Statistical Process Control Concepts
Download
Report
Transcript Statistical Process Control Concepts
Statistical Process Control
Concepts
Hampton Roads Section, ASQ
June 15, 2011
Scott Rutherford
Master Black Belt, Norfolk Naval Shipyard
Learning Objectives
Understand the three general areas of
SPC
Understand the proper use of SPC in
your Quality program
Understand SPC’s pitfalls and success
factors
History
1920’s: “Created” by Walter Shewhart for
Bell Labs
WWII: W. Edwards Deming applies
concepts to munitions manufacturing
After WWII: Deming teaches SPC to
Japanese to help build back their
industrial base
1988: SPC applied to software industry
becomes CMMI (Causal Maturity Model
Integration)
Shewhart’s original concept
Physical process data is usually NOT
normally distributed
Processes either display:
◦ Controlled variation that is natural to the
process
◦ Uncontrolled variation that is generated from
outside the “process causal system” (assignable
cause)
Attacking process variation produces a
product that consistently conforms to
specifications
Three Elements of SPC
Understanding the Process
Understanding the Causes of Variation
Eliminating the sources of Assignable
(Special) Cause variation
Techniques for Understanding the
Process
Process Mapping
Determine Measure of Quality
◦ From customer’s perspective
◦ Usually an output measure
Determine Measure of Quality Predictor
◦ What In-process Measure is the most reliable
predictor of the Quality Measure?
Measure the Predictor
◦ Control Charts / Run Charts
Understanding Causes of Variation
Root Cause Analysis - Reactive
◦ Causal Mapping – 5 Whys
◦ Switch Theory
◦ Human Factors Analysis and Classification System
(HFACS) – Swiss Cheese Analysis
◦ Kepner series (Tregoe, Fourie)
◦ Fishbone
◦ Others (Ford Global 8D, DuPont, etc.)
Failure Modes & Effects Analysis (FMEA) Proactive
Eliminating Special Cause Variation
Identification
◦ Control Chart monitoring
◦ Control Chart Analysis (“rules”)
Response
◦ Immediate Action (Control or Reaction Plan)
– Getting back to the norm
◦ Short-term actions – Preventing reoccurrence
◦ Long-term actions – Eliminating cause
Preventing Reoccurrence
Immediate Knowledge capture
Start of Work / Shift briefings
Verbal / “Pen & Ink” Procedure changes
Training
Periodic self-assessments
Proper monitoring of the process
Eliminating Reoccurrence
Lean Tools
◦ 5S
◦ Standard Work codification
◦ Mistake-proofing
FMEA Analysis
Process Knowledge Use
◦ R&D
◦ Next generation design
Behavioral / Cultural Change
Why SPC Initiatives often Fail
“It’s All about the charts”
“More important to train”
“Don’t have time to figure out the source,
get back to work!”
“The control plan should have caught that!”
“Sort out the bad to get enough good
product to sell”
“We have the time to do it over, but not
enough time to do it right the first time”
SPC succeeds when:
Supervisors empowering front line
workers in monitoring control charts and
executing control plans
There is a dedicated workforce to spend
the time to analyze sources of variation
Eliminating variation is a strategic initiative
What other comments or questions
do you have?
[email protected]
[email protected]