

I

C

REP Short Form Variable Gage R&R Sample
A short form gage R&R can be used on an existing process. If the gage R&R results are >14%, use the long form gage R&R.

This is an example of an acceptable Gage R&R. Must be <14%.

For Reference Only

D

M

A

I

C

Minitab Sample Analysis
Components of Vari ati on
100
%C ontr ibution %Study Var
1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4

By Part

Pr et ec n

50

0
Gage R &R R epeat R epr od Par t- to- Par t

Part
1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4

1

2

3

4

5

6

7

8

9

10

R Chart by Operator
0.15
1

By Operator
3

2

Sml Rne a p ag e

U C L=0.1252
0.10 0.05 0.00 0

R =0.03833

LC L=0

Operat or

1

2

3

Xbar Chart by Operator
1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3
1

Operator*Part Interac ti on
3

2

Source
0

VarComp 0.004437 0.001292 0.003146 0.000912 0.002234 0.037164 0.041602

%Contribution (of VarComp) 10.67 3.10 7.56 2.19 5.37 89.33 100.00

Aeae vr g

U C L=0.8796 Mean=0.8075 LC L=0.7354

1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4

Operat or
1 2 3

Sml Ma a p en e

Part

1

2

3

4

5

6

7

8

9

10

Total Gage R&R Repeatability Reproducibility Operator Operator*Part Part-To-Part Total Variation

Failed
%Study Var (%SV) 32.66 17.62 27.50 14.81 23.17 94.52 100.00

Source Total Gage R&R Repeatability Reproducibility Operator Operator*Part Part-To-Part Total Variation

StdDev (SD) 0.066615 0.035940 0.056088 0.030200 0.047263 0.192781 0.203965

Study Var (5.15*SD) 0.34306 0.18509 0.28885 0.15553 0.24340 0.99282 1.05042

Minitab output gives graphical output and tabular. This provides the percentage for each category: % Contribution % Study Variation % Tolerance
For Reference Only

D

M

A

I

C

Six Sigma R&R Guidelines Acceptability Summary
Tabular Method

% Contribution 9% 1%

Process Control % Study Variation 30% 10%

Product Control % Tolerance 30% 10%

Number of Distinct Categories 5 10

Desirable to Have All 4 Indicators Green or Yellow
? 2001 Six Sigma Academy For Reference Only

D

M

A

I

C

Attribute Gage R&R Sample Data Collection
SCORING REPORT
Attribute Legend 1 pass 2 fail
5 (used in com putations)

DATE: NAME: PRODUCT: BUSINESS:
Operator #2 Try #1 Try #2 Operator #3 Try #1 Try #2

All operators agree within and between each Other All Operators agree with standard

Known Population Sample # Attribute 1 2 3 4 5 6 7 8 9 10

Operator #1 Try #1 Try #2

Y/N Agree

Y/N Agree

Examples of Attribute data would be: true/false, yes/no, go/no-go, etc. It is less precise than variable, so a typical sample size would be between 30 - 100 pieces. The following pages have information and links in order to use this form.
For Reference Only

D

M

A

I

C

Attribute Gage R&R Analysis

1. Looks at whether an operator agrees with himself on both trials. 2. Looks at whether an operator agrees with both trials AND the Known Attribute. 3. Looks to see if all operators agree within and between themselves. 4. Looks to see if all operators agree within and between themselves AND with the Known Attribute.
Click here for Link to blank Excel Attribute Gage R&R Form
For Reference Only

Data Entry

Attribute Gage R & R Effectiveness
SCORING REPORT
Attribute Legend 1 pass 2 fail
5 (used in computations)

DATE: NAME: PRODUCT: BUSINESS:
Operator #2 Try #1 Try #2 Operator #3 Try #1 Try #2

All operators agree within and between each Other All Operators agree with standard

Known Population Sample # Attribute 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Operator #1 Try #1 Try #2

Y/N Agree

Y/N Agree

% APPRAISER SCORE(1) -> % SCORE VS. ATTRIBUTE(2) ->

#DIV/0! Known

#DIV/0! Known

0% Known

SCREEN % EFFECTIVE SCORE(3) -> 0.00% SCREEN % EFFECTIVE SCORE vs. ATTRIBUTE (4) -> Note:
(1) (2) (3) (4) (5)

0.00%

Operator agrees with him/herself on both trials Operator agrees on both trials with the known standard All operators agreed within and between themselves All operators agreed within and between themselves AND agreed with the known standard Enter Pass/Fail, Good/Bad, Accept/Reject or other labels which indicate status of inspection

Page 1

D

M

A

I

C

Attribute Gage R&R Analysis
? ? ? A typical acceptance range for attribute data is between 80 – 100% (with 100% being the goal.) If the scores drop below 80%, evaluate training for the gage, gage function or sample size. The more samples, the better. The sample below shows all scores ≥80%.

For Reference Only

D

M

A

I

C

Allows the Team to monitor, control, and improve process performance over time.
Averages

Control Charts

Ranges

Upper Control Limit

Upper Control Limit

X

R

Batch Number

Batch Number

? ?

