culture clash
Jan 15, 2016
culture clash
2
how we think
▲ Fast:▲ Enumerativ
e▲ Heuristics▲ Surface
▲ Slow:▲ Analytic▲ Understanding▲ Under the
surface
Hard wired to operate here.
Science /C.I. operates here
3
thinking about things Political / Enumerative
Scientific / Analytic
describe population focus / inference
analyze process
static frame of reference dynamic
fast mode of thinking slow
theory over data truthdata over
theory
command and control
management model leadership
5
cultural dimension
CEO
Vice-President Vice-president Vice-President Vice-President
Manager Manager Manager ManagerManagerManagerManagerManagerManager Manager
Staff StaffStaffStaffStaffStaffStaffStaffStaffStaff
ConsumerResearch
CONSUMERS
Design &Redesign
Receipt ofInputsA
B
C
Systems of Production & Service
Tests of Processes
6
Organizational ExpressionCharacteristic Professional
ManagementPracticing Management
Management Model/Thinking
Hierarchical Command and Control
Systems Thinking and Leadership
Ethos Dogmatic, ideological, rules, control
Pragmatic, scientific, experiments, learning
Location Executive Suite, 42th floor of head office
Gemba, the heart and soul of the operation
Information processing / understanding
Enumerative, descriptive/inspection
Analytic, explanatory/predictive
Primary activity Reading reports and writing directions
Running the business and providing leadership
hierarchy and positions organizational model systems and processes
7
Systems Thinking: Complexity & Variation
T HREE E LEMENT S YSTEM S EVEN E LEMENT S YSTEM
S YSTEM S TATES = 64 S YSTEM S TATES = 4,398,046,511,104
8
the challenge: improvement & innovation
▲ The boss and the business are hardwired against change (despite endless talk to the contrary).
▲ Further, they are hardwired against analysis and evidence-based arguments (despite endless claims to the contrary).
▲ Make improvements. Just don’t change anything.
PRACTICAL TOOLS TO CHALLENGE THE STATUS QUO THINKING
1. Kahneman’s Human Psychology2. Systems Map3. Capacity/Little’s Law4. Control Chart/Performance
Measurement5. Taguchi Nominal Targets
10
1. Kahneman Heuristics
Heuristics▲ Representativeness
▲ Anchoring & Adjustment
▲ Availability
Strategies▲ Metaphors,
Comparative Similes, Stories
▲ Accurate Baselining, Detailing the Variation, Models.
▲ Broaden the Search Field, Provide Context, Use ModelsEnough doubt must be raised to convince people
to give up their evolutionary tendencies and expend energy in thinking slow about the problem and solution.
11
2. Systems Diagram
Source: revised from Vanguard Consulting Group, with permission
Process StepProcess Step
Customer Touch/Interaction Point
Process StepProcess StepProcess Step
Process StepCustomer Touch/Interaction Point
Process StepCustomer Touch/Interaction Point
Process StepCustomer Touch/Interaction Point
CUSTOMER
system conditions
management thinking
process efficiency & flexibility
what matters to customers
demand/response capability
purpose of the system1
2
3
4
5
6
12
3. Understand Capacity/Little’s Law
▲ Capacity is dynamic concept
▲ Metaphor spirit, let’s talk about pipes
▲ Little’s Law: TH = WIP / CT
13
3. Understand Capacity:Process Dynamics
ProcessBacklogQueue
Arriving Work/Inputs
Work (Uin)
CompletedWork
Finished Product/Services
TH =WIP
CT
Rework or Failure Demand (Urwrk)
Scrap (Uscrp)
Uout =
First Time Yield = (Uin – Uscrp – Urwrk) / Uin
Failure Demand may represent the largest source of demand in the
system
Yield = (Uin – Uscrp) / Uin
CTq CTe
CT = (CTq + CT e)
WIP = (WIPq + WIPe)
WIPq WIPe
WIP: Total number of people or units of work in the system
WIPe: how many people/units are in-process.
WIPq: how many units/people are waiting for
service.
CT: Total time it takes one unit to traverse the system, time from order to delivery CTe: processing time per
person or unit of work.
CTq: waiting time per person of unit of work.
Little's Law; The fundamental law governing the performance behaviour of all processes
TH will equal for a system that is in balance.
