Rootcauseanalysispresentation 090728111941 Phpapp01

Module Outline Chronic vs. Sporadic Problems Improvement (Breakthrough) vs. Troubleshooting Addressing Chronic Problems - Breakthrough Definitions Diagnosis Why “Root Cause”? Tie to MBF (Management by Fact) Addressing Sporadic Problems - Troubleshooting Summary Resources ?

Module Objectives You will be able to: Distinguish between chronic vs. sporadic problems. Define Problem/Symptoms/Causes/Diagnosis. Describe approaches to find root causes of chronic problems. Describe the link between Root Cause Analysis and MBF. Describe approaches to find root causes of sporadic problems.

Improvement vs. Fire Fighting Sporadic Chronic Gain Time Performance Good Bad

Chronic Problems Existed for some time Usually “lived with” or accepted Cost more $$$$$ than sporadic problems Don’t sound alarms (allowances for problem have been made in the system) Competitors may be enduring same losses Improvement is needed when a chronic problem is occurring Requires improvement PROJECTS - investment of time and resources Fire Prevention - attack the chronic level

Sporadic Problems A “spike” in performance “ Alarm” usually goes off Troubleshooting used when a sporadic problem happens (a departure from the normal level) Aimed at restoring the status quo May include Interim (temporary) fixes Corrective Action - eliminate the cause of the problem Remove the change that produced the deviation or new change to offset deviation Fire Fighting - focus on control

Improvement vs. Troubleshooting Improvement: Focused on attaining a new level of performance that is superior to any previous level. Accomplished through BREAKTHROUGH. Improves chronic level of performance. Fire Prevention Troubleshooting: Focused on solving a sudden change in performance. Accomplished with CORRECTIVE ACTION to put out a fire. Improves sporadic level of performance. Fire Fighting

Sequence for Breakthrough Proving the need Identifying Projects Organizing project teams Establishing project (mission/vision/responsibilities) Diagnosis - to find the causes Developing remedies - based on knowledge of the causes Proving remedies & effectiveness - under operating conditions Dealing with resistance to change Controlling at the new level Focus Here Set-up Carry-out

Definitions Problem: Any deviation from the standard, expected, or desired which is outside the accepted tolerance, norm, or benchmark. Project: a problem scheduled for solution! Symptom: An observable indicator, cue, or event directing attention to a problem. Arising from and accompanying a problem. Theory: Unproved assertion as to reasons for the problem and symptoms.

Definitions - Continued Cause: “Something” that happened to produce a deviation of the actual from the expected or desired. Proved reason for existence of problem. Often “multiple causes”. Dominant or Root Cause: a major contributor to existence of problem which must be fixed before there is an adequate solution. Remedy: a change that can successfully eliminate or neutralize the cause of a problem.

Definition - Diagnosis Diagnosis: the process of studying symptoms, theorizing as to causes, testing theories, and discovering causes. Diagnosis - to find the causes Diagnostic Journey = “symptom to cause” Followed by Remedial Journey = “cause to remedy” “ Symptom to cause” is DIFFICULT!

Diagnosis for Improvement Diagnostic Journey: 1. Study the symptoms surrounding the problem to serve as a basis for theorizing about causes. 2. Theorize the causes of these symptoms. 3. Collect data and do analysis to test the theories and determine the causes. When the problem can be switched on & off at will - the journey is over!

Studying Symptoms Two forms of “Evidence”: 1. Words/documentation describing problem 2. Physical measurements & analysis of product/process Agreement on “terms/descriptions” for symptoms Recording data - established to test specific theories Quantifying symptoms: Frequency Intensity Use of Pareto Charts/Analysis

Check Sheets Concentration Diagram “ Autopsies” Tools for Symptoms “ Glossaries” for symptom terms Pareto Analysis Check Sheet Example: Scratches IIII II II IIII I 15 Runs IIII IIII IIII IIII IIII I IIII IIII 34 Dirt IIII I IIII 11 Wrong Color IIII I I 7 Paint Peeling I I 2 Total 32 15 22 69 Problem Total Order 1 Order 2 Order 3

Theorizing Causes Diagnosis is made theory by theory Need to affirm or deny the validity of theories This is “IDENTIFYING POTENTIAL CAUSES” 3 Steps 1. Generate Theories e.g. Brainstorm, Ask Why 5 times 2. Arrangement of Theories e.g. Matrix, Tables, Cause & Effect Diagrams 3. Choosing Theories to be Tested e.g. Data Collection, Pareto Analysis Establish Priorities for testing theories

Ask Why 5+ times Brainstorming* Cause & Effect Diagrams* Force Field Analysis* Affinity Diagram* Structure Tree Diagram* Interrelationship Digraph* Program Decision Process Chart (PDPC)* Matrix Diagram* Check Sheet* Pareto Analysis* Designed Experiments ETC. Tools for Theorizing Causes * Included in Memory Jogger II Pareto Chart Example:

Tool Example - Ask Why 5 Times Generate Theory: Follow with Remedy! Failures from Supplier A are 2X the industry average. There is excessive damage in transit. Packaging is insufficient. Packaging specifications are incomplete. Why? Why? Why? Why? Real Root Cause There are no technical specifications for packaging available. Why?