? ? ?

Control charts are a method of Statistical Process Control, SPC. Upper and lower control limits are calculated for a process and sampled measures are regularly plotted about a central line between the two sets of limits. The plotted line corresponds to the stability/trend of the process. Action can be taken based on trend rather than on individual variation. Prevents over-correction/compensation for random variation, which would lead to many rejects.
Measure Phase For Reference Only

D

M

A

I

C

Control Chart Trending Examples
Violation of the Control Limits
If a point falls above or below the control limit, the point needs to be circled & commented on the back of the chart. LCL

Process Shift

UCL

UCL X-BAR LCL

X-BAR

LCL

OOC

Process Shift

7th Point needs to be circled & commented on back of

Trending Up or Down

UCL

7th Point trending up needs to be circled & commented on back of chart

UCL

X-BAR LCL

X-BAR

LCL
7 Points in a row above the X-

For Reference Only

D

M

A

I

C

Data Collection Plans help you collect the right data for your needs and ensure the data you collect is useful and meaningful.
Data-Collection Plan
PROJECT_________________________________________________________________________ What questions do you want to answer? DATA What Measure type/ Data type Attribute: Count the frequency of occurance for a given process characteristic. *VALIDATE* How measured Go/No Go Pass/Fail Yes/No Categories Good/Defective Operational Definition and Procedures Related Conditons Sampling Notes How/Where Recorded to Record attach form Sample size: 50-100 pieces

Data Collection Plan

Variable: Describe the process characteristics in terms of it's weight, time, or size.

Continuous Data Time Length Width

Sample size: Min of 30 Pieces

How will you ensure consistency and stability? Remember to verify your collection method prior to putting it in place MSA--Gage R & R must be done on any tool being used. NOTES: 1) Include the unit of measurement where appropriate Be sure

What is your plan for starting data collection? (Attach details if necessary) When will you start/end the collection? Who will be responsible for collecting data? (filling in the forms) Are they trained to collect the Data accurately? Who will be responsible for collecting data forms as they are being filled in?
For Reference Only

Data-Collection Plan
PROJECT________________________________________________________________________ What questions do you want to answer? DATA What Measure type/ Data type Attribute: Count the frequency of occurance for a given process characteristic. *VALIDATE* How measured Go/No Go Pass/Fail Yes/No Categories Good/Defective Operational Definition and Procedures Related Conditons How/Where Recorded Sampling Notes to Record attach form Sample size: 50-100 pieces

Variable: Describe the process characteristics in terms of it's weight, time, or size.

Continuous Data Time Length Width

Sample size: Min of 30 Pieces

How will you ensure consistency and stability? Remember to verify your collection method prior to putting it in place MSA--Gage R & R must be done on any tool being used. NOTES: 1) Include the unit of measurement where appropriate. Be sure to test and monitor any measurement procedures/instruments. 2) Related conditions are stratification factors or potential causes you want to monitor as you collect data.

What is your plan for starting data collection? (Attach details if necessary) When will you start/end the collection? Who will be responsible for collecting data? (filling in the forms) Are they trained to collect the Data accurately? Who will be responsible for collecting data forms as they are being filled in? Don't forget to follow-up!! Check to be sure that everyone is following the procedure and that changes are made as needed. How will the data be displayed? (Sketch on additional sheet)

Be sure to collect the right data and ensure that the data you collect is useful and meaningful. ~Saves your Team time and effort ~Structures your data collection so everyone can understand what and how. ~Helps you think about how you can sort data in a manner that might provide clues about a problems cause.
For Reference Only

D

M

A

I

C

Process Flow Mapping
? After completing a high level look at the process (SIPOC), walk the process using your flow charts from your DCP or create one if one does not exist. (Click here to see an example.) ? You may also use a Job Observation form at this point to ensure the task being done matches the existing process map. ? There are 3 types of process maps:
– What we THINK it is. – What it REALLY is. (Discover through line walks and job observation - includes hidden factory and waste) – What it SHOULD be. (Removes hidden factory and nonvalue added work.)

For Reference Only

D

M

A

I

C

Allows the Team to summarize data from a process that has been collected over a period of time.
The plot to the left shows a normal distribution, where the plot below shows a bi-modal. This may generate some clues. Start asking ‘why’. 2 distributions

Histogram

? ? ? ?

A Histogram is a graphic summary of variation in a set of data. It enables us to see patterns that are difficult to see in a simple table of numbers. Can be analysed to draw conclusions about the data set. A histogram is a graph in which the continuous variable is clustered into categories and the value of each cluster is plotted to give a series of bars.
For Reference Only

D

M

A

I

C

Operational Definitions
Operational definition is a clear statement describing your problem or defect allowing you to measure it correctly. Having a clear operational definition eliminates confusion and will keep you focused on fixing the problem. Example: How many blue shirts are there in the room? As you look around the room, you can see that each person can have a different definition of a blue shirt. Some people may say any shirt that has blue on it – whether it's stripes, polka dots, solid or checked. What about the shade of blue? Do you need to define "shirt"? Short sleeved, long sleeved, length? When describing your problem, make sure you have established the operational definition with the team.

