Step 1 Problem Identification 1 Give a detailed description of the problem and its impact
Step 2 Suppliers Initial Response & Containment 2 Actions required to ensure that the customer is protected from receiving further nonconformances Response required within 1 Business Day
Step 3 Delphi Accepts or Rejects Initial Response 3 Reply required within 1 Business Day
Step 4- Supplier Determines the Root Cause 4 Detailed investigation of what caused the problem. Delphi prescribes 5 why as the tool to determine root cause
Step 5 Supplier Develops Corrective Action/Solution 5 Actions taken to eliminate the root cause
1 November 6, 2001
PRR Training
Problem Reporting & Resolution What are the Key Steps of the Problem Resolution Process?
Step 10 Delphi Closes the PRR 10Full PRR cycle within 30 calendar days!!!
2 November 6, 2001
PRR Training
Supplier Problem Identified Owner(Delphi) Issues PR/R Major Disruption, Repeats, Customer Impacted no Supplier Initiates Containment & Provide details
1 business day Owner Provides feedback to Supplier reject 1 business days 15 calendar days
Owner Reviews Suppliers Response accept Suppliers Final Answer including 5 Why
Extension Request
Owner or SQ Provides feedback to Supplier Owner (with SQ for MD, Repeats, CI) reviews RC, CA, and Implementation Plan All accepted After Implementation, Supplier will provide at Delphis request 15 days of Verification evidence Any item rejected
2 business days
reject
PRR Training
PRR Responsiveness
The new PRR process was designed to reduce the cycle time from PRR issuance to closure The new PRR system (Covisint Problem Solver) will monitor PRR responsiveness of the Supplier and Delphi Supplier response requirements: Initial Response - 1 Business Day Final Response - 15 Calendar Days Extended Request - As approved by Delphi Delphi response requirements: Suppliers initial response reply- 1 Business Day Suppliers final response reply - 2 Business Days Final Closure - 30 Calendar days if no extension
4 PRR Training November 6, 2001
5 Why Analysis
Understanding Quick
Critique 5-
Wrap
PRR Training
5-WHY
a supplier has submitted an initial response and containment plan (Step # 2 in the PRR process), a detailed Step investigation is necessary to determine what caused the problem. Step # 4 (Supplier determines the root cause) requires a 5-Why analysis to help in identifying the root cause of the problem. back to one of the elements within the Purpose of a PRR to facilitate problem resolution, 5-Why is the prescribed tool for determining the root cause of the problem to facilitate problem resolution.
8
Going
PRR Training
November 6, 2001
5-Why:
Specific problem:
Why did we have the problem? Why did the problem reach the Customer? Why did our system allow it to occur?
System failure:
Basic Cause/Effect Investigation Where in the process is the problem occurring? Go See the problem.
q
Area of Cause Located Point of Cause (PoC) Why? 1 Cause Why? 2 Cause Why? 3 Cause Why? 4 Cause Why? 5
C A U S E
I N V E S T I G A T I O N
q
Root Cause
q
Corrective Action
q
Lessons Learned
10 PRR Training November 6, 2001
Root Causes
WHY?
Use this path to investigate why the problem was not detected.
WHY?
WHY?
WHY?
WHY?
WHY?
C
Ref. No. (Spill, PR/R) WHY? WHY?
B
Date of Spill WHY?
Customer
WHY?
Product / Process
Delphi Location
WHY?
C
Problem Resolution Complete Communicate to Customer Date: Process Change Break Point Date: Implement System Change Date:
Lessons Learned:
11 November 6, 2001
PRR Training
Generic Information for 5-Why Example: Regular 5Cola Soft Drink vs. Diet Cola
Quick 5-Why Exercise as a group 5The plant received a PR/R from a customer. (We use 5-Why Analysis to answer every PR/R.) The PR/R states that the customer received Regular Cola in the right container (same for both products) with the Diet Cola label. The order called for Regular Cola. The plant has two identical lines that are capable of running either of our two products. The lines are located immediately beside each other. The only differences in the products are the syrup and the labels. The plant runs both lines 24 hours per day. There are three shifts that run 7:00 a.m. to 3:00 p.m., 3:00 p.m. to 11:00 p.m., and 11:00 p.m. to 7:00 a.m. The date code indicates that the defective product was manufactured at 3:03 p.m. Defective product has been contained and sorted.
