Aerojet Rocketdyne’s
InThinking Network
Better Thinking About...
Webinar Series
Better Thinking About Tools & Processes
Six Hats, SPC, and Problem Analysis
Presented by Tim Higgins
Timothy.higgins@rocket.com
November 14, 2013
1
Transformation is a shift in perception
Getting to China in 1200 AD
Clever tools elegantly used yield
- lots of “go east” alternatives
- never a “go west” idea
“Go west” alternatives emerge
only
within a round earth perception
Prevailing style
of management
Appreciation
for a system
Theory of
Knowledge
Understanding
variation
Psychology
2
Tools without Profound Knowledge
Effects of tool application in the absence of knowledge:
•
•
•
•
•
Discovering faster ways to dig deeper the pit we are in
More efficient execution of practices that ought to be abandoned
Creative ways to rate and rank people
Improving one part better and making the whole worse
Continuous tampering
3
Competitive Advantage
Each tool has a purpose
Using it on for a different purpose will decrease effectiveness of results
Capability to apply multiple tools can increase effectiveness
Can any organization learn and apply Six Thinking Hats?
Can any organization learn and apply Kepner-Tregoe?
Can any organization learn and apply Statistical Process Control?
Wherein does competitive advantage lie?
4
Competitive Advantage
What if I could join Six Thinking Hats and Kepner Tregoe?
What if I could join Hats, K-T, and SPC?
Could I improve the effectiveness of results over using them separately?
Wherein does competitive advantage lie?
Transformation in perception
Integrated tool application
Both
5
Aim: demonstrate an integration of three tools:
̶
̶
̶
The statistical control of quality
Kepner-Tregoe Problem Analysis
Six Thinking Hats
About introducing the possibility
Introducing a possibility merely facilitates the start of turning that possibility into an
actual accomplishment. Developing a possibility into an actual accomplishment will
require learning the tools, practice, and PDSA – elements this webinar is not designed to
address.
6
SPC as a Mental Model
The statistical
control of quality
=
Knowledge about
common causes
and special causes
SPC is ultimately a way of thinking a catalyst for seeking understanding.
7
SPC as a Tool
Without an understanding of “Production Viewed as a System” and
without an understanding of how to use the Deming Learning and
Improvement Cycle [PDSA] – the techniques are of little use.
Wheeler and Chambers, Understanding Statistical Process Control
May-09
Feb-09
Nov-08
Aug-08
May-08
Feb-08
Nov-07
Aug-07
May-07
Feb-07
Nov-06
Aug-06
May-06
Feb-06
Nov-05
Aug-05
May-05
Feb-05
Nov-04
Aug-04
May-04
Feb-04
Nov-03
Aug-03
4.0
2.0
0.0
Rate Delta
Rate
Cal Ops OSHA Recordable Rate
last 70 months
mR (moving range)
4.0
2.0
0.0
1
3
5
7
9
11
13
15
17
19
21
23
25
27
29
31
33
35
37
39
41
43
45
47
49
51
53
55
57
59
61
63
8
Six Thinking Hats
White Hat
Yellow Hat
Black Hat
Green Hat
Blue Hat
Caution, Difficulties, Weaknesses
Logical Negative
Alternatives and Options
Managing the Thinking Process
Red Hat
Hunches, Intuition and Gut Feelings
Information
Available and Needed
Benefits
Logical Positive
Exploration in parallel
9
Hats Sequences
A sequence for IMPROVEMENT
What are weaknesses?
What are alternatives that would overcome the weaknesses?
Another sequence for IMPROVEMENT (no better than the first)
Review current process
What are other people’s view on current process
What’s working well
What are the weaknesses
Generate ideas to overcome weaknesses
Choose most appropriate ideas
Decide on next steps.
Knowing the Hats and using them in sequences them
is a relatively small competitive advantage
10
Hats Sequences
A sequence for PROBLEM SOLVING (flopping about)
What could be causing the problem?
Generate ideas to solve
Chose best ideas
Benefits of each
Weaknesses of each
Overcome weaknesses
What needs doing by when
How well would this sequence aid us in the following cases?
11
Red Sweat Case
http://www.slideshare.net/GCU-Tech-Training/the-case-of-the-red-sweat-mystery
On Eastern Airlines flights over the Atlantic between New York City and Florida, flight attendants begin to
contract a strange, malady. The mystery malady: red spots, exuding a blood-like fluid, appear randomly on
faces, hands, and necks of flight attendants. Physicians and scientists have been unable to explain the
condition, which so far has manifested itself almost exclusively on A300 flights between New York and
Florida. Flight attendants are in a panic. Some believe they are sweating blood, though an Eastern
spokesperson said there is no evidence the fluid is blood. Still, Eastern has launched a major effort at finding
the cause of the malady. The fluid escaping from their pores looks like blood, though it is not, and so the
condition is called “red sweat.” Others are stricken by reddish blotches of pinprick-size dots. But either way,
before a doctor can diagnose it, the mysterious condition disappears – until, perhaps, the next flight over the
Atlantic to Florida. So far at least 60 flight attendants have reported cases of the “red sweat” and many have
had it more than once. Records show 120 incidences in the last two months, though a few cases can be
traced back as far as two years. No passengers or other crew have shown symptoms, and no such condition
has been reported on inland flights. No other airline has reported a similar problem. Airbus, makers of the
A300, says foreign carriers have not encountered anything similar.