For Reference Only

D

M

A

I

C

Allows the Team to focus efforts on the problems that offer the greatest potential for improvement.
? The Pareto principle suggests that most effects come from relatively few causes. In quantitative terms: 80% of the problems come from 20% of the causes (machines, raw materials, operators etc.)
% Rejects

Pareto Chart

80%

?

10%

3% C

3% D

? ?

Therefore effort aimed at the right 20% can solve 80% of the problems. Double (back to back) Pareto charts can be used to compare 'before and after' situations.

1% E

1% F

A

B

For Reference Only

D

M

A

I

C

Allows the Team to study data obtained from a checksheet or other data-collection source to analyze trends or patterns.
50% 40% 30% 20% 10% 0%
Jan Feb Mar Apr May Jun

Run Chart

Improvement Action A Improvement Action B

? Run charts display process performance over time. ? In a run chart, events, shown on the y axis, are graphed against a time period on the x axis. ? Run charts can also be used to track improvements that have been put into place, checking to determine their success. ? Also, an average line can be added to a run chart to clarify movement of the data away from the average.
For Reference Only

D

M

A

I

C

Process Sigma / DPMO
Step 1:
? ? ? ? Number of units processed Total number of Defects made N = 500 D = 57 O=3

(include defects made and later fixed)

Number of defect Opportunities per unit Calculate DPMO using the formula below:
= Total number of defects Total units X Opp’s per unit X 1,000,000

Step 2:
DPMO

DPMO =

57 500 X 3

X 1,000,000

DPMO = 38,000

?

Step 3

Convert DPMO to a Sigma Value using a Z-Table Conversion Table. DPMO of 38,000 gives you a Sigma Value of 3.27

For Reference Only

D

M

A

I

C

Z-Value - DPMO - Cpk

a σ value _ X σ value

a

For Reference Only

D

M

A

I

C

Review the FMEA to determine if any of the Potential Failures already identified, offer any clues for your project.
P ro c e s s # : D e s c ript io n: LV L E ngine A s s e m bly P a rt # P a rt N a m e : P ro cess Functio n C l a s s

FMEA

Potential Failure Mode and Potential Effects of failure
S e v P otential O Cause(s)/ c Mechanism(s) c of Failure Current P rocess Controls P revention Current Process Controls Detection D e t R P N

Original Date:___________________________ Last Revisio n Date:______________________ Rev# : ________________________________ Dept # ________________________________ Dept Name:____________________________

Potential Failure Mode

Potential Ef f ect(s) of Failure

Recommended Action(s)

Requirements

Resp. & Target Complition Date

Action Results S O D c e e A ctions c v t Taken

R P N

See next page for further explanation.
For Reference Only

D
P ro c e s s # : D e s c ript io n: LV L E ngine A s s e m bly P a rt # P a rt N a m e : P ro cess Functio n C l a s s

M

A

I

C

Original Date:___________________________ Last Revisio n Date:______________________

Refer to the Rating Tables on the Following Pages

Rev# : ________________________________ Dept # ________________________________ Dept Name:____________________________

Potential Failure Mode

Potential Effect(s) of Failure

S e v

Requirements 340R Install Water P ump & H/S (4) B o lts- Install Idler wheel & R/D B o lt D01 Co rrect Water 9 P ump Installed

O Potential c Cause(s)/ Mechanism(s) c of Failure

Current Process Controls Prevention

Current Process Controls Detection

D e t

R Recommended Action(s) P N

Resp. & Target Complition Date

Action Results S O D e c e Actions v c t Taken

R P N

INSPECT
Water pump no t installed Engine reject at auto R/D statio n 1 0 OTH Operato r Erro r

DETECT
2 >Char Crtl 1 Operato r >Char Crtl: 2 P ro be : 1 00% Training/Visual aids HM I Display, part change ackno wledgement 2 40 No ne

TAKE ACTION
No ne

What is the effect(s)?

How bad is it?

What can be done? ~Design Changes ~Process Changes ~Special Controls

What can go wrong? ~No function
What is the Function?

What is the cause(s)?

Risk Priorty Number (RPN) = the SEVERITY times the OCCURANCES times the DETECTION If the # equals 100 or more the team should be looking for ways to improve the process. (Ie errorproofing, detection etc)

How often does it happen?

~Partial Function ~Intermittent Function

How can this be found?

How good is the method of finding it?