12 PRR Training November 6, 2001
BOTTLES
WATER
SYRUP
B O T T L I N G
13
LIDS
INSPECT
LABELS
PRR Training
November 6, 2001
Defi e
r lem 2
WHY
Root C
Changeover " Use t is at to yt e investigate problem as not detected. Operator not certified. (
WHY
WHY
No for'alized procedure. ) (
WHY
No for'al certification procedure. Use t is pat to investigate t e systemic root cause Operator dependent process; no supervision. Relying of operator to follow up procedure. 2
Ref. No. ( p ill, PR/R) WHY
PRR Training
@8
998
o ns L
r ned:
14 November 6, 2001
Proble
Resolution Complete
Product / Process
Delphi Location
WHY
Date of
p ill
WHY
Budgetary constrains.
WHY
WHY
C B
C
Implement y stem Change Date:
Date
0 0
% #% # $ #
General Guidelines:
A.) Dont jump to conclusions!; B.) Be absolutely objective. C.) Dont assume the answer is obvious. D.) If you are not thoroughly familiar with the process yourself, assemble a cross-functional team to complete the analysis.
Step 1: Problem Statement Is the analysis being reported on the problem as the Customer sees it? Step 2: Three Paths (Dimensional, Detection, Systemic) -Are there any leaps in logic? -Is this as far as the Whys lead? Can you still ask one, two, three more whys)? -Is there a true cause-and-effect path from beginning to end of each path? Is there statistical data/evidence to prove it? ---Can the problem be turned off and on? -Does the path make sense when read in reverse from cause to cause? (e.g.We did this, so this happened, so this happened, and so on, which resulted in the original problem.) -Do the whys go back to the actual error? -Does the systemic path tie back to management systems/issues? -Does the dimensional path ties back to issues such as design, operational, tier-n management, etc? -Does the detection path ties back to issues such as protect the customer, control plans, etc? Step 3: Corrective Actions -Does each corrective action address the root cause from a path? -Is there a separate corrective action for each root cause? If not, does it make sense that the corrective action applies to more than one root cause? -Is each corrective action possible to implement? -Are there corrective actions that affect the Customer or require customer approval? How will they be communicated to the Customer? -Is there evidence and documentation to support the validity of the corrective actions? -Are the corrective actions irreversible? If not, are there corrective actions in place that address containment?
15 November 6, 2001
PRR Training
Step 4: Lessons Learned -How could this problem have been foreseen? -How will this information be implemented: a.) on the line or in the plant? b.) at the point of detection? c.) cross-functionally at the Supplier? d.) other product/plants? -Are there lessons learned for the Customer?
Step 5: Overall -Do there seem to be big holes where ideas, causes, corrective actions, or lessons learned are being avoided? -Where things are missed or not documented? -Do the corrective actions address what are the actions the supplier owns? -How many iterations has the supplier gone through so far in preparing this 5-why (It doesnt happen on the first try!) -Who prepared the 5-why?
Matl Flow
17 PRR Training November 6, 2001
Defi e rD lem End-cuts mixed with good parts. F F F A 1. 100 % visual containment before packing. 2. Dock audit each lot. 3. Posted quality alert at cutting station. End-cut not properly removed from cut station. G
WHY
Root C
ses
Use t is pat to investigate yt e problem as not detected. Operators not able to detect. Q
WHY
Operator mistake. Q
WHY
B 1. Posted quality alerts at cutting and packing operations. 2. Modified operator instructions to include the quality alerts. A
12340000 V W
Date of pill WHY
11/30/00 V
Product / Process Delphi Location Content Latest Rev Date WHY
Flint, MI.