How well would the
sequence aid us in this
case?
12
Dead Batteries
Some dealers reported that for the new model car, not all the cars could be started after
being rolled off the delivery trailers. They all suffered from dead batteries. Cars were
being transported from the assembly plant to dealers on car carrier trailers. Each trailer
holds eight cars. Reports of dead batteries came from dealers as close as 450 miles
from the assembly plant and as far away as 1600 miles. No reports came from dealers
closer to the assembly plant. A compilation of the data associated with the car deliveries
indicated that for dealers reporting dead batteries, 400 cars had been delivered and 50
had dead batteries. Several of the dealers who had received multiple deliveries reported
that the only one of the cars in each delivery suffered from the problem, and two of them
reported that only the last car to be off loaded from each trailer had a dead battery.
How well would the sequence aid us in this case?
13
Failed Relay
An assembly designed to control electrical operations failed to function properly during its
operation on a test stand. The failure happened after the unit had been submitted to vibration that
is a standard part of the test operations activity. The controller assembly was removed from the
stand, and an operational check was performed on it.
For each of four relay sequence boards (K1, K2, K3, K4, and K5) and four ordinance boards (L1,
L2, L3, L4, L5) current is supposed to flow between certain contacts when specific circuits are
energized on a relay. When Y1-Y2 is energized on a sequence board current is supposed to flow
from contacts A2 to A3, B2 to B3, and C2 to C3. During the checkout, when Y1-Y2 was energized
on relay sequence board K4, there was no current flow from B2 to B3. The checkout revealed no
other electrical anomalies on K4 and no anomalies on K1, K2, K3, or K5. The controller assembly
was taken apart, as well as each of the relay subassemblies, and contamination was found in
relay K4. No other anomalies were found with the visual examinations.
How well would the sequence aid us in the following cases?
14
Problem Analysis
Kepner Tregoe
Structure of a problem
SHOULD performance, condition
DEVIATION
ACTUAL from SHOULD
ACTUAL condition, performance
What type of
variation does this
structure fit?
15
Problem Analysis
Kepner Tregoe
Signpost for utilizing PROBLEM ANALYSIS
Is there a DEVIATION from the SHOULD?
Is the cause UNKNOWN?
Do I need to know cause in order to take effective actions?
“YES” to all three equals candidate for Problem Analysis
16
Problem Analysis
Kepner Tregoe
State the problem
Object - deviation
Specify the problem
Is
and Is not
WHAT WHERE
WHEN EXTENT
Generate possible causes
from knowledge and experience
from distinctions and changes
Evaluate possible causes
Select most probable cause
Confirm true cause
Aim - finding immediate or proximate causes.
“Root” causeS lie in institutional practices. The K-T Poblem Analysis
result is the starting point for exploring the institutional causes.
17
PROBLEM: Contact failed to pass current
IS
Y1-Y2 energized
Contact B2 to B3
IS NOT
Contacts A2 to A3, C2 to C3
No output
Incorrect output, intermittent output
Relay K4 Sequence Board
Relays K1,2,3,5 Sequence board & L1-5
Ordnance board
At test facility
At sub-assy supplier, at end item
manufacturer
When?
After vibration / shock test
Sporadic
Before test
Continuous, cyclic
Extent?
One relay, one contact
More relays, more contacts
What?
Where?
18
POSSIBLE CAUSES
1.0 Contact load too low
2.0 Magnet too weak
3.0 Bent or malformed contact point
4.0 Transient contamination between contact points
5.0 Coil fails to transfer
6.0 Pulse width too short
7.0 Contact resistance too high
8.0 No command
9.0 Sticky relay
10.0 Vibration/shock/sine pulse test damage to relay
11.0 Indicator (hand held meter) malfunction
12.0 Circuit design flaw
13.0 Open circuit inside relay
19
EVALUATE CAUSES - IF __________ is the cause of contact failed to pass current,
then how does it explain both the IS and the IS NOT conditions?
1.0 Contact load too low
DOES NOT EXPLAIN why two contacts work
DOES NOT EXPLAIN why failure was not seen at sub-assy supplier or end item supplier
2.0 Magnet too weak
DOES NOT EXPLAIN why two contacts work
DOES NOT EXPLAIN why failure was not seen at sub-assy supplier or end item supplier
3.0 Bent or malformed contact point
EXPLAINS ONLY IF failure was not seen at sub-assy supplier or end item supplier because the
vibe/shock on the test stand caused damaged to the contact.