Table 4 Customer Impact Rating

Table 5 # of Times Failure has Occured

Table 6 Is there a process in place to detect the faiure and how reliable is the process
For Reference Only

D

M

A

I

C

ANALYZE
The Goal of the Analyze phase is analyze data collected and determine root cause(s). PURPOSE: ? To allow the Team to target improvement opportunities by taking a closer look at the data collected. TOOLS: (not all tools will be used) ? ? ? ? ? ? ? Brainstorming Fishbone Box Plot Histogram Scatter Diagram DOE (Design of Experiment) Hypothesis Testing

For Reference Only

D

M

A

I

C

Tool Box Structure
? Each phase has an introduction statement, a purpose and a list of tools for that phase. ? Each of the tools has a level of difficulty associated with it:
Easy (should be familiar with now or can learn in one day) Intermediate (may need some help from Green Belt coordinator or BB Mentor) Advanced (may need assistance from BB mentor or Master BB)

For Reference Only

D

M

A

I

C

Analyze Phase
TOOL
Brainstorming

DESCRIPTION
Enables your team to creatively and efficiently generate a large number of ideas about causes of errors. Enables your team to identify, explore, and graphically display, in increasing detail, all the possible causes related to a problem. The deeper you are able to push for causes, the more likely your solutions will be long-lasting ones. Generate clues from your data by graphically displaying the data. A stratified Histogram helps you identify process characteristics that might confirm patterns. Like Scatter Diagrams, they help you understand relationships that can confirm an underlying cause of a problem. Used to show the relationship between two variables. It can help your team verify causal relationships. Provides a method for testing multiple potential causes for error at the same time, enabling your team to reach conclusions about the primary causes. Verify Statistical differences in your data.
For Reference Only

Fishbone (Cause and Effect Diagram) Box Plot

Histogram Scatter Diagram DOE (Design of Experiments) Hypothesis Testing

D

M

A

I

C

Brainstorming
Brainstorming encourages open thinking when the team is stuck in "same old way" thinking. The common method for brainstorming is to write down the idea on a flip chart or dry erase board and each team member takes a turn giving ideas. Common rules: ? No idea is a bad idea – no criticism ? Write down all ideas ? 5 – 20 minutes is typically long enough (varies based on complexity of problem) ? Review the list and discard any duplicates Alternate 6-3-5 Method – Each member of the team (usually 6) has 5 minutes to write down 3 ideas on a sheet of paper. Then each person passes their sheet to the next person who then takes 5 more minutes to add three more ideas that build on the first three ideas. The rotation continues as many times as there are team members. After all ideas have been recorded, vote on the one to pursue first.
For Reference Only

D

M

A

I

C

Fishbone
Allows the Team to identify, explore, and graphically display all of the possible causes related to the problem.
Measure Method Machine Problem Statement People Environment Materials
(Cause and Effect Diagram/Ishakawa Diagram)

? ? ? ? ?

The cause-and-effect diagram is a method for analyzing process dispersion. The diagram's purpose is to relate causes and effects. Effect = problem to be resolved, opportunity to be grasped, result to be achieved. Excellent for capturing team brainstorming output and for filling in from the 'wide picture'. Helps organize and relate factors, providing a sequential view. Click here for an example.
For Reference Only

D

M

A

I

C

Box Plot
120 100 80 60 40 20 0 Group A Group B Group C
q1

min

median

max

q3

Look for clues using this tool. For example, look at Group C. That process has a shift in the median and is running with less variability. Start asking “Why”.

? ? ?

The outline of the box indicates the middle 50% of the data. The line inside the box represents the median. The lines extending on either side of the box, represent an upper limit and a lower limit.
For Reference Only

D

M

A

I

C

Allows the Team to summarize data from a process that has been collected over a period of time.
The plot to the left shows a normal distribution, where the plot below shows a bi-modal. This may generate some clues. Start asking ‘why’. 2 distributions

Histogram

? ? ? ?

A Histogram is a graphic summary of variation in a set of data. It enables us to see patterns that are difficult to see in a simple table of numbers. Can be analysed to draw conclusions about the data set. A histogram is a graph in which the continuous variable is clustered into categories and the value of each cluster is plotted to give a series of bars.
For Reference Only

D

M

A

I

C

Scatter Diagram
Relationship
50 40 30 20 10 0 0 5 10 15 20

No Relationship

?A scatter plot is effectively a line graph with no line - i.e. the point intersections between the two data sets are plotted but no attempt is made to physically draw a line. ?The Y axis is conventionally used for the characteristic whose behavior we would like to predict. ?Used to define the area of relationship between two variables - not cause and effect.
For Reference Only

D

M

A

I

C

Design of Experiment - DOE
? Running a DOE allows you to see the effect certain factors have on your process by purposely changing the factors and measuring your output. ? An example may be altering pressure and temperature in your process to see the effect on tool life:
RunOrder Pressure Temperature Tool life 1 3.5 200 2 2.5 200 3 2.5 200 4 2.5 150 5 3.5 150 6 3.5 150 7 3.5 200 8 2.5 150

* Contact a Black Belt to use this tool and analyze data.
For Reference Only

D

M

A

I

C

Hypothesis Testing
? Hypothesis testing is used to analyze data. It will help determine if there are statistically significant differences between 2 sets of data or populations. ? Hypothesis testing will help determine if there is a statistical difference in the data “before” a change or fix and “after.” – Used for : ? Variable Data
– Mean or Median (Target) – Variance (variation)

? Proportions Data
For Reference Only

D

M

A

I

C

Improve
The goal of the Improve phase is to develop, try out, and implement solutions that address root causes. PURPOSE:
? Allow Team to develop, implement, and validate improvement alternatives that will achieve the desired performance levels as defined by CTQ’s.