14
3. Understanding Capacity: Variation & Congestion
Cyc
le T
ime
CapacityUtilization
100%
Non
Fea
sibl
e R
egio
n
Cyc
le T
ime
CapacityUtilization
100%
15
3. Understanding Capacity: Variation & Congestion
U U
Term
Processing time in hours
(te)
Wait Time in hours
(CTq)
Total Cycle Time in hours
70% 2.33 2.4 5.6 8.0 80% 4.0 2.4 9.6 12.090% 9.0 2.4 21.6 24.095% 19.0 2.4 45.6 48.0100% infinite 2.4 infinite infinite
16
4. Performance Measurement
What gets measured gets done.--Current Management Insight
If you cant measure it, you can’t manage it. --Current Management Wisdom
Measure the meaningless and misinterpret what is measured.--Current Management Practice
17
4. AHS Performance Dashboard
YTD Target compared to Actual
This Qtr. compared to previous Qtr.
YTD compared to YTD last year
18
4. Executive Trends . . .
any two data points
(-)
per
form
ance
(+
)
time(-
) p
erfo
rman
ce
(+)
time
(-)
per
form
ance
(+
)
time
performance is increasing(positive trend)
performance is flat(neutral trend)
performance is decreasing(negative trend)
19
4. Hospital ScorecardPerformanc
e RankLOS
Comparison to
System Average
Target
Comparison to Target
Hospital 1 1 3.03 2.5 Hospital 2 3 3.07 2.5 Hospital 3 4 3.12 2.5 Hospital 4 2 3.03 2.5 Hospital 5 5 3.15 2.5
Overall System Average
3.08 2.5
Legend: Better than
expected Within +/- 5% Worse than
expected
20
4. PM Hospital ANOVA
Minitab Output:No significant differences among the 5 hospitals
21
4. PM: No Difference?
Highlights and Conclusions
LOS1: rising trend
LOS2: Likely two shifts at periods 16 and 24.
LOS3: reliable performance, high random variation
LOS4: reliable performance
LOS5: multiple special causes
28252219161310741
8
7
6
5
4
3
2
1
0
Period
Indiv
idual V
alu
e
_X=3.027UCL=3.595
LCL=2.458
111
1111
111
1
2222
2222
13
1111
11
1
I Chart of LOS1
Converge Consulting Group Inc.
28252219161310741
8
7
6
5
4
3
2
1
0
Period
Indiv
idual V
alu
e
_X=3.073
UCL=4.119
LCL=2.028
11
111
I Chart of LOS2
Converge Consulting Group Inc.
28252219161310741
8
7
6
5
4
3
2
1
0
Period
Indiv
idual V
alu
e_X=3.127
UCL=6.016
LCL=0.238
I Chart of LOS3
Converge Consulting Group Inc.
28252219161310741
8
7
6
5
4
3
2
1
0
Period
Indiv
idual V
alu
e
_X=3.033
UCL=4.794
LCL=1.273
I Chart of LOS 4
Converge Consulting Group Inc.
28252219161310741
8
7
6
5
4
3
2
1
0
Period
Indiv
idual V
alu
e
_X=3.15
UCL=6.433
LCL=-0.133
11
2
1
I Chart of LOS 5
Converge Consulting Group Inc.
Run and Control charts are the only ways of communicating any performance metric.
23
5. Performance TargetsWhere did this come from?
Nominal is the place to be.
24
5. Taguchi Nominal
▲ Provides the cost at any target distance from nominal.
▲ EL(x) = k(s2 + barx2) smaller is better
▲ k is a constant = cost(scrap) /(xscrap-T)2
25
5. Emergency Service
▲ k = cost(scrap) /(xscrap-T)2 = .007▲ cost(scrap) = 1 life
▲ Xscrap = 20 minutes▲ T = 8 minutes
▲ EL(x) = .007 (s2 + barx2) = 0.28▲ s = 2▲ Barx = 6
26
5. Taguchi ImplicationsThis target builds in a 28% death rate arising from slow ambulance response rates.
CONCLUSIONS
Understand that you are coming from a whole different perspective .
People are hardwired to reject your approach and your reasoning
To shift people from Fast to Slow, use facts to address their operating heuristics
Many Lean, Lean Six Sigma, Continuous Improvement tools are designed to do just that.