Tool Example - Cause & Effect Generate Theory: Follow with Test of Theories based on priorities Out of Date Not Manufacturable Part Problem Cutting oil Bar Stock Capability Scheduling cpk Tolerances Procedures Set-up Maintenance Cost Inaccurate Training Engineering Support Operator Errors Adjustments Materials Machinery Methods Manpower Wrong Material Over Sized Wrong Oil Packaging Print Inaccurate

Tool Example - Structure Tree Elbow Failures Problem Subproblems Theories Most Likely Root Causes Crimp makes inadequate contact Cross Thread Pin not fully seated Cable not fully seated Defective Elbow Operator Material Methods Tools & Equipment Guide Thread Lack of Verification for Alignment # of Revolutions not counted

Testing Theories to Find Cause Determine which cause is ROOT cause Identifying most likely cause DATA COLLECTION & ANALYSIS! Methods 1. Product/Process “Dissection” e.g. Flow Map, Process Capability Study, Time to Time 2. Collecting New Data e.g. Measuring within the process 3. Experiments - create & process trials to test validity e.g. Designed experiments, pilots, trials 4. Measure for Diagnosis e.g. Variables vs. attributes, increase precision, new methods

Historical Review Flow Diagram/Map* Process Capability Analysis Time to Time Analysis Stream to Stream Analysis Histograms* Control Charts* Piece to Piece Analysis Within Piece to Piece Analysis Multi-Vari Analysis Check Sheet* Pareto Analysis* Designed Experiments Correlation Studies Measurements within process Study of Work Methods Measurement Capability Studies ETC. Tools for Testing Theories * Included in Memory Jogger II

Tool Example - Histogram & Run Chart Test Theory: Follow with Remedy! Histograms: Run Charts:

Tool Example - Capability Test Theory: - Structured Experimentation Follow with Remedy! Goal: Collect data on all parts of the process so that: Capture all contributors to variability Variability can be partitioned into parts due to each contributor = Total Machine Cavity Time Sample + + +

Diagnosis of Failures in Systems Historical Review - objective to improve segment of system used to attain quality e.g.: Product Development, Supplier Relations Based on analysis of # of past problems Interview and Document: 1. Brief description of problem 2. Stage where problem 1st observed 3. Earliest stage at which problem COULD have been discovered 4. Reasons for NOT discovering problem at earlier stage 5. What could have been done to have found the problem earlier or avoided it completely

Diagnosis of Failures in Systems Examples: 1. Historical review of past product development problems: Symptoms & Cause: Review revealed use of FMEA/FMECA (Failure Mode, Effect, and Criticality Analysis) would have identified problems upfront. Remedy: Require FMEA upfront for new products. 2. Historical review of manufacturing defects: Symptoms & Cause: Review revealed lack of set-up procedure led to high lot to lot variation and defects. Remedy: Document and implement set-up procedure.

Diagnostic Journey Root Cause Analysis Begins with Collecting Data on Symptoms Ends with agreement on Cause(s)

Why Find the “Root Cause”?

What is MBF Used For? Address gaps in business performance Customer Satisfaction Employee Satisfaction Business models Operational processes Manufacturing processes Simplify reporting Link problems and root causes with action and results Link a reporting format to a problem solving model Report on projects that address a specific problem

Management by Fact (MBF) Remedies Root Cause Analysis Diagnosis Direction Setting Direction Deployment Management Process Policy Deployment & Management by Fact Policy Deployment Monitor Performance Gap Detected No Yes

2. Root Cause Categorization / Analysis (RCA ) Analyze Problem 2 Separate “beliefs” from “real” problem Determine, categorize, and analyze Root Causes Use basic quality tools Cause and Effect Diagram Affinity diagram Ask “WHY” 5 times

MBF Form Root Cause Analysis/Diagnosis Remedies PROBLEM STATEMENT/OBJECTIVES & PERFORMANCE TRENDS: (indicate a brief statement of fact here) Insert graph of performance over time here (e.g., Pareto Chart / Current State vs. Goal) Insert graph of more detailed information here (Supportive Info. as necessary - e.g., Costs associated with the problem statement)

Troubleshooting Diagnosis & remedial action applied to sporadic problems Generally receives immediate attention Result of adverse change Similar to chronic problem journey, but often simpler Journey is discovering what the change was and removing or adjusting for it Journey: Symptom to Cause to Remedy If troubleshooting is ineffective, the result will not be maintained

Troubleshooting Ask: What, Where, When, How Much, Who Compare Good vs. Bad, time to time, ETC. Ask what “IS & IS NOT” Reconstruct time sequence Take corrective action to remedy the problem Use a structured approach: 1. State deviation. 2. Specify the deviation (is & is not, what, where, etc.). 3. Identify unique characteristics of deviation. 4. Search for changes. 5. Develop possible causes. 6. Test the possible causes against the requirement. 7. Verify the cause (duplicate or eliminate by removing cause).

Summary Need to distinguish between sporadic & chronic problems Need to be persistent - get to root cause! For chronic problems: Use BREAKTHROUGH Use Root Cause Analysis (Diagnosis) For sporadic problems - use Troubleshooting Improvement requires use of multiple quality tools! To find true (root) causes you need to be a: Doctor Detective Driver Gardner Firefighter

References Books (+ many others): “ Quality Planning and Analysis”, J. M. Juran & Frank M. Gryna, Third Edition, McGraw Hill. “ Juran’s Quality Control Handbook” 4th Edition, McGraw Hill. The Memory Jogger II , Michael Brassard & Diane Ritter, GOAL/QPC, 1994. Analytic Trouble Shooting, Kepner Tregoe, 3rd Edition, 1978, Princeton Research Press

Rootcauseanalysispresentation 090728111941 Phpapp01

More Related Content

What's hot (20)

Similar to Rootcauseanalysispresentation 090728111941 Phpapp01 (20)

Recently uploaded (20)

Rootcauseanalysispresentation 090728111941 Phpapp01

Editor's Notes