TOOLS: (not all tools will be used) ? Brainstorming ? Control Charts – SPC ? Histograms ? Pareto Charts ? Run Charts ? PDSA Cycle/Pilot ? FMEA ? Process Sigma
For Reference Only

D

M

A

I

C

Tool Box Structure
? Each phase has an introduction statement, a purpose and a list of tools for that phase. ? Each of the tools has a level of difficulty associated with it:
Easy (should be familiar with now or can learn in one day) Intermediate (may need some help from Green Belt coordinator or BB Mentor) Advanced (may need assistance from BB mentor or Master BB)

For Reference Only

D

M

A

I

C

Improve Phase (1 of 2)
TOOL
Brainstorming

DESCRIPTION
Enables your team to creatively and efficiently generate a large number possible solutions. In the Improve step, Control charts are used to show past and present performance of an indicator. Since Control Charts (and Run Charts) show plots of results as time passes, they are excellent tools for determining whether a solution has any real, lasting effect on your process. (Good or Bad) Comparing before and after, Histograms show how much progress has been made. As with Histograms, comparing before and after Charts is a way to objectively see how much progress has been made. Plots data from Check Sheets and other sources and helps you look for patterns over time in process variation.
For Reference Only

Control Charts Histograms

Pareto Chart

Run Chart

D

M

A

I

C

Improve Phase (2 of 2)
TOOL DESCRIPTION

PDSA Cycle/Pilot

The Plan-Do-Study-Act Cycle is the basic methodology behind a pilot. A pilot is a test of the whole system on a small scale to evaluate a solution and to make its fullscale implementation more effective.

FMEA

Used to anticipate potential problems, allowing your team to take counter-measures to reduce or eliminate risks.

Process Sigma

The new gauge of the effectiveness of any solution will show up in the process-sigma level.
For Reference Only

D

M

A

I

C

Brainstorming
Brainstorming encourages open thinking when the team is stuck in "same old way" thinking. The common method for brainstorming is to write down the idea on a flip chart or dry erase board and each team member takes a turn giving ideas. Common rules: ? No idea is a bad idea – no criticism ? Write down all ideas ? 5 – 20 minutes is typically long enough (varies based on complexity of problem) ? Review the list and discard any duplicates Alternate 6-3-5 Method – Each member of the team (usually 6) has 5 minutes to write down 3 ideas on a sheet of paper. Then each person passes their sheet to the next person who then takes 5 more minutes to add three more ideas that build on the first three ideas. The rotation continues as many times as there are team members. After all ideas have been recorded, vote on the one to pursue first.
For Reference Only

D

M

A

I

C

Allows the Team to monitor, control, and improve process performance over time.
Averages

Control Charts

Ranges

Upper Control Limit

Upper Control Limit

X

R

Batch Number

Batch Number

? ?

? ? ?

Control charts are a method of Statistical Process Control, SPC. Upper and lower control limits are calculated for a process and sampled measures are regularly plotted about a central line between the two sets of limits. The plotted line corresponds to the stability/trend of the process. Action can be taken based on trend rather than on individual variation. Prevents over-correction/compensation for random variation, which would lead to many rejects.
Improve Phase For Reference Only

D

M

A

I

C

Control Chart Trending Examples
Violation of the Control Limits
If a point falls above or below the control limit, the point needs to be circled & commented on the back of the chart. LCL

Process Shift

UCL

UCL X-BAR LCL

X-BAR

LCL

OOC

Process Shift

7th Point needs to be circled & commented on back of

Trending Up or Down

UCL

7th Point trending up needs to be circled & commented on back of chart

UCL

X-BAR LCL

X-BAR

LCL
7 Points in a row above the X-

For Reference Only

D

M

A

I

C

Allows the Team to summarize data from a process that has been collected over a period of time.
The plot to the left shows a normal distribution, where the plot below shows a bi-modal. This may generate some clues. Start asking ‘why’. 2 distributions

Histogram

? ? ? ?

A Histogram is a graphic summary of variation in a set of data. It enables us to see patterns that are difficult to see in a simple table of numbers. Can be analysed to draw conclusions about the data set. A histogram is a graph in which the continuous variable is clustered into categories and the value of each cluster is plotted to give a series of bars.
For Reference Only

D

M

A

I

C

Allows the Team to focus efforts on the problems that offer the greatest potential for improvement.
? The Pareto principle suggests that most effects come from relatively few causes. In quantitative terms: 80% of the problems come from 20% of the causes (machines, raw materials, operators etc.) Therefore effort aimed at the right 20% can solve 80% of the problems. Double (back to back) Pareto charts can be used to compare 'before and after' situations.
For Reference Only

Pareto Chart

80%

% Rejects

?