DOES NOT EXPLAIN why failure is not continuous
4.0 Transient contamination between contact points
EXPLAINS THE DATA ONLY IF
Other relays do not have contamination OR contamination in other relays did not move to prevent
contact connection
5.0 Coil fails to transfer
DOES NOT EXPLAIN why two of the contacts pass current
20
6.0 Pulse width too short
DOES NOT EXPLAIN why two of the contacts pass current
7.0 Contact resistance too high
EXPLAINS ONLY IF the contact load is too low OR if contamination prevents contact
8.0 No command
DOES NOT EXPLAIN why two of the contacts pass current
9.0 Sticky relay
DOES NOT EXPLAIN why two of the contacts pass current
10.0 Vibration/shock/sine pulse test damage to relay
EXPLAINS ONLY IF the contact load is too low
DOES NOT EXPLAIN why two of the contacts pass current
11.0 Indicator (hand held meter) malfunction
DOES NOT EXPLAIN why meter functions for two contacts that pass current
LAB TEST CONFIRMED FUNCTIONING METER
12.0 Circuit design flaw
DOES NOT EXPLAIN why during checkouts subsequent to test current began to pass
DOES NOT EXPLAIN why two of the contacts pass current
13.0 Open circuit inside relay
EXPLAINS ONLY IF contamination held contact open
21
CAUSES NOT INVALIDATED DURING EVALUATION
Transient contamination between contact points
EXPLAINS THE DATA ONLY IF
Other relays do not have contamination OR contamination in other relays did not move to prevent contact
connection
Contact resistance too high
EXPLAINS ONLY IF the contact load is too low OR if contamination prevents contact
Open circuit inside relay
EXPLAINS ONLY IF contamination held contact open
MOST PROBABLE CAUSE
Transient contamination between contact points.
CONFIRMATION OF CAUSE
Remove contamination and functionally check relay K4
22
INTEGRATION
23
COMPARE
Hats sequence
Problem Analysis
What could be causing the problem?
Generate ideas to solve
Chose best ideas
Benefits of each
Weaknesses of each
Overcome weaknesses
What needs doing by when
Where does competitive advantage lie?
24
Both: Transformation in perception & integrated tool application
Without an understanding of “Production Viewed as a System” and without an understanding of how
to use the Deming Learning and Improvement Cycle no tool / technique is of much use.
Statistical control of quality
Transformation in perception
Tool for charting to stimulate thinking
Six Thinking Hats Tool for EXPLORING a subject
Tool for finding proximate cause of
Problem Analysis
Kepner-Tregoe Rational Process special cause variation
25
Problem Analysis
Describe problem
State the problem
Specify the problem
Identify Possible Causes
Use knowledge and experience, or use
distinctions and changes to develop
possible cause statements
Evaluate Possible Causes
Test the possible causes against the
aaIS and IS NOT specification
Determine the most probable cause
Confirm True Cause
Verify assumptions, observe,
aaexperiment, or try a fix and monitor
Decision Analysis
Situation Appraisal
Identify Concerns
List threats and opportunities
Separate and clarify concerns
Set Priority
Consider seriousness,
urgency, and growth
Plan Next Steps
Determine analysis needed
Plan Involvement
Determine help needed
Clarify Purpose
State the decision
Develop objectives
Classify objectives into MUSTs and WANTs
Weigh the WANTs
Assess Risks
Identify adverse consequences
Evaluate Alternatives
Generate alternatives
Screen alternatives through the MUSTs
Compare alternatives against the WANTs
Make Decision
Make the best balanced choice
Potential Problem (Opportunity) Analysis
Identify Potential
aaProblems (Opportunities)
State the action
List the potential problems (opportunities)
Identify Likely Causes
Consider causes for the
aapotential problem (opportunity)
Take Preventive (Promoting) Action
Take actions to address likely causes
Plan Contingent (Capitalizing)
aaAction and Set Triggers
Prepare action to reduce (enhance)
aalikely effects
Set triggers for contingent
26
aa(capitalizing) actions
Kepner Tregoe Rational Processes
27
Situation Appraisal
Identify Concerns
List threats and opportunities
Separate and clarify concerns
Set Priority
Consider seriousness,
urgency, and growth
Plan Next Steps
Determine analysis needed
Plan Involvement
Determine help needed
28
Decision Analysis
Clarify Purpose
State the decision
Develop objectives
Classify objectives into MUSTs and WANTs
Weigh the WANTs
Assess Risks
Identify adverse consequences
Evaluate Alternatives
Generate alternatives
Screen alternatives through the MUSTs
Compare alternatives against the WANTs
Make Decision
Make the best balanced choice
29
Potential Problem Analysis
Take Preventive Action
Identify Potential Problems
Take actions to address likely causes
State the action
List the potential problems (opportunities)
Identify Likely Causes
Consider causes for the
potential problem
Plan Contingent Action and Set Triggers
aa
Prepare action to reduce (enhance) likely effects
Set triggers for contingent actions
aa
aa
30
Potential Opportunity Analysis
Identify Potential Opportunities
State the action
List the potential opportunities
Take Promoting Action
Take actions to address likely causes
Identify Likely Causes
Consider causes for the
potential opportunity
Plan
aa Capitalizing Action and Set Triggers
Prepare action to enhance likely effects
Set
aa triggers for capitalizing actions
aa
aa
31