Communicate to Delphi Date:
18 November 6, 2001
PRR Training
11/28/00
WHY
C
Implement ystem Change Date:
Date
R R
I HI H H
C B W
12/01/00
12/01/00 12/01/00
12/01/00
12/01/00
12/01/00
Corrective Actions
A
1. 100 % visual containment before packing. 2. Dock audit each lot. 3. Posted quality alert at cutting station.
End-cuts misloaded on good exit conveyor. End-cuts not properly removed from cut station.
Root Cause
Operator failed to remove. Operator mistake.
Problem definition, is it as the customer sees it? - customer stated mixed/foreign material in shipment. Is the 5 Why response acceptable? - Process of removing end-cut is operator dependent, yet no actions taken for error-proofing this area. Is this as far as the Whys lead? Is the last statement within each path the root-cause? Does the analysis tell the entire story? - Low frequency/intermittent problem, though no actions identified to address issue from ever re-occurring. Actions still dependent upon the hope that an operator finds the problem.
19 PRR Training November 6, 2001
Corrective Actions
B
1. Posted quality alert at cutting and packing stations 2. Modified operator instructions to include the quality alerts.
Root Cause
Problem definition, is it as the customer sees it? - customer stated mixed/foreign material in shipment. Is the 5 Why response acceptable? - Process of removing end-cut is operator dependent, yet no actions taken for error-proofing this area. Is this as far as the Whys lead? Is the last statement within each path the root-cause? Does the analysis tell the entire story? - Low frequency/intermittent problem, though no actions identified to address issue from ever re-occurring. Actions still dependent upon the hope that an operator finds the problem.
20 PRR Training November 6, 2001
Corrective Actions
C
1. Added improper end-cut removal to PFMEA.
Root Cause
Problem definition, is it as the customer sees it? - customer stated mixed/foreign material in shipment. Is the 5 Why response acceptable? - Process of removing end-cut is operator dependent, yet no actions taken for error-proofing this area. Is this as far as the Whys lead? Is the last statement within each path the root-cause? Does the analysis tell the entire story? - Low frequency/intermittent problem, though no actions identified to address issue from ever re-occurring. Actions still dependent upon the hope that an operator finds the problem.
Date
Mixed/foreign material.
WHY
WHY
WHY
Assumed 2 PPM is good enough. Use t is pat to investigate t e systemic root cause Operator dependent process (end-cut removal) not addressed with error prevention.
WHY WHY
WHY
Not Considered.
C
WHY
1. Process design standard now includes end-cut handling best practice. B 2. APQP process to identify and address all customer quality issues and process causes.
12340000
Date of pill
11/30/00
Product / Process Delphi Location Content Latest Rev Date WHY
Flint, MI.
Communicate to Delphi Date:
C 11/28/00
Process Change Break Point Date: Implement ystem Change Date:
Lessons Learned: 1.) Operator dependent process must be supported by error prevention methods. 2.) PFMEA must include all process failure modes and associated causes.
22
PRR Training
g g f f f f
q q
cb
Defi e r
a ` u
Root Causes
A 1. 100 % visual containment before packing. 2. Dock audit each lot. 3. Posted quality alert at cutting station.
12/01/00
4. Add closed end-cut chute with parts pass sensors to single proper end-cut removal, cutting station stops if not. 5. Added lock box to hold end-cuts. Operator failed to remove. 6. End-cuts made larger than "good" Orings; no pass bar installed on exit conveyor to stop any end-cuts.
12/08/00 12/08/00
B
Operator mistake. 1. Added end-cut mishandling to PFMEA. No error prevention/detection inplace. A 2. Added verification of error detection methods to Process Control Plan (PCP), et-up, and Preventive Maintenance (PM) instructions.
12/08/00
12/08/00
12/08/00
November 6, 2001