10%

? ?

3% C

3% D

1% E

1% F

A

B

D

M

A

I

C

Allows the Team to study data obtained from a checksheet or other data-collection source to analyze trends or patterns.
50% 40% 30% 20% 10% 0%
Jan Feb Mar Apr May Jun

Run Chart

Improvement Action A Improvement Action B

? ? ? ?

Run charts display process performance over time. In a run chart, events, shown on the y axis, are graphed against a time period on the x axis. Run charts can also be used to track improvements that have been put into place, checking to determine their success. Also, an average line can be added to a run chart to clarify movement of the data away from the average.
For Reference Only

D

M

A

I

C

PDSA Cycle/Pilot
Use to evaluate and improve the solutions you want to implement. PLAN ? Study the problem and PLAN a solution. DO ? Implement the solution. STUDY ? Study the results of the changes implemented. ACT ? When the study shows that root cause has been isolated and removed, the changes that you implemented, are made permanent.

ACT STUDY

PLAN DO

For Reference Only

D

M

A

I

C

Review the FMEA to determine if any of the Potential Failures already identified, offer any clues for your project.
P ro c e s s # : D e s c ript io n: LV L E ngine A s s e m bly P a rt # P a rt N a m e :

FMEA

Potential Failure Mode and Potential Effects of failure
S e v
C l a s s

Original Date:___________________________ Last Revisio n Date:______________________ Rev# : ________________________________ Dept # ________________________________ Dept Name:____________________________

P ro cess Functio n

Potential Failure Mode

Potential Ef f ect(s) of Failure

Requirements

O Potential c Cause(s)/ Mechanism(s) c of Failure

Current Process Controls Prevention

Current Process Controls Detection

D e t

R P N

Recommended Action(s)

Resp. & Target Complition Date

Action Results S O D e e c A ctions t v c Taken

R P N

See next page for further explanation.
For Reference Only

D
P ro c e s s # : D e s c ript io n: LV L E ngine A s s e m bly P a rt # P a rt N a m e : P ro cess Functio n C l a s s

M

A

I

C

Original Date:___________________________ Last Revisio n Date:______________________

Refer to the Rating Tables on the Following Pages

Rev# : ________________________________ Dept # ________________________________ Dept Name:____________________________

Potential Failure Mode

Potential Effect(s) of Failure

S e v

Requirements 340R Install Water P ump & H/S (4) B o lts- Install Idler wheel & R/D B o lt D01 Co rrect Water 9 P ump Installed

O Potential c Cause(s)/ Mechanism(s) c of Failure

Current Process Controls Prevention

Current Process Controls Detection

D e t

R Recommended Action(s) P N

Resp. & Target Complition Date

Action Results S O D e c e Actions v c t Taken

R P N

INSPECT
Water pump no t installed Engine reject at auto R/D statio n 1 0 OTH Operato r Erro r

DETECT
2 >Char Crtl 1 Operato r >Char Crtl: 2 P ro be : 1 00% Training/Visual aids HM I Display, part change ackno wledgement 2 40 No ne

TAKE ACTION
No ne

What is the effect(s)?

How bad is it?

What can be done? ~Design Changes ~Process Changes ~Special Controls

What can go wrong? ~No function
What is the Function?

What is the cause(s)?

Risk Priorty Number (RPN) = the SEVERITY times the OCCURANCES times the DETECTION If the # equals 100 or more the team should be looking for ways to improve the process. (Ie errorproofing, detection etc)

How often does it happen?

~Partial Function ~Intermittent Function

How can this be found?

How good is the method of finding it?

Table 4 Customer Impact Rating

Table 5 # of Times Failure has Occured

Table 6 Is there a process in place to detect the faiure and how reliable is the process
For Reference Only

D

M

A

I

C

Process Sigma / DPMO
Step 1:
? ? ? ? Number of units processed Total number of Defects made N = 500 D = 57 O=3

(include defects made and later fixed)

Number of defect Opportunities per unit Calculate DPMO using the formula below:
= Total number of defects Total units X Opp’s per unit X 1,000,000

Step 2:
DPMO

DPMO =

57 500 X 3

X 1,000,000

DPMO = 38,000

?

Step 3

Convert DPMO to a Sigma Value using a Z-Table Conversion Table. DPMO of 38,000 gives you a Sigma Value of 3.27

For Reference Only

D

M

A

I

C

CONTROL
The Goal of the Control phase is to maintain the gains made by standardizing work methods or process.

PURPOSE:
? ? Institutionalize process/product improvements, including error proofing, and monitor ongoing performance in order to sustain the gains achieved in the Improve Stage. Develop DCP, FMEA and Reaction Plans.

TOOLS: (not all tools will be used)
? ? ? ? ? Run Charts Error Proofing PDSA Cycle Control Charts/Mini Tab FMEA ? ? ?

DCP - Dynamic Control Plans Preventive Maintenance Six Sigma Storyboard/Presentation

For Reference Only

D

M

A

I

C

Tool Box Structure
? Each phase has an introduction statement, a purpose and a list of tools for that phase. ? Each of the tools has a level of difficulty associated with it:

Easy (should be familiar with now or can learn in one day) Intermediate (may need some help from Green Belt coordinator or BB Mentor) Advanced (may need assistance from BB mentor or Master BB)
For Reference Only

D

M

A

I

C

Control Phase (1 of 2)
TOOL
Run Chart Error Proofing

DESCRIPTION
Monitors progress over time after a project is completed. It is typically part of the Process Management Chart. Devices that are used to catch various types of defects. Serves as a reminder to think of improvement as being continual: Where can you go next to make the process even better? Monitor progress over time after your project is completed. They can help your team continually quantify the capability of your process and identify when special causes interrupt normal operations. It is typically part of the Process Management Chart. Used to anticipate potential problems, allowing your team to take counter-measures to reduce or eliminate risks. Used to show how potential problems are controlled in your process - Prevented or Detected
For Reference Only

PDSA Cycle

Control Charts FMEA DCP (Dynamic Control Plans)

D

M

A

I

C

Control Phase (2 of 2)
TOOL DESCRIPTION

PM (Preventative Maintenance)

Used to schedule tasks in order to monitor equipment / process for maintenance prior to failure.

Six Sigma Storyboard

A pictorial record of your project. Typical documents include a succinct final report and a completed Storyboard that captures the project in a graphical form. (Powerpoint Presentation)
For Reference Only

D

M

A

I

C

Allows the Team to study data obtained from a checksheet or other datacollection source to analyze trends or patterns.
50% 40% 30% 20% 10% 0%
Jan Feb Mar Apr May Jun

Run Chart

Improvement Action A Improvement Action B

? ? ? ?

Run charts display process performance over time. In a run chart, events, shown on the y axis, are graphed against a time period on the x axis. Run charts can also be used to track improvements that have been put into place, checking to determine their success. Also, an average line can be added to a run chart to clarify movement of the data away from the average.
For Reference Only

D

M

A

I

C

Error Proofing
? Error proofing devices are identified, and installed in critical places in the process to catch various types of defects. ? Probes, scanners, cameras, and lasers are some of the options. ? Daily challenges are done to verify the integrity of each device to prevent these defects from being passed on to the customer. ? A challenge may entail placing a bad part on a good engine and running it through the error proofing device.
For Reference Only

D

M

A

I

C

PDSA Cycle/Pilot
Use to evaluate and improve the solutions you want to implement. PLAN ? Study the problem and PLAN a solution. DO ? Implement the solution. STUDY ? Study the results of the changes implemented. ACT ? When the study shows that root cause has been isolated and removed, the changes that you implemented, are made permanent.

ACT STUDY

PLAN DO

For Reference Only

D

M

A

I

C

Allows the Team to monitor, control, and improve process performance over time.
Averages

Control Charts

Ranges

Upper Control Limit

Upper Control Limit

X

R

Batch Number

Batch Number

? ?

? ? ?

Control charts are a method of Statistical Process Control, SPC. Upper and lower control limits are calculated for a process and sampled measures are regularly plotted about a central line between the two sets of limits. The plotted line corresponds to the stability/trend of the process. Action can be taken based on trend rather than on individual variation. Prevents over-correction/compensation for random variation, which would lead to many rejects.
Control Phase For Reference Only

D

M

A

I

C

Control Chart Trending Examples
Violation of the Control Limits
If a point falls above or below the control limit, the point needs to be circled & commented on the back of the chart. LCL

Process Shift

UCL

UCL X-BAR LCL

X-BAR

LCL

OOC

Process Shift

7th Point needs to be circled & commented on back of

Trending Up or Down

UCL

7th Point trending up needs to be circled & commented on back of chart

UCL

X-BAR LCL

X-BAR

LCL
7 Points in a row above the X-

For Reference Only

D

M

A

I

C

Review the FMEA to determine if any of the Potential Failures already identified, offer any clues for your project.
P ro c e s s # : D e s c ript io n: LV L E ngine A s s e m bly P a rt # P a rt N a m e : P ro cess Functio n C l a s s

FMEA

Potential Failure Mode and Potential Effects of failure
S e v P otential O Cause(s)/ c Mechanism(s) c of Failure Current P rocess Controls P revention Current Process Controls Detection D e t R P N

Original Date:___________________________ Last Revisio n Date:______________________ Rev# : ________________________________ Dept # ________________________________ Dept N ame:____________________________

Potential Failure Mode

Potential Ef f ect(s) of Failure

Recommended Action(s)

Requirements

Resp. & Target Complition Date

Action Results S O D c e e A ctions c v t Taken

R P N

See next page for further explanation.
For Reference Only

D
P ro c e s s # : D e s c ript io n: LV L E ngine A s s e m bly P a rt # P a rt N a m e : P ro cess Functio n C l a s s

M

A

I

C

Original Date:___________________________ Last Revisio n Date:______________________

Refer to the Rating Tables on the Following Pages

Rev# : ________________________________ Dept # ________________________________ Dept Name:____________________________

Potential Failure Mode

Potential Effect(s) of Failure

S e v

Requirements 340R Install Water P ump & H/S (4) B o lts- Install Idler wheel & R/D B o lt D01 Co rrect Water 9 P ump Installed

O Potential c Cause(s)/ Mechanism(s) c of Failure

Current Process Controls Prevention

Current Process Controls Detection

D e t

R Recommended Action(s) P N

Resp. & Target Complition Date

Action Results S O D e c e Actions v c t Taken

R P N

INSPECT
Water pump no t installed Engine reject at auto R/D statio n 1 0 OTH Operato r Erro r

DETECT
2 >Char Crtl 1 Operato r >Char Crtl: 2 P ro be : 1 00% Training/Visual aids HM I Display, part change ackno wledgement 2 40 No ne

TAKE ACTION
No ne

What is the effect(s)?

How bad is it?

What can be done? ~Design Changes ~Process Changes ~Special Controls

What can go wrong? ~No function
What is the Function?

What is the cause(s)?

Risk Priorty Number (RPN) = the SEVERITY times the OCCURANCES times the DETECTION If the # equals 100 or more the team should be looking for ways to improve the process. (Ie errorproofing, detection etc)

How often does it happen?

~Partial Function ~Intermittent Function

How can this be found?

How good is the method of finding it?

Table 4 Customer Impact Rating

Table 5 # of Times Failure has Occured

Table 6 Is there a process in place to detect the faiure and how reliable is the process
For Reference Only

D

M

A

I

C

DCP - Dynamic Control Plans

This is an example of a control plan. If you have made changes to your process as part of improvement actions, this needs to be updated as well as the FMEA. Improvement actions should attempt to move controls from Prevention to Detection if the defect cannot be eliminated altogether. Be sure to look at the Special Characteristic Classification.
For Reference Only

D

M

A

I

C

Preventative Maintenance
? If the root cause you have identified requires scheduled maintenance to sustain the improvements, meet with your maintenance department to develop a permanent schedule.
– For example, if one of your X’s in Y = f(X) is socket wear, the team would need to determine cycle life and place the socket on a regular schedule to be changed. – Other examples of PM items would be seals, hoses, error proofing devices or any part touching details.

For Reference Only

D

M

A

I

C

Six Sigma Storyboard / PowerPoint Presentation
? Use to track progress through out project ? Click here to see an example of a Six Sigma Storyboard from the Ford Intranet at the 6-Sigma website - Black Belt Resources page.

For Reference Only


相关文章:
Course Outline_DMAIC the way to solve problem_Chine...
2页 1财富值 6Sigma-DMAIC Problem Sol... 99页 20财富值... Course Outline_DMAIC the way to solve problem_Chinese_Rev_03.10 隐藏>>...
论文模板
Six Sigma DMAIC problem-solving paths and lean manufacturing and Six Sigma ...For many, lean is the set of "tools" that assist in the identification ...
11 Quality - Problem Solving Tools
Problem Solving & 7 QCC ... 暂无评价 38页 5财富值 On tools for analysis of... 2页 免费 6Sigma-DMAIC Problem Sol... 99页 20财富值如要投诉...
6sigma复习
是产品质量和过程能力的衡量指标,目标是每百万次机会 3.4 个缺陷,即短期σ水 平达到 5、的关注点:过程输入变量 6、DMAIC 和 DMADV 的区别 新型...
以六西格玛管理的DMAIC模型对人力资源培训与开发管理进...
六西格玛管理的DMAIC模型对人力资源培训与开发管理进行分析_经管营销_专业资料。以六西格玛管理的DMAIC模型对人力资源培训与开发管理进行分析...
6 Sigma 简介
这一目标要通过两套 6 Sigma 系统方法来完成: DMAIC 和 DMADV。6 Sigma DMAIC 过程(D-定义、M-测量、A-分析、I-改善和 C-控制)主要针对不能满足要求的过程...
【精编完整版】六西格玛DMAIC模型在变速器噪声中的应用毕业论文...
项目组通过建立 6sigma DMAIC(Define,Measure, Analyze,Improve,Contro1)业绩改进...(VOC), and conduct research for the customer to reflect the problems to ...
6sigma全套课程1
· 是一种哲学,是方法,也是战略 中所用的 DMAIC 业绩改进模型 ...(1) 6SIGMA 管理的计划和实施(上)□ 王金德 6SIGMA 管理其实是一项以顾客为...
6Sigma教程
也是战略 中所用的 DMAIC 业绩改进模型 Define — Measure — Analyze —...从 20 世纪 90 年代起, 6SIGMA 这个使用频率日益增多的词汇引 起了企业界和...
Six Sigma六西格玛(管理必看)
六西格玛是一项以数据为基础,追求几乎完美的质量管理方法。西格玛是一个希腊字母 ...(DMAIC) * 确立运营流程的计量标准,并进行严格和连续不断的检验 * 深入见效的...
更多相